| Server IP : 156.238.232.47 / Your IP : 216.73.216.150 Web Server : nginx/1.25.3 System : Linux C202504152095410 3.10.0-1160.119.1.el7.x86_64 #1 SMP Tue Jun 4 14:43:51 UTC 2024 x86_64 User : www ( 1000) PHP Version : 8.3.25 Disable Function : passthru,exec,system,putenv,chroot,chgrp,chown,shell_exec,popen,proc_open,pcntl_exec,ini_alter,ini_restore,dl,openlog,syslog,readlink,symlink,popepassthru,pcntl_alarm,pcntl_fork,pcntl_waitpid,pcntl_wait,pcntl_wifexited,pcntl_wifstopped,pcntl_wifsignaled,pcntl_wifcontinued,pcntl_wexitstatus,pcntl_wtermsig,pcntl_wstopsig,pcntl_signal,pcntl_signal_dispatch,pcntl_get_last_error,pcntl_strerror,pcntl_sigprocmask,pcntl_sigwaitinfo,pcntl_sigtimedwait,pcntl_exec,pcntl_getpriority,pcntl_setpriority,imap_open,apache_setenv MySQL : OFF | cURL : ON | WGET : ON | Perl : ON | Python : ON | Sudo : ON | Pkexec : ON Directory : /www/data/wwwroot/2025_07_28_02/data/ |
Upload File : |
SQLite format 3 @ � � � � -�) � p������ R \� 6� a J
?
� � E
33cn0中文1adminadmin2017-11-30 13:55:372018-04-13 11:40:49
j j � �R� 8 q&/static/upload/file/20221107/1667827385971592.docx8 q%/static/upload/file/20221107/1667827385971592.docx8 q$/static/upload/file/20221107/1667827385971592.docx8 q#/static/upload/file/20221107/1667827385971592.docx8 q"/static/upload/file/20221107/1667827385971592.docx�q �a 2<p class="ds-markdown-paragraph" style="font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255); margin-top: 0px !important;"><span style="font-weight: 600;">Brand Name</span><br/>TOREX</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Model Number</span><br/>XC6220B331MR-G</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Part Number</span><br/>C86534</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Package</span><br/>SOT-25</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Packaging Method</span><br/>Tape &amp; Reel</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Gross Weight</span><br/>0.04g (grams)</p><p><br/></p>
� � �n U[ #''1 q1
cn湖南某某网络科技有限公司湖南长� � @ k
cnShenzhen Pengcheng Wanli Semiconductor Co., Ltd
� 8$��W*����`*
�
�
�
�
o
C
� � � o E � �a � ]
�
�
^
3
� � �� x Y�����}f�C� �F$ 8 3
tpl_html_cache_time9002 �7 )
tpl_html_cache12 �3 )+%watermark_textwww.adminbuy.cn2 �水印文本 D ;K
con4
sn_user2 �!2 ;
content_keyword_replace2 �6 +
message_rqlogin02 �+ !
close_site02 �65 #O
licensecodeOEI4RkY5MDM1MCwzRDU1M0Y5OTIzLw==B2 �!3 7
sn8B8FF90350,3D553F99232 �1 #
main_domain2 �0 )
to_main_domain02 �/
to_https02 �.
spiderlog02 � D-
lgautosw12 �, +
close_site_note2 �&( %
-tpl_html_dir2 �模板HTML目录%& )%message_status12 �留言开关
� *
#ip_allow2 �IP白名单 )
#ip_deny2 �IP黑名单"' #%form_status12 �表单开关$% '%url_rule_type32 �路径类型($ 1lock_time9002 �登录锁定时间'# !1lock_count52 �登陆锁定阈值)" ++form_check_code02 �表单验证码%! )%message_verify12 �留言审核 1%watermark_) 1%watermark_position42 �水印位置U 'q%watermark_pic/static/upload/image/20221107/1667811208970920.png2 �水印图片A 5#=watermark_text_color100,100,1002 �水印文本字体颜色1 31watermark_text_size202 �水印文本字号/ 3
1watermark_text_font2 �水印文本字体% )%watermark_open02 �水印开关. )7baidu_xzh_type02 �熊掌号推送类型+ )1form_send_mail02 �表单发送邮件( 1
%smtp_username_test2 �测试邮箱= =
Ccontent_tags_replace_num2 �内容关键字替换次数 !gzip02 �GZIP压缩& !
1wap_domain2 �手机绑定域名' +
)baidu_xzh_token2 �熊掌号token' +
)baidu_xzh_appid2 �熊掌号appid* )
1baidu_zz_token2 �百度站长密钥# !
+api_secret2 �API认证密钥"
+api_appid2 �API认证用户" +api_auth12 �API强制认证
api_open02 �API开关. +
7message_send_to1 �留言发送到邮箱4 /=message_send_mail11 �留言发送邮件开关#
'
%weixin_secret2 �微信SECRET! %
#weixin_appid2 �微信APPID* -+admin_check_code11 �后台验证码) '
1smtp_password2 �邮件发送密码) '
1smtp_username2 �邮件发送账号+ =smtp_ssl11 �邮件是否安全连接& -smtp_port4652 �邮件SMTP端口3 ##3smtp_serversmtp.qq.com2 �邮件SMTP服务器, 1+message_check_code11 �留言验证码 open_wap01 �手机版� � s������������������������� �z!
A
3
�M
33cn9高级Linux运维工程师#333333admin本站2018-04-12 10:38:09<p style="line-height: 150%"><strong><span style="font-size:16px;line-height: 150%;font-family:宋体">岗位职责：</span></strong></p><p>1、负责公司服务器基础环境的部署、配置、日常巡检、维护、故障的应急响应和问题处理；</p><p>2、负责公司kvm虚拟化平台的管理工作，基础环境部署，性能容量管理，漏洞扫描、安全加固，保证其稳定、高效运行；</p><p>3、负责维护公司集中监控系统，根据业务需求调整监控策略、告警阀值，处理告警信息和问题跟踪；</p><p>4、编写系统维护文档，完善并更新运维流程文档；</p><p style="line-height:150%"><span style="font-size: 16px;line-height:150%"> </span></p><p style="line-height:150%"><strong><span style="font-size:16px;line-height:150%;font-family:宋体">任职要求：</span></strong></p><p>1、计算机等相关专业，本科以上学历，2年以上linux系统管理工作经验，经验丰富可适�V$
'
3a
A
33
cn3荣誉证书5#333333admin本站2019-08-26 02:29:58/uploads/allimg/190826/1-1ZR61030060-L.jpg荣誉证书,荣誉证书, �1000 �adminadmin2019-08-26 02:30:072019-08-26 02:29:584�V$
'
3a
A
33
cn3荣誉证书4#333333admin本站2019-08-26 02:29:51/uploads/allimg/190826/1-1ZR61029550-L.jpg荣誉证书,荣誉证书, �1000 �adminadmin2019-08-26 02:29:572019-08-26 02:29:514�V$
'
3a
A
33
cn3荣誉证书3#333333admin本站2019-08-26 02:29:44/uploads/allimg/190826/1-1ZR61029480-L.jpg荣誉证书,荣誉证书, �1000 �adminadmin2019-08-26 02:29:502019-08-26 02:29:444�V$
'
3a
A
33
cn3荣誉证书2#333333admin本站2019-08-26 02:29:37/uploads/allimg/190826/1-1ZR61029410-L.jpg荣誉证书,荣誉证书, �1000 �adminadmin2019-08-26 02:29:432019-08-26 02:29:374�U$
'
3a
A
33
cn3荣誉证书1#333333admin本站2019-08-26 02:29:2 x s n �
� d _ Z U P � �L H< XN �P �O �S }K� 0 +
� � ����������������� ! '
150基础版红色,橙色" ' 2999专业版黄色,绿色"<;:9876543
2 "#$% &�
� W������������ G ! # 3 C x u �
� � � x �R�/���EEEEEE& �+ %/5
3
%
35
# � � � � x �2
!
3
q
35
cn101ABOUT USabout.html1THE PERMANENT WORKS/static/upload/image/20221107/1667822515313163.jpgaboutadminadmin2022-10-30 11:13:05 2025-07-29 17:49:1440
��L � k � %
/
%
35
cn1013联系我们index_lianxi.html1?lxwm/联系 � � � � � � � o 1-%
35
cn348条码软件系列productli � v +
!
35
cn112Company Profileabout.html0jianjieadminadmin2022-10-30 11:13:05 2025-07-30 18:37:2840�D -%
+
q
35
cn304PRODUCTSproductlist.htmlproduct.html1PRODUCT DISPLAY/static/upload/image/20221107/1667826021942384.jpgproductsadminadmin2022-10-30 11:13:05 2025-07-30 09:25:3340 x w )
%
35
cn11315Online Messagemessage.html1gbookadminadmin2022-10-30 11:13:05 2025-07-30 09:16:1640�'
!
%
q
35
cn1013CONTACT UScontact.html1/static/upload/image/20221107/1667827698556640.jpgcontactadminadmin2022-10-30 11:13:05 2025-07-29 17:55:1940�+ '
q
35
cn209NEWSnewslist.htmlnews.html1NEWS/static/upload/image/20221107/1667826964340852.jpgnewsadminadmin2022-10-30 11:13:05 2025-07-29 17:54:5240
� ��^ D _3ext_color4红色,橙色,黄色,绿色,蓝色,紫色颜色 � \ G3ext_type4�# #
%4ext_wenjian6文件上传 �
%3ext_cpms8产品描述 �
� �V V %-332搜集电话ay_diy_telephoneadminadmin2018-11-30 15:17:402018-11-30 15:17:40P %!331在线留言ay_messageadminadmin2018-04-11 17:31:292018-04-11 17:31:29
� ! �) 3 ( !KK�� �1=!!�-tableay_contentay_contentCREATE TABLE "ay_content" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"scode" TEXT(20) NOT NULL,
"subscode" TEXT(20) NOT NULL,
"title" TEXT(100) NOT NULL,
"titlecolor" TEXT(7) NOT NULL,
"subtitle" TEXT(100) NOT NULL,
"filename" TEXT(50) NOT NULL,
"author" TEXT(30) NOT NULL,
"source" TEXT(30) NOT NULL,
"outlink" TEXT(100) NOT NULL,
"date" TEXT NOT NULL,
"ico" TEXT(100) NOT NULL,
"pics" TEXT(1000) NOT NULL,
"content" TEXT NOT NULL,
"tags" TEXT(500) NOT NULL,
"enclosure" TEXT(100) NOT NULL,
"keywords" TEXT(200) NOT NULL,
"description" TEXT(500) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"status" TEXT(1) NOT NULL,
"istop" TEXT(1) NOT NULL,
"isrecommend" TEXT(1) NOT NULL,
"isheadline" TEXT(1) NOT NULL,
"visits" INTEGER(10) NOT NULL,
"likes" INTEGER(10) NOT NULL,
"oppose" INTEGER(10) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(20) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
, gtype TEXT(1) NOT NULL DEFAULT '4', gid TEXT(20) NOT NULL DEFAULT '', gnote TEXT(100) NOT NULL DEFAULT '', picstitle TEXT(1000) NOT NULL DEFAULT '')
8v##� tableay_extfielday_extfield CREATE TABLE "ay_extfield" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"mcode" TEXT(20) NOT NULL,
"name" TEXT(30) NOT NULL,
"type" TEXT(1) NOT NULL,
"value" TEXT(500) NOT NULL,
"description" TEXT(30) NOT NULL,
"sorting" INTEGER(11) NOT NULL
)�4@--�tableay_diy_telephoneay_diy_telephone
CREATE TABLE "ay_diy_tel�j?++� tableay_content_sortay_content_sort CREATE TABLE "ay_content_sort" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"mcode" TEXT(20) NOT NULL,
"pcode" TEXT(20) NOT NULL,
"scode" TEXT(20) NOT NULL,
"name" TEXT(100) NOT NULL,
"listtpl" TEXT(50) NOT NULL,
"contenttpl" TEXT(50) NOT NULL,
"status" TEXT(1) NOT NULL,
"outlink" TEXT(100) NOT NULL,
"subname" TEXT(200) NOT NULL,
"ico" TEXT(100) NOT NULL,
"pic" TEXT(100) NOT NULL,
"title" TEXT(100) NOT NULL,
"keywords" TEXT(200) NOT NULL,
"description" TEXT(500) NOT NULL,
"filename" TEXT(30) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
, gtype TEXT(1) NOT NULL DEFAULT '4', gid TEXT(20) NOT NULL DEFAULT '', gnote TEXT(100) NOT NULL DEFAULT '', def1 TEXT(1000) NOT NULL DEFAULT '', def2 TEXT(1000) NOT NULL DEFAULT '', def3 TEXT(1000) NOT NULL DEFAULT '') �))�tableay_content_extay_content_extCREATE TABLE "ay_content_ext" (
"extid" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"contentid" INTEGER(10) NOT NULL,
"ext_price" TEXT(100),
"ext_type" TEXT(100),
"ext_color" TEXT(100)
)�< �_tableay_configay_configCREATE TABLE "ay_config" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"name" TEXT(30) NOT NULL,
"value" TEXT(200) NOT NULL,
"type" TEXT(1) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"description" TEXT(30) NOT NULL
)�s;!!�1tableay_companyay_companyCREATE TABLE "ay_company" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"name" TEXT(100) NOT NULL,
"address" TEXT(200) NOT NULL,
"postcode" TEXT(6) NOT NULL,
"contact" TEXT(10) NOT NULL,
"mobile" TEXT(50) NOT NULL,
"phone" TEXT(50) NOT NULL,
"fax" TEXT(50) NOT NULL,
"email" TEXT(30) NOT NULL,
"qq" TEXT(50) NOT NULL,
"weixin" TEXT(100) NOT NULL,
"blicense" TEXT(20) NOT NULL,
"other" TEXT(200) NOT NULL
)�:�[tableay_areaay_areaCREATE TABLE "ay_area" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"pcode" TEXT(20) NOT NULL,
"name" TEXT(50) NOT NULL,
"domain" TEXT(100) NOT NULL,
"is_default" TEXT(1) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)P9++Ytablesqlite_sequencesqlite_sequence&CREATE TABLE sqlite_sequence(name,seq)
� � � � P� H!!�utableay_messageay_messageCREATE TABLE "ay_message" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"contacts" TEXT(10),
"mobile" TEXT(12),
"content" TEXT(500),
"user_ip" TEXT(11) NOT NULL,
"user_os" TEXT(30) NOT NULL,
"user_bs" TEXT(30) NOT NULL,
"recontent" TEXT(500) NOT NULL,
"status" TEXT(1) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
, uid INTEGER(10) NOT NULL DEFAULT 0) 0�[tableay_siteay_siteCREATE TABLE "ay_site" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"title" TEXT(100) NOT NULL,
"subtitle" TEXT(200) NOT NULL,
"domain" TEXT(50) NOT NULL,
"logo" TEXT(100) NOT NULL,
"keywords" TEXT(200) NOT NULL,
"description" TEXT(500) NOT NULL,
"icp�+A##� tableay_extfielday_extfield CREATE TABLE "ay_extfield" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"mcode" TEXT(20) NOT NULL,
"name" TEXT(30) NOT NULL,
"type" TEXT(1) NOT NULL,
"value" TEXT(500) NOT NULL,
"description" TEXT(30) NOT NULL,
"sorting" INTEGER(11) NOT NULL
)�4@--�tableay_diy_telephoneay_diy_telephone
CREATE TABLE "ay_diy_telephone" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"create_time" TEXT NOT NULL,
"tel" TEXT(20)
)�HJ�gtableay_roleay_roleCREATE TABLE "ay_role" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"rcode" TEXT(20) NOT NULL,
"name" TEXT(30) NOT NULL,
"description" TEXT(50) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)�8G))�+tableay_menu_actionay_menu_actionCREATE TABLE "ay_menu_action" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"mcode" TEXT(20) NOT NULL,
"action" TEXT(20) NOT NULL
)�WF�tableay_menuay_menuCREATE TABLE "ay_menu" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"mcode" TEXT(20) NOT NULL,
"pcode" TEXT(20) NOT NULL,
"name" TEXT(50) NOT NULL,
"url" TEXT(100) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"status" TEXT(1) NOT NULL,
"shortcut" TEXT(1) NOT NULL,
"ico" TEXT(30) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)� E�tableay_linkay_linkCREATE TABLE "ay_link" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"gid" INTEGER(10) NOT NULL,
"name" TEXT(50) NOT NULL,
"link" TEXT(100) NOT NULL,
"logo" TEXT(100) NOT NULL,
"sorting" INTEGER(11) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)�hD �#tableay_labelay_labelCREATE TABLE "ay_label" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"name" TEXT(100) NOT NULL,
"value" TEXT(500) NOT NULL,
"type" TEXT(1) NOT NULL,
"description" TEXT(30) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(20) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)�<C''�7tableay_form_fielday_form_fieldCREATE TABLE "ay_form_field" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"fcode" TEXT(20) NOT NULL,
"name" TEXT(30) NOT NULL,
"length" INTEGER(10) NOT NULL,
"required" TEXT(1) NOT NULL,
"description" TEXT(30) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)�LB�otableay_formay_form CREATE TABLE "ay_form" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"fcode" TEXT(20) NOT NULL,
"form_name" TEXT(30) NOT NULL,
"table_name" TEXT(30) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)
n �c�n! R %331cont�0留言内容 �adminadmin2022-10-30 08:36:492022-10-30 0S %351cont�1留言内容 �adminadmin2022-10-30 08:36:49 2022-10-30 08:39:41M 351email1邮箱 �adminadmin2022-10-30 08:36:43 2022-10-30 08:39:41O 351company1公司 �adminadmin2022-10-30 08:36:36 2022-10-30 08:39:40M 351phone1电话 �adminadmin2022-10-30 08:36:18 2022-10-30 08:39:39L 351name1姓名 �adminadmin2022-10-30 08:36:09 2022-10-30 08:39:38
� � z
c%33downli
5135zuobiao118.379591,35.0954411百度地图坐标adminadmin2022-10-30 08:35:38 2022-10-30 08:35:42
� �2�m
� e %?
33cn在线工具https://tool.adminbuy.cn/ �adminadmin2022-11-07 22:22:122022-11-07 22:22:12^ =
33cn仿站http://fang.adminbuy.cn/ �adminadmin2022-11-07 22:22:022022-11-07 22:22:02_ ;
33cnjs特效https://js.adminbuy.cn/ �adminadmin2022-11-07 22:21:522022-11-07 22:21:52b %9
33cn图标下载http://sc.adminbuy.cn/ �adminadmin2022-11-07 22:21:432022-11-07 22:21:43c #=
33cnAB模板网https://www.adminbuy.cn/ �adminadmin2022-11-07 22:21:352022-11-07 22:21:35g )=
35cnPbootCMS模板https://www.adminbuy.cn/ �adminadmin2018-04-12 10:53:06 2022-11-07 21:10:15
N �!�:
�
T � i � ~
�
� 8�K�U�p�+���>�G�N �B %A+33M1005M1001文章评论/admin/MemberComment/index\10fa-commenting-oadminadmin2019-10-04 08:25:412019-10-04 08:26:23sA %3 33M1004M1001会员管理/admin/Member/index[10fa-usersadminadmin2019-10-04 08:25:412019-10-04 08:26:23z@ %=!33M1003M1001会员字段/admin/MemberField/indexZ10fa-wpformsadminadmin2019-10-04 08:25:412019-10-04 08:26:23y? %= 33M1002M1001会员等级/admin/MemberGroup/indexY10fa-signaladminadmin2019-10-04 08:25:412019-10-04 08:26:23o> %1 33M10010会员中心/admin/M1001/indexX10fa-user-oadminadmin2019-10-04 08:25:412019-10-04 08:26:23z< +/-33M160M157自定义表单/admin/Form/index�11fa-plus-square-oadminadmin2018-05-30 18:25:412018-05-31 23:55:10q9 %/'33M1570扩展内容/admin/M157/index�10fa-arrows-altadminadmin2018-03-25 17:27:572018-04-30 14:50:34{: %7-33M158M156模型字段/admin/ExtField/index �11fa-external-linkadminadmin2018-03-25 21:24:432018-04-07 18:44:49l8 %/ 33M1560全局配置/admin/M156/index �10fa-globeadminadmin2018-03-25 17:20:432018-04-30 14:43:56r7 %1!33M155M156模型管理/admin/Model/index �11fa-codepenadminadmin2018-03-25 17:16:062018-04-07 18:44:40q= %/ 33M1000M157文章内链/admin/Tags/index�10fa-randomadminadmin2019-07-12 08:25:412019-07-12 08:26:23s5 %3!33M153M156配置参数/admin/Config/index �11fa-slidersadminadmin2018-03-21 14:52:052018-04-07 18:44:02n4 %/33M152M157友情链接/admin/Link/index�11fa-linkadminadmin2018-03-01 14:58:452018-04-07 18:47:16t3 %1%33M151M157轮播图片/admin/Slide/index�11fa-picture-oadminadmin2018-03-01 14:57:412018-04-07 18:47:07~2 %5533M150M157留言信息/admin/Message/index�11fa-question-circle-oadminadmin2018-02-01 13:20:172018-07-07 23:45:09q$ %1 33M136M156定制标签/admin/Label/index �11fa-wrenchadminadmin2018-01-03 11:52:402018-04-07 18:44:31x %5)33M132M130列表内容/admin/Content/index�00fa-file-text-oadminadmin2017-12-26 10:48:172018-04-07 21:52:15r %3 33M131M130单页内容/admin/Single/index�00fa-file-oadminadmin2017-12-26 10:46:352018-04-07 18:46:35r %/)33M1300文章内容/admin/M130/index�10fa-file-text-oadminadmin2017-12-26 10:45:362018-04-30 14:49:47u %=33M129M110内容栏目/admin/ContentSort/index/11fa-barsadminadmin2017-12-26 10:42:402018-04-07 18:46:25v %5%33M112M110公司信息/admin/Company/index.11fa-copyrightadminadmin0000-00-00 00:00:002018-04-07 18:46:09m %/33M111M110站点信息/admin/Site/index-11fa-cogadminadmin0000-00-00 00:00:002018-04-07 18:45:57n
%/!33M1100基础内容/admin/M110/index,10fa-slidersadminadmin2017-11-28 11:13:052018-04-30 14:48:29y +1)33M109M101服务器信息/admin/Site/server�11fa-info-circleadminadmin0000-00-00 00:00:002018-04-30 14:55:34y +7#33M108M101数据库管理/admin/Database/index�11fa-databaseadminadmin0000-00-00 00:00:002018-04-30 14:55:24n %/33M107M101类型管理/admin/Type/index�00fa-tagsadminadmin0000-00-00 00:00:002018-04-30 14:55:13s %3!33M106M101系统日志/admin/Syslog/index�11fa-historyadminadmin0000-00-00 00:00:002018-04-30 14:55:00o %/ 33M105M101系统用户/admin/User/index�11fa-usersadminadmin0000-00-00 00:00:002018-04-30 14:54:51u %/)33M104M101系统角色/admin/Role/index�11fa-hand-stop-oadminadmin0000-00-00 00:00:002018-04-30 14:54:43n %/33M103M101系统菜单/admin/Menu/index�00fa-barsadminadmin0000-00-00 00:00:002018-04-30 14:54:35q %/!33M102M101数据区域/admin/Area/index�11fa-sitemapadminadmin0000-00-00 00:00:002018-04-30 14:54:23j %/33M1010系统管理/admin/M101/index�10fa-cogadminadmin0000-00-00 00:00:002018-04-30 14:52:57
N � ���������~qdWJ=0# ����������zm`SF9,
�
�
�
�
�
�
�
�
�
�
v
i
\
O
B
5
(
� � � � � � � � � { m _ Q C 5 ' � N M1005del M M1004mod L M1004del K M1004add J M1003mod I M1003del H M1003add G M1002mod F M1002del E M1002add D M1000mod C M1000del B M1000add A M160mod @ M160del ? M160add > M158add = M158del < M158mod ; M155add : M155del 9 M155mod 8 M152add 7 M152del 6 M152mod 5 M151add 4 M151del 3 M151mod 2 M150mod 1 M150del 0 M145add / M145del . M145mod - M144add , M144del + M144mod * M143add ) M143del ( M143mod ' M142add & M142del % M142mod $ M141add # M141del " M141mod ! M136add M136del M136mod M132add M132del M132mod M131mod M129add M129del M129mod M120add M120del M120mod M114add M114del M114mod M112mod M111mod M107add M107del
M107mod M105add M105del
M105mod M104add M104del M104mod M103add M103del M103mod M102add M102del M102mod
# ��#: � �
'S| �Q�z
�3�4�4 � 6 �a=+�)indexay_content_sort_filenameay_content_sort5CREATE INDEX "ay_content_sort_filename"
ON "ay_content_sort" ("filename" ASC)y`7+� indexay_content_sort_acodeay_content_sort4CREATE INDEX "ay_content_sort_acode"
ON "ay_content_sort" ("acode" ASC)�_=)�)indexay_content_ext_contentiday_content_ext3CREATE INDEX "ay_content_ext_contentid"
ON "ay_content_ext" ("contentid" ASC)n^3!�indexay_content_subscodeay_content2CREATE INDEX "ay_content_subscode"
ON "ay_content" ("subscode" ASC)h]/!�
indexay_content_statusay_content1CREATE INDEX "ay_content_status"
ON "ay_content" ("status" ASC)k\1!�indexay_content_sortingay_content0CREATE INDEX "ay_content_sorting"
ON "ay_content" ("sorting" ASC)e[-!� indexay_content_scodeay_content/CREATE INDEX "ay_content_scode"
ON "ay_content" ("scode" ASC)nZ3!�indexay_content_filenameay_content.CREATE INDEX "ay_conte�BM�[tableay_siteay_siteCREATE TABLE "ay_site" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"title" TEXT(100) NOT NULL,
"subtitle" TEXT(200) NOT NULL,
"domain" TEXT(50) NOT NULL,
"logo" TEXT(100) NOT NULL,
"keywords" TEXT(200) NOT NULL,
"description" TEXT(500) NOT NULL,
"icp" TEXT(30) NOT NULL,
"theme" TEXT(30) NOT NULL,
"statistical" TEXT(500) NOT NULL,
"copyright" TEXT(200) NOT NULL
)�4L''�'tableay_role_levelay_role_levelCREATE TABLE "ay_role_level" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"rcode" TEXT(20) NOT NULL,
"level" TEXT(50) NOT NULL
)�1K%%�%tableay_role_areaay_role_areaCREATE TABLE "ay_role_area" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"rcode" TEXT(20) NOT NULL,
"acode" TEXT(20) NOT NULL
)�QR�ytableay_useray_user%CREATE TABLE "ay_user" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"ucode" TEXT(20) NOT NULL,
"username" TEXT(30) NOT NULL,
"realname" TEXT(30) NOT NULL,
"password" TEXT(32) NOT NULL,
"status" TEXT(1) NOT NULL,
"login_count" INTEGER(10) NOT NULL,
"last_login_ip" TEXT(11) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)� Q�Wtableay_typeay_type$CREATE TABLE "ay_type" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"tcode" TEXT(20) NOT NULL,
"name" TEXT(30) NOT NULL,
"item" TEXT(30) NOT NULL,
"value" TEXT(20) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)�BP�[tableay_tagsay_tags"CREATE TABLE "ay_tags" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"name" TEXT(50) NOT NULL,
"link" TEXT(200) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)�HO �_tableay_syslogay_syslog CREATE TABLE "ay_syslog" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"level" TEXT(20) NOT NULL,
"event" TEXT(200) NOT NULL,
"user_ip" TEXT(11) NOT NULL,
"user_os" TEXT(30) NOT NULL,
"user_bs" TEXT(30) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL
)�CN �Ytableay_slideay_slideCREATE TABLE "ay_slide" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"acode" TEXT(20) NOT NULL,
"gid" INTEGER(10) NOT NULL,
"pic" TEXT(100) NOT NULL,
"link" TEXT(100) NOT NULL,
"title" TEXT(50) NOT NULL,
"subtitle" TEXT(100) NOT NULL,
"sorting" INTEGER(11) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)
� y +g33R102内容管理员内容管理员具有基本内容管理权限adminadmin2017-06-01 00:32:022019-08-05 11:22:12m +O33R101系统管理员系统管理员具有所有权限adminadmin2017-03-22 11:33:322019-08-05 11:22:02
� ��
R102cn
R101cn
n� y � � � �
7
Q
k
�
�
�
�
� 1 J h � � � � # A ] | � � � �
+
H
e
�
�
�
�
� <Vp�����,F`z����fL2�����tZ@ �����1Rs���� &C`|����
( C _����eI/� �W +R102/admin/Tags/mod�V +R102/admin/Tags/del�U +R102/admin/Tags/add�T /R102/admin/Tags/index�S +R102/admin/Form/mod�R +R102/admin/Form/del�Q +R102/admin/Form/add�P /R102/admin/Form/index�O +R102/admin/Link/mod�N +R102/admin/Link/del�3 1R101/admin/Site/server �2 7R101/admin/Database/index�1 +R101/admin/Type/mod�0 +R101/admin/Type/del�/ +R101/admin/Type/add�. /R101/admin/Type/index�- 3R101/admin/Syslog/index�, +R101/admin/User/mod�+ +R101/admin/User/del�* +R101/admin/User/add�) /R101/admin/User/index�( +R101/admin/Role/mod�' +R101/admin/Role/del�& +R101/admin/Role/add�4 /R102/admin/M110/index�5 /R102/admin/Site/index�6 +R102/admin/Site/mod �7 5R102/admin/Company/index�8 1R102/admin/Company/mod �9 =R102/admin/ContentSort/index �: 9R102/admin/ContentSort/add �; 9R102/admin/ContentSort/del �< 9R102/admin/ContentSort/mod�= /R102/admin/M130/index�> 3R102/admin/Single/index�? /R102/admin/Single/mod �@ 5R102/admin/Content/index�A 1R102/admin/Content/add�B 1R102/admin/Content/del�C 1R102/admin/Content/mod�D /R102/admin/M157/index �E 5R102/admin/Message/index�F 1R102/admin/Message/del�G 1R102/admin/Message/mod�H 1R102/admin/Slide/index�I -R102/admin/Slide/add�J -R102/admin/Slide/del�K -R102/admin/Slide/mod�L /R102/admin/Link/index�M +R102/admin/Link/addj /R101/admin/M156/indexk 3R101/admin/Config/indexl 1R101/admin/Label/indexm -R101/admin/Label/addn -R101/admin/Label/delo -R101/admin/Label/modp 1R101/admin/Model/indexq -R101/admin/Model/addr -R101/admin/Model/dels -R101/admin/Model/mod t 7R101/admin/ExtField/indexu 3R101/admin/ExtField/addv 3R101/admin/ExtField/delw 3R101/admin/ExtField/modx /R101/admin/M110/indexy /R101/admin/Site/indexz +R101/admin/Site/mod { 5R101/admin/Company/index| 1R101/admin/Company/mod } =R101/admin/ContentSort/index ~ 9R101/admin/ContentSort/add 9R101/admin/ContentSort/del � 9R101/admin/ContentSort/mod� /R101/admin/M130/index� 3R101/admin/Single/index� /R101/admin/Single/mod � 5R101/admin/Content/index� 1R101/admin/Content/add� 1R101/admin/Content/del� 1R101/admin/Content/mod� /R101/admin/M157/index � 5R101/admin/Message/index�
1R101/admin/Message/del� 1R101/admin/Message/mod� 1R101/admin/Slide/index�
-R101/admin/Slide/add� -R101/admin/Slide/del� -R101/admin/Slide/mod� /R101/admin/Link/index� +R101/admin/Link/add� +R101/admin/Link/del� +R101/admin/Link/mod� /R101/admin/Form/index� +R101/admin/Form/add� +R101/admin/Form/del� +R101/admin/Form/mod� /R101/admin/Tags/index� +R101/admin/Tags/add� +R101/admin/Tags/del� +R101/admin/Tags/mod� /R101/admin/M101/index� /R101/admin/Area/index� +R101/admin/Area/add� +R101/admin/Area/del� +R101/admin/Area/mod�! /R101/admin/Menu/index�" +R101/admin/Menu/add�# +R101/admin/Menu/del�$ +R101/admin/Menu/mod�% /R101/admin/Role/index
� � �=�: �';c�5
ucnPbootCMS永久开源免费的PHP企业网站开发建设管理系统http://pb.com/static/images/logo.pngcms,免费cms,开源cms,企业cms,建站cmsPbootCMS是一套全新内核且永久开源免费的PHP企业网站开发建设管理系统，是一� � � �
cndefault
\ \ L��)
�
b � � ' � P
� � \ \ �a=+�)indexay_content_sort_filenameay_content_sort5CREATE INDEX "ay_content_sort_filename"
ON "ay_content_sort" ("filename" ASC)y`7+� indexay_content_sort_acodeay_content_sort4CREATE INDEX "ay_content_sort_acode"
ON "ay_content_sort" ("acode" ASC) �=)�)indexay_content_ext_contentiday_content_ext3CREATE INDEX "ay_content_ext_contentid"
ON "ay_content_ext" ("contentid" ASC)n^3!�indexay_content_subscodeay_content2CREATE INDEX "ay_content_subscode"
ON "ay_content" ("subscode" ASC)h]/!�
indexay_content_statusay_content1CREATE INDEX "ay_content_status"
ON "ay_content" ("status" ASC)k\1!�indexay_content_sortingay_content0CREATE INDEX "ay_content_sorting"
ON "ay_content" ("sorting" ASC)e[-!� indexay_content_scodeay_content/CREATE INDEX "ay_content_scode"
ON "ay_content" ("scode" ASC)nZ3!�indexay_content_filenameay_content.CREATE INDEX "ay_content_filename"
ON "ay_content" ("filename" ASC)bY+!�indexay_content_dateay_content-CREATE INDEX "ay_content_date"
ON "ay_content" ("date" ASC)eX-!� indexay_content_acodeay_content,CREATE INDEX "ay_content_acode"
ON "ay_content" ("acode" ASC)^W) �indexay_config_nameay_config+CREATE INDEX "ay_config_name"
ON "ay_config" ("name" ASC)eV-!� indexay_company_acodeay_company*CREATE INDEX "ay_company_acode"
ON "ay_company" ("acode" ASC)XU'}indexay_area_pcodeay_area)CREATE INDEX "ay_area_pcode"
ON "ay_area" ("pcode" ASC)`T'� indexay_area_acodeay_area(CREATE UNIQUE INDEX "ay_area_acode"
ON "ay_area" ("acode" ASC)�1S%%�%tableay_user_roleay_user_role'CREATE TABLE "ay_user_role" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"ucode" TEXT(20) NOT NULL,
"rcode" TEXT(20) NOT NULL
)
� Y Y�� |
q
35cn/static/upload/image/20221108/1667869312906019.jpg#banner1 �adminadmin2022-10-30 08:40:13 2022-11-08 09:01:59 � =
33cn/skin/images/banner3.jpg#banner3 �adminadmin2022-10-30 08:40:252022-10-30 08:40:25 H =
33cn/skin/images/banner2.jpg#banner2 �adminadmin2022-10-30 |
q
35cn/static/upload/image/20221108/1667869326803408.jpg#banner2 �adminadmin2022-10-30 08:40:18 2022-11-08 09:02:07 � ���� ��$ t�w �J
% % �X+&
+
3q
�o
�K35
cn2Company Profile#333333admin本站2022-10-30 11:15:46/static/upload/image/20221108/1667869569139116.jpg<p><strong style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255); font-family: arial, helvetica, sans-serif;">Shenzhen Pengcheng Wanli Semiconductor Co., Ltd</strong><span style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-family: 微软雅黑, &quot;Microsoft YaHei&quot;; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255);">&nbsp;</span><span style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-family: 微软雅黑, &quot;Microsoft YaHei&quot;; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(64, 64, 64); background-color: rgb(255, 255, 255);">是一家专注于高性能半导体芯片设计与研发的创新型企业。总部位于中国深圳，在上海、北京及美国硅谷设有研发中心。公司以“驱动万物互联，赋能人工智能”为使命，致力于为全球客户提供低功耗、高算力的尖端芯片解决方案，覆盖消费电子、汽车电子、物联网（IoT）、5G通信及AI加速等领域。</span><span style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-family: 微软雅黑, &quot;Microsoft YaHei&quot;; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255);">同时为客户提供全套解决方案并秉着客户优先的宗旨以及自主研发、统一生产、灵活销售</span><span style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-family: 微软雅黑, &quot;Microsoft YaHei&quot;; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255);">的特点为客户量身定做产品及个性化服务。“以人为本、依托人才”的管理理念使公司拥有专业的技术骨干、规范的制造管理模式、精锐的营销服务团队；与此同时，以“诚信、创新、合作、共赢”为经营理念，不断提高产品的生产水平以及生产规模，奠定了新一轮发展的坚实基础。“更快捷、更精确、更稳定”是我们对客户的承诺，更是</span><strong style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255); font-family: arial, helvetica, sans-serif;">Shenzhen Pengcheng Wanli Semiconductor Co., Ltd</strong><span style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-family: 微软雅黑, &quot;Microsoft YaHei&quot;; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255);">&nbsp;</span><span style="box-sizing: border-box; outline: 0px; margin: 0px; padding: 0px; font-family: 微软雅黑, &quot;Microsoft YaHei&quot;; font-size: 18px; text-indent: 26px; text-wrap-mode: wrap; color: rgb(101, 101, 101); background-color: rgb(255, 255, 255);">不断追求的目标。</span></p>Shenzhen Pengcheng Wanli Semiconductor Co., Ltd是一家专注于高性能半导体芯片设计与研发的创新型企业。总部位于中国深圳，在上海、北京及美国硅谷设有研发中心。公司以“驱动万物互联，赋能人工智能”为使命，致力于为全球客户提供低功耗、高算力的尖端芯片解决方案，覆盖消费电子、汽车电子、物联网（IoT）、5G通 �10005adminadmin2022-10-30 11:15:46 2025-07-30 11:19:054
� i
y � e < e � ��0$
!
#
3
35
cn13Contact Us#333333AB模板网本站2022-11-07 21:28:19 �1000tadminadmin2022-11-07 21:28:19 2025-07-30 11:18:2140�S/&
#
3
�=
�W35
cn1ABOUT US#333333AB模板网本站2025-07-30 09:55:38<p class="ds-markdown-paragraph" style="font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-seri ��.$
)
3
35
cn15Online Message#333333admin本站2022-10-30 11:15:46 �1000jadminadmin2022-10-30 11:15:46 2025-07-30 09:15:554
+ � �W��N
�
\
� H � � 4
�
|
� h
�N��=�w�M��*�q�`�X � S�J 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 06:42:20S�I 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 06:36:44U�H ; !3info修改文章49成功！245066568Windows 10Chromeadmin2025-07-29 06:20:01U�G ; !3info修改文章50成功！245066568Windows 10Chromeadmin2025-07-29 06:19:11U�F ; !3info修改文章49成功！245066568Windows 10Chromeadmin2025-07-29 06:18:36U�E ; !3info修改文章50成功！245066568Windows 10Chromeadmin2025-07-29 06:18:24S�D 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 06:17:23`�C Q !3info修改数据内容栏目4成功！245066568Windows 10Chromeadmin2025-07-29 05:45:22`�B Q !3info修改数据内容栏目4成功！245066568Windows 10Chromeadmin2025-07-29 05:44:10S�A 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 05:42:48`�@ Q !3info删除数据内容栏目5成功！245066568Windows 10Chromeadmin2025-07-29 05:37:22Y�? C !3info批量删除文章成功！245066568Windows 10Chromeadmin2025-07-29 05:36:57a�> S !3info删除数据内容栏目16成功！245066568Windows 10Chromeadmin2025-07-29 05:35:27a�= S !3info删除数据内容栏目10成功！245066568Windows 10Chromeadmin2025-07-29 05:35:01`�< Q !3info删除数据内容栏目3成功！245066568Windows 10Chromeadmin2025-07-29 05:34:48`�; Q !3info删除数据内容栏目6成功！245066568Windows 10Chromeadmin2025-07-29 05:34:34`�: Q !3info删除数据内容栏目7成功！245066568Windows 10Chromeadmin2025-07-29 05:34:24`�9 Q !3info删除数据内容栏目8成功！245066568Windows 10Chromeadmin2025-07-29 05:34:15Y�8 C !3info批量删除文章成功！245066568Windows 10Chromeadmin2025-07-29 02:35:15T�7 9 !3info删除文章1成功！245066568Windows 10Chromeadmin2025-07-29 02:34:55[�6 G !3info修改单页内容43成功！245066568Windows 10Chromeadmin2025-07-29 02:33:23[�5 G !3info修改单页内容47成功！245066568Windows 10Chromeadmin2025-07-29 02:32:43[�4 G !3info修改单页内容43成功！245066568Windows 10Chromeadmin2025-07-29 02:28:38[�3 G !3info修改单页内容47成功！245066568Windows 10Chromeadmin2025-07-29 02:25:18Y�2 C !3info修改公司信息成功！245066568Windows 10Chromeadmin2025-07-29 02:04:30Y�1 C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:49:17Y�0 C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:42:06Y�/ C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:40:17Y�. C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:37:39Y�- C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:11:12Y�, C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:09:29Y�+ C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:03:34Y�* C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 01:02:01Y�) C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 00:54:02Y�( C !3info修改站点信息成功！245066568Windows 10Chromeadmin2025-07-29 00:51:18K�' ' !3info登录成功!245066568Windows 10Chromeadmin2025-07-29 00:10:36S�& 7 !3info用户资料成功！245066568Windows 10Chromeadmin2025-07-29 00:10:10K�% ' !3info登录成功!245066568Windows 10Chromeadmin2025-07-29 00:05:45Z�$ C!!3info修改参数配置成功！1728159377Windows 10Chromeadmin2025-07-28 20:35:43L�# '!!3info登录成功!1728159377Windows 10Chromeadmin2025-07-28 20:35:37Z�" C!!3info修改参数配置成功！2130706433Windows 10Chromeadmin2025-01-22 11:30:25L�! '!!3info登录成功!2130706433Windows 10Chromeadmin2025-01-22 11:30:21W� =!!3info备份数据库成功！2130706433Windows 10Chromeadmin2022-11-08 09:26:02
� � � � � :<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #fff; --tw-ring-color: rgba(59 �
� � � � � 9<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #fff; --tw-ring-color: rgba(59,130, �
� � a );Z =35cn网站https://www.adminbuy.cn/adminadmin2019-07-12 14:33:13 2022-11-07 22:22:21
D D � �1 �a 8<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #fff; --tw-ring-color: rgba(59,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-back �
H �S��H X %33T101菜单功能导入importadminadmin2017-04-27 07:31:382017-08-09 15:23:27X %33T101菜单功能导出exportadminadmin2017-04-27 07:30:422017-08-09 15:23:29U %33T101菜单功能修改modadminadmin2017-04-27 07:29:342017-08-09 15:23:32U %33T101菜单功能删除deladminadmin2017-04-27 07:29:082017-08-09 15:23:34T % 33T101菜单功能新增addadminadmin2017-04-27 07:28:342017-08-09 15:25:56
� � �
� t
#M
3310001adminAB模板网c3284d0f94606de1fd2af172aba15bf31adminadmin2017-05-08 18:50:302026-08-18 18:20:25
K
����l^F2��������/g=R���� #!ay_message+ay_content_sort ay_model ay_member| +ay_member_group+ay_member_field$ #ay_extfield% ay_user_role
ay_user ay_type
ay_tags ay_slide
ay_site'ay_role_level �%ay_role_area ay_roleS ay_linkS ay_model
)ay_menu_actionN ay_menuB ay_lin
ay_label 'ay_form_field ay_form ! ay_syslogK#!ay_message!ay_contentV")ay_content_ext
ay_config8
! ay_company
ay_area
� ��
10001R101
� � cn
� � 0
� � cn
8 �������yf � � �I5"
� _
� �� K�����|l_M6 { o � � � 8
%$
�
�
�
�
�
�
q
Y
H
6 3tpl_html_cache_time8)tpl_html_cache7+message_rqlogin6#licensecode5 sn_user4sn3;content_keyword_replace2#main_domain1)to_main_domain0 to_https/
spiderlog. lgautosw-+close_site_note,!close_site+ ip_allow* ip_deny)%tpl_html_dir('weixin_secret
%weixin_appid 3watermark_text_size 3watermark_text_font 5watermark_text_color )watermark_text1watermark_position 'watermark_pic )watermark_open!wap_domain #form_status')message_status&'url_rule_type%1smtp_username_test'smtp_username smtp_ssl#smtp_server
smtp_port'smtp_password open_wap)message_verify!+message_send_to /message_send_mail 1message_check_code
lock_time$!lock_count#gzip)form_send_mail+form_check_code" =content_tags_replace_num)baidu_zz_token)baidu_xzh_type+baidu_xzh_token+baidu_xzh_appid!api_secret api_open
api_auth
api_appid-admin_check_code
� � ���������������QJC<5.' ������������������ cn.� cn/cn+cn0X � � � cnVcnUcnTcnScnRcnQcnPcnOcnNcnMcnLcnKcnJ [ � � � � � � � cn<cn;cn:cn9cn8cn7cn6cn5cn4cn3cn2
8 @ �
`
� X p(@Xp
�
� @
�
� � � � �
�
x
0
H������������������� � q A A 8 / 8 32025-07-24 09:45:43L32025-05-30 17:20:08U32025-06-04 12:13:25T32025-07-29 07:07:01732025-07-29 06:42:23532025-07-29 07:02:06632025-07-29 07:12:51832025-07-29 06:36:47432025-07-29 06:20:03332025-07-29 06:08:14232025-07-29 16:31:38<32025-07-30 09:55:38/32022-10-30 11:15:46.32025-07-29 16:29:13;32025-07-29 07:21:18932025-07-12 16:12:51N32025-07-18 15:06:53M32025-07-28 11:46:08K32022-11-07 21:28:19032025-07-29 16:28:31:32025-07-07 08:30:05O� � � 8 � � � � � � 32025-05-22 18:17:30V32025-06-12 17:15:18S32025-06-17 10:43:34R32025-06-27 11:30:49Q32025-07-02 10:19:18P H 0 32025-07-30 11:35:44J32022-10-30 11:15:46+
*C�>jeV[QL`�to�9y��~����4/%%�����������������
.H
V
+
U
T
K
2
/�
8
7
5
6
9
4
3
<
;
L
O
N
M
J
:
P� 2 � < � 7 � � 7 � � � � # �
S
R
Q�
�
0
� �����RLF4:.(" X
����@���|��� 13015.^ 9V2+1/9O9N9M9L9K9J9P4:4;49484745469Q4443424<� | p � � 9U9T9S9R l f ` Z T � � B � � 0 * $
� � ���������������QJC<5.' ������������������ �.� �/ �+ �0X � � � �V �U �T �S �R �Q �P �O �N �M �L �K �J [ � � � � � � � �< �; �: �9 �8 �7 �6 �5 �4 �3 �2
� �XR|F@:(."4d�^L�vjp����� ������������������� # � � 1L1U1T171516181413121<1/1.1;191M1N1K101:1O� 6 H H � N � � � * � 1V1S1R1Q1P 1J1+
*C�>jeV[QL`�to�9y��~����4/%%�����������������
.H
V
+
U
T
K
2
/�
8
7
5
6
9
4
3
<
;
L
O
N
M
J
:
P� 2 � < � 7 � � 7 � � � � # �
S
R
Q�
�
0
� �k �U 7<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #ff �
� � ���������������� � cn� cncn
� � � cn � � cn cn cn
� | �|����tttttttttt [ [
� ! �� � � � contact
news products about C A 8lm4 lm3lm2 lm1�
jianjie gbook
9 ��W��/
�
C � � 4 � x
� � h��,�u�_�� 9��� v � d7+�+indexay_content_sort_scodeay_content_sort9CREATE UNIQUE INDEX "ay_content_sort_scode"
ON "ay_content_sort" ("scode" ASC)yc7+� indexay_content_sort_pcodeay_content_sort8CREATE INDEX "ay_content_sort_pcode"
ON "ay_content_sort" ("pcode" ASC)d�) �indexay_model_mcodeay_modelYCREATE UNIQUE INDEX "ay_model_mcode"
ON "ay_model" ("mcode" ASC)�`� �tableay_modelay_modelICREATE TABLE "ay_model" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"mcode" TEXT(20) NOT NULL,
"name" TEXT(50) NOT NULL,
"type" TEXT(1) NOT NULL,
"urlname" TEXT(100) NOT NULL,
"listtpl" TEXT(50) NOT NULL,
"contenttpl" TEXT(50) NOT NULL,
"status" TEXT(1) NOT NULL,
"issystem" TEXT(1) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)m1%�indexay_user_role_ucodeay_user_roleVCREATE INDEX "ay_user_role_ucode"
ON "ay_user_role" ("ucode" ASC)m~1%�indexay_user_role_rcodeay_user_roleUCREATE INDEX "ay_user_role_rcode"
ON "ay_user_role" ("rcode" ASC)b}-� indexay_user_usernameay_userTCREATE INDEX "ay_user_username"
ON "ay_user" ("username" ASC)`|'� indexay_user_ucodeay_userSCREATE UNIQUE INDEX "ay_user_ucode"
ON "ay_user" ("ucode" ASC)X{'}indexay_type_tcodeay_typeRCREATE INDEX "ay_type_tcode"
ON "ay_type" ("tcode" ASC)Xz'}indexay_tags_acodeay_tagsQCREATE INDEX "ay_tags_acode"
ON "ay_tags" ("acode" ASC)Vy% yindexay_slide_giday_slidePCREATE INDEX "ay_slide_gid"
ON "ay_slide" ("gid" ASC)]x) �indexay_slide_acodeay_slideOCREATE INDEX "ay_slide_acode"
ON "ay_slide" ("acode" ASC)Xw'}indexay_site_acodeay_siteNCREATE INDEX "ay_site_acode"
ON "ay_site" ("acode" ASC)qv3'�indexay_role_level_rcodeay_role_levelMCREATE INDEX "ay_role_level_rcode"
ON "ay_role_level" ("rcode" ASC)mu1%�indexay_role_area_rcodeay_role_areaLCREATE INDEX "ay_role_area_rcode"
ON "ay_role_area" ("rcode" ASC)mt1%�indexay_role_area_acodeay_role_areaKCREATE INDEX "ay_role_area_acode"
ON "ay_role_area" ("acode" ASC)`s'� indexay_role_rcodeay_roleJCREATE UNIQUE INDEX "ay_role_rcode"
ON "ay_role" ("rcode" ASC)� g-!� indexay_message_acodeay_messageHCREATE INDEX "ay_message_acode"
ON "ay_message" ("acode" ASC)up5)�indexay_menu_action_mcodeay_menu_actionFCREATE INDEX "ay_menu_action_mcode"
ON "ay_menu_action" ("mcode" ASC)_o+�indexay_menu_sortingay_menuECREATE INDEX "ay_menu_sorting"
ON "ay_menu" ("sorting" ASC)Xn'}indexay_menu_pcodeay_menuDCREATE INDEX "ay_menu_pcode"
ON "ay_menu" ("pcode" ASC)`m'� indexay_menu_mcodeay_menuBCREATE UNIQUE INDEX "ay_menu_mcode"
ON "ay_menu" ("mcode" ASC)_l+�indexay_link_sortingay_linkACREATE INDEX "ay_link_sorting"
ON "ay_link" ("sorting" ASC)Rk#uindexay_link_giday_link@CREATE INDEX "ay_link_gid"
ON "ay_link" ("gid" ASC)Xj'}indexay_link_acodeay_link?CREATE INDEX "ay_link_acode"
ON "ay_link" ("acode" ASC)wi7'� indexay_form_field_sortingay_form_field>CREATE INDEX "ay_form_field_sorting"
ON "ay_form_field" ("sorting" ASC)qh3'�indexay_form_field_fcodeay_form_field=CREATE INDEX "ay_form_field_fcode"
ON "ay_form_field" ("fcode" ASC)`g'� indexay_form_fcodeay_form<CREATE UNIQUE INDEX "ay_form_fcode"
ON "ay_form" ("fcode" ASC)~ k/#�
indexay_extfield_mcodeay_extfield;CREATE INDEX "ay_extfield_mcode"
ON "ay_extfield" ("mcode" ASC)e;+�%indexay_content_sort_sortingay_content_sort:CREATE INDEX "ay_content_sort_sorting"
ON "ay_content_sort" ("sorting" ASC) yb7+� indexay_content_sort_mcodeay_content_sort7CREATE INDEX "ay_content_sort_mcode"
ON "ay_content_sort" ("mcode" ASC)
� ����������������� 2� � 11
2 1 ) � $ � � � 31
� ����������������� 0� 010
0 0 . � $ � � � � 13
� � ���������������� � 15� 1413
� � � 9 � � 4 2 1
� � ���������������� � �
� � � � �
� ��� 343
� �� 2 1
� � ������ 1 111 1 11
� � ������ � � � � �
� ������ cncncncncn cn
� ������
� ������ � � � � � �
� ������������������ypg^ULC:1(
M1005B M1004A M1003@ M1002? M1001>M160<M158:M1579M1568M1557M1535M1524M1513M1502M136$M132 M131 M130 M129 M112 M111 M110
M109 M108M107M106M105M104M103M102 M101 M1000=
� �j �S 6<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #f �
� �����
�������������wne\SJA8/& M1001B M1001A M1001@ M1001?0>M157=M157<M1574M1573M1572M156:M1567M1565M156$M130 M130 M110 M110 M110 M101 M101M101M101M101M101M101M10109080 0
0
! ������������������=6/(!|ung`YRKD \B[AZ@Y?X>� ������� ��=�<�4�3�2�9� � � / . - ,
�: �7 �$ �5 �8
N
6 ���
�
�
|
r
h
^
T
J
@
6�����������wne\SJA8/& ��������������{ri`WNE<3*!
�
�
�
�
�
�
�
�
�
�
�
� M1005N M1004M M1004L M1004K M1003J M1003I M1003H M1002G M1002F M1002EM160AM160@M160?M158>M158=M158<M155;M155:M1559M1528M1527M1526M1515M1514M1513M1502M1501M1450M145/M145.M144-M144,M144+M143*M143)M143(M142'M142&M142%M141$M141#M141"M136!M136 M136 M132 M132 M132 M131M129M129M129M120M120M120M114M114M114M112M111M107M107M107
M105 M105 M105
M104 M104M104M103M103M103M102M102 M102 M1000D M1000C M1000B
� � � � � ;<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #fff; --tw-ring-color: rg �
� l �
q
�l2d
� B:3
�� �K5$
'
#
3q
35
cn4STM32F401RET6#333333AB模板网本站2025-07-29 06:42:23/static/upload/image/20250729/1753743720861674.png �1110adminadmin2025-07-29 07:02:02 2025-07-30 10:27:2540�K;$
'
#
3q
35
cn4STM32F103C8T6#333333AB模板网本站2025-07-29 16:29:13/static/upload/image/20250729/1753777893458287.png �1110adminadmin2025-07-29 16:31:36 2025-07-30 11:10:1440�K<$
'
#
3q
35
cn4STM32H743VIH6#333333AB模板网本站2025-07-29 16:31:38/static/upload/image/20250729/1753778079961131.png �1110#adminadmin2025-07-29 16:34:42 2025-07-30 09:39:5240�H9$
!
#
3q
35
cn4LIS2DW12TR#333333AB模板网本站2025-07-29 07:21:18/static/upload/image/20250729/1753746510704253.png �1110adminadmin2025-07-29 07:48:32 2025-07-30 11:08:3740�L2$
)
#
3q
35
cn4XC6220B331MR-G#333333AB模板网本站2025-07-29 06:08:14/static/upload/image/20250729/1753741039119012.png �1111,adminadmin2025-07-29 06:17:23 2025-07-30 09:49:5440�K3$
'
#
3q
35
cn4STM32F103RDT6#333333AB模板网本站2025-07-29 06:20:03/static/upload/image/20250729/1753742201206044.png �11114adminadmin2025-07-29 06:36:44 2025-07-30 10:16:0140 � �E:$
#
3q
35
cn4TDA7850#333333AB模板网本站2025-07-29 16:28:31/static/upload/image/20250729/1753777738495229.png �1110adminadmin2025-07-29 16:29:11 2025-07-30 11:11:5740�F8$
#
3q
35
cn4LIS3DHTR#333333AB模板网本站2025-07-29 07:12:51/static/upload/image/20250729/1753744836585008.png �1110adminadmin2025-07-29 07:21:15 2025-07-30 10:31:0040 � �F4$
#
3q
35
cn4ULN2804A#333333AB模板网本站2025-07-29 06:36:47/static/upload/image/20250729/1753742538418633.png �1110adminadmin2025-07-29 06:42:20 2025-07-30 10:23:1440�K7$
'
#
3q
35
cn4STM32F072RBT6#333333AB模板网本站2025-07-29 07:07:01/static/upload/image/20250729/1753744337912672.png �1110adminadmin2025-07-29 07:12:48 2025-07-30 10:29:0540�K6$
'
#
3q
35
cn4STM8S003K3T6C#333333AB模板网本站2025-07-29 07:02:06/static/upload/image/20250729/1753743896273936.png �1110 adminadmin2025-07-29 07:04:58 2025-07-30 10:26:5740
�
� g�
�3�
� j '-353产品2listlist_pro.htmlarticle_pro.html10adminadmin2018-04k ++356伙伴2listrongyulist.htmlrongyulist.html00adminadmin2022-11-08 09:22:55 2025-07-29 17:14:05i -%353产品2listproductlist.htmlproduct.html10adminadmin2018-04-11 17:17:46 2022-11-07 16:54:08X
!351专题1pagesabout.html11adminadmin2018-04-11 17:16:01 2022-11-07 16:53:51 k ++355荣誉2listrongyulist.htmlrongyulist.html00adminadmin2022-11-07 19:56:26 2025-07-29 17:14:04g ''354下载2listdownlist.htmldownlist.html00adminadmin2018-04-11 17:19:53 2025-07-29 17:14:03b ' 332新闻2listnewslist.htmlnews.html11adminadmin2018-04-11 17:17:162019-08-05 11:12:04
� �� R102 R101
� �� cncn
� �� R102R101
n � ��������������ypg^ULC:0& �������������|rh^TJ@6,"
�
�
�
�
�
�
�
�
�
�
�
�
�
x
n
d
Z
P
F
<
2
(
� � � � � � � � � � � � ~ t j ` V L B 8 . $ � � � � � � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R102 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 � R101 �R101R101~R101}R101|R101{R101zR101yR101xR101wR101vR101uR101tR101sR101rR101qR101pR101oR101nR101mR101lR101kR101j
� � cn
� ��� cncncn
� ���
� � cn
� ����� T101T101T101T101 T101
� �� 10001
� �� admin
� �� R101
� �� 10001
� � �x �o <<p class="ds-markdown-paragraph" style="font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255); margin-top: 0px !important;"><span style="font-weight: 600;">Brand Name</span><br/>ST (STMicroelectronics)</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Product Model</span><br/>STM32H743VIH6</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Product Number</span><br/>C964146</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Package Type</span><br/>TFBGA-100</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600;">Packaging Method</span><br/>Tray</p><p class="ds-markdown-paragraph" style="font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255); margin-bottom: 0px !important;"><span style="font-weight: 600;">Gross Weight</span><br/>2.6 g (grams)</p><p><br/></p>
� � � ��oN(
�?
#
3
��A
�e35
cn9Knocking on the door of L3, where will the new decade of intelligent assisted driving go?#333333AB模板网本站2025-07-12 16:12:51<p>At the inception of any technology, debates surrounding its development path always arise, and intelligent assisted driving technology is no exception.</p><p>Should we opt for a gradual progression from L2 to L4, or leapfrog directly to the L4 endpoint? Should we pursue the ultimate simplification of a pure vision technology route, or adhere to the redundant safety of multi-sensor fusion? How can we balance the unification of industry standards with the customization of automobile companies? Where will the game between vertical integration and open alliances lead?</p><p>These unresolved propositions precisely demonstrate the vigorous vitality of technological innovation. Through continuous exploration and practical verification, the future evolution roadmap of intelligent assisted driving is gradually becoming clearer.</p><p>The leap in computing power under the three-tier architecture</p><p>Over the past deca w�vM&
#
3q
�'
�m35
cn9Meta is testing its first self-developed AI training chip#333333AB模板网本站2025-07-18 15:06:53/static/upload/image/20250730/1753855715737801.png<p><span style="text-wrap-mode: nowrap;"><br/></span></p><p>Breaking News: Another tech giant is accelerating its AI R&amp;D process. This time, the protagonist is Meta. According to reports, the company is testing its first self-developed AI training chip. This move aims to reduce its huge infrastructure costs and decrease its dependence on NVIDIA (it is said that NVIDIA often makes Zuckerberg use &quot;adult language&quot;). If everything goes smoothly, Meta hopes to use it for training in 2026.</p><p>According to reports, Meta has begun a small-scale deployment of this dedicated accelerator chip, which is specifically designed for AI tasks (hence more energy-efficient than general-purpose NVIDIA GPUs). Immediately after completing its first &quot;tape-out&quot;, the company began deploying it. Tape-out is a stage in silicon chip development where the complete design is sent for manufacturing testing.</p><p>This chip belongs to the Meta Training and Inference Accelerator (MTIA) series, which is the company&#39;s self-developed chip series focusing on generative AI, recommendation systems, and advanced research.</p><p>Last year, the company began using MTIA chips for inference, which is the prediction process carried out by AI models in the background. Meta has already started using inference chips in the news feed recommendation systems of Facebook and Instagram. According to reports, the company also plans to start using training chips. The long-term plan for both types of chips is said to start with recommendation systems and eventually be used in generative products like Meta&#39;s AI chatbots.</p><p>After ordering billions of dollars worth of GPUs in 2022, the company became one of NVIDIA&#39;s largest customers. This marked a turning point for Meta, as the company had previously abandoned a self-developed inference chip that failed in small-scale test deployments - similar to the training chip tests currently underway.</p><p><br/></p>Breaking News: Another tech giant is accelerating its AI R&amp;D process. This time, the protagonist is Meta. According to reports, the company is testing its first self-developed AI training chip. This move aims to reduce its huge infrastructure costs and decrease its dependence on NVIDIA (it is sa �1000<adminadmin2025-07-30 14:08:37 2025-07-30 15:26:1840
� ������ 65432 1
� � � v %U333高级会员高级会员具备全部特殊权限1' T �adminadmin2020-06-25 00:00:002020-06-25 00:00:00r %U332中级会员中级会员具备部分特殊权限1�'adminadmin2020-06-25 00:00:002020-06-25 00:00:00m %O331初级会员初级会员具备基本的权限1�adminadmin2020-06-25 00:00:002020-06-25 00:00:00
F � u ��x����B�g
�
� }�9/� indexay_member_comment_puiday_member_commentmCREATE INDEX "ay_member_comment_puid" ON "ay_member_comment" ("puid" ASC)z�7/�indexay_member_comment_uiday_member_commentlCREATE INDEX "ay_member_comment_uid" ON "ay_member_comment" ("uid" ASC)� �C/�3indexay_member_comment_contentiday_member_commentkCREATE INDEX "ay_member_comment_contentid" ON "ay_member_comment" ("contentid" ASC)z�
7/�indexay_member_comment_piday_member_commentjCREATE INDEX "ay_member_comment_pid" ON "ay_member_comment" ("pid" ASC)�o� //�
tableay_member_commentay_member_commentiCREATE TABLE "ay_member_comment" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"pid" INTEGER(10) NOT NULL DEFAULT 0,
"contentid" INTEGER(10) NOT NULL,
"comment" TEXT(1000) NOT NULL,
"uid" INTEGER(10) NOT NULL,
"puid" INTEGER(10) NOT NULL,
"likes" INTEGER(10) NOT NULL DEFAULT 0,
"oppose" INTEGER(10) NOT NULL DEFAULT 0,
"status" TEXT(1) NOT NULL,
"user_ip" TEXT(11) NOT NULL,
"user_os" TEXT(30) NOT NULL,
"user_bs" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_user" TEXT(30) NOT NULL,
"update_time" TEXT NOT NULL
)�M� ++�Qtableay_member_fielday_member_fieldhCREATE TABLE "ay_member_field" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"name" TEXT(30) NOT NULL,
"length" INTEGER(10) NOT NULL,
"required" TEXT(1) NOT NULL,
"description" TEXT(30) NOT NULL,
"sorting" INTEGER(10) NOT NULL,
"status" TEXT(1) NOT NULL,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
) � �
� ��� 32 1
ions and traffic rules. It may decide to immediately slow down and stop, wait for the pedestrian to cross the road and the signal light to turn green, and then continue driving.</p><p>Most importantly, the reasoning process of VLA is fully differentiable, enabling it to explain the driving logic to users through on-board displays, thereby enhancing user trust.</p><p>Industry insiders have indicated that the VLA model holds significant importance for the evolution of intelligent driving. With its enhanced ability to comprehend the world end-to-end, VLA may become the cornerstone of the next generation of intelligent assisted driving technology, facilitating the leap from L2 assisted driving to L4 autonomous driving.</p><p>However, in the short term, VLA faces two major challenges in transitioning from &quot;usable&quot; to &quot;easy to use&quot;. Firstly, there is insufficient computing power at the vehicle end. For example, the dual Orin-X chips in the ideal vehicle already struggle to run &quot;end-to-end + VLM&quot; computations. The model structure of VLA is more complex, and the parameters will be further expanded, placing higher demands on hardware computing power and the iterative capability of the data loop.</p><p>In the future, enterprises with stronger chip integration and vertical integration capabilities are expected to gain a first-mover advantage in the competition for VLA technology. Currently, leading industry players such as Tesla, Huawei, Xiaopeng, Li Xiang, and Momenta are gradually moving towards the integration of hardware and software, developing more suitable intelligent driving chips for their proprietary models, allowing for maximum optimization.</p><p>Another major challenge lies in how to deeply integrate data and information from end-to-end and multimodal large models. This tests the intelligent driving team&#39;s ability to define model frameworks and rapidly iterate models.</p><p>To address the aforementioned issues, various technical approaches are being explored. For instance, introducing interpretable modules or posterior visualization tools to enhance the transparency of the decision-making process; utilizing Diffusion models to optimize trajectory generation, ensuring the smoothness and stability of control instructions. Simultaneously, combining VLA with traditional rule engines or Model Predictive Control (MPC) to improve safety redundancy and system robustness through a hybrid architecture has also become a popular direction.</p><p>With the continuous advancements in large model technology, edge computing, and in-vehicle hardware, VLA is poised to play a more pivotal role in the field of intelligent assisted driving. It not only offers a smarter driving solution for complex urban roads but also extends to various application scenarios such as fleet coordination, remote control, and human-machine interaction.</p><p>The technological revolution triggered by the VLA model is reshaping the landscape of the intelligent assisted driving industry. The ultimate outcome of this revolution may not be the complete victory of a certain technological path, but rather the emergence of a tiered market. Of course, it is still uncertain whether VLA can become the &quot;ultimate destination&quot; of intelligent assisted driving. From CNN to Transformer, and then to VLM and VLA, the speed of technological iteration is astonishing, and new breakthroughs may already be brewing in the future. When software-defined vehicles enter the 2.0 era, the real competition has just begun.</p><p><br/></p>Recently, the Gemini family welcomed a new member - Gemini Robotics On-Device. This is Google DeepMind&#39;s first Vision-Language-Action (VLA) model that can be directly deployed on robots. The model does not rely on data networks during operation and can run entirely offline on the robot device lo �1000adminadmin2025-07-30 14:27:32 2025-07-30 15:27:3640 `isted driving technology</p><p>The working principle of the VLA model can be divided into three main steps: visual perception, language understanding and decision generation, and action control, supported by multiple key modules, including visual encoder, language encoder, cross-modal fusion module, and action generation module.</p><p>The visual encoder is responsible for extracting high-level visual features from images or videos, while the language encoder processes natural language inputs. The cross-modal fusion module integrates visual and language features, and the action generation module generates control commands for the vehicle based on the fused information.</p><p>In the visual perception stage, sensors such as cameras and radars of the vehicle act like the &quot;eyes&quot; of the VLA model. They continuously collect visual information around the vehicle, including road conditions, traffic signs, and the locations of other vehicles and pedestrians. These raw data are input into the model&#39;s visual processing module, which utilizes advanced deep learning algorithms to extract and analyze image features, transforming complex visual scenes into feature vectors that computers can understand.</p><p>For example, it can recognize the red round traffic signal ahead, determine the speed and direction of the vehicle next to it, and detect pedestrians on the side of the road approaching the roadway.</p><p>The language comprehension and decision-making generation stage is the core operational part of the &quot;brain&quot; of the VLA model. After visual information is processed, the model combines language instructions and its own knowledge base for analysis and reasoning.</p><p>If the passenger issues a voice command of &quot;turn right at the next intersection&quot;, the VLA model will first understand this linguistic information and then integrate it with the current visual perception information.</p><p>It analyzes the traffic conditions at the intersection ahead, such as the presence of vehicles and pedestrians, the status of traffic lights, etc. Meanwhile, it considers traffic rules and safety factors, employs &quot;chain of thought&quot; technology to conduct human-like reasoning, and ultimately generates reasonable driving decisions. In this example, the model may decide to slow down in advance, turn on the right turn signal, and complete the right turn maneuver at the appropriate time, ensuring safety.</p><p>The action control phase is the process where the VLA model converts decisions into actual actions. The driving decisions generated by the model, such as acceleration, deceleration, steering, and other commands, are sent to the vehicle&#39;s execution system, just like giving operational commands to the various &quot;organs&quot; of the vehicle.</p><p>Based on these instructions, the execution system precisely controls the vehicle&#39;s throttle, brakes, steering wheel, and other components to ensure safe driving. For example, when the model decides to slow down, the execution system will control the braking system to gradually reduce the vehicle&#39;s speed; when a turn is required, it will precisely adjust the angle of the steering wheel to ensure that the vehicle follows the predetermined trajectory.</p><p>Taking the driving scenario in complex urban road conditions as an example, the working process of the VLA model becomes clearer. When a vehicle is driving on a crowded urban street, there may be various vehicles, pedestrians, bicycles, as well as constantly changing traffic lights and complex traffic signs on the road.</p><p>After perceiving this information through cameras and radar, the VLA model quickly analyzes the scene. If it encounters a pedestrian crossing the road ahead and the traffic light is about to turn red, the model will understand this complex situation and make decisions based on language instruct aes as an auxiliary system, providing understanding and semantic analysis of complex traffic scenarios, but the two are relatively independent.</p><p>For example, the ideal &quot;end-to-end + VLM&quot; dual-system architecture solution is based on Daniel Kahneman&#39;s theory of two human thinking systems proposed in &quot;Thinking, Fast and Slow&quot;. It integrates the end-to-end system (equivalent to System 1) and the VLM model (equivalent to System 2) into the autonomous driving technology solution, endowing the vehicle-end model with higher performance limits and development potential.</p><p>Among them, System 1, also known as the end-to-end model, is an intuitive and fast-reacting mechanism that directly maps sensor inputs (such as camera and lidar data) to driving trajectory outputs without any intermediate processes. It is an integrated One Model. System 2, on the other hand, is implemented by a large visual-language model (VLM) with 2.2 billion parameters. Its output is integrated with System 1 to form the final driving decision.</p><p>Xpeng Motors divides its cloud-based model factory into four workshops, which carry out model pre-training, post-training, model distillation, and vehicle deployment in sequence. Ideal AI, on the other hand, chooses to first pre-train a visual-language base model, then proceed with model distillation, and finally conduct post-training and reinforcement learning using driving scene data. These two different technical approaches result in varying training costs and efficiencies, which is precisely what creates a stark contrast between the two automakers in the market.</p><p>Although &quot;end-to-end + VLM&quot; has significantly improved the level of intelligent driving, there are still many issues. For example, it is difficult to conduct joint training for end-to-end and VLM. In addition, there are problems such as insufficient understanding of 3D space, insufficient driving knowledge and memory bandwidth, and difficulty in dealing with the multimodal nature of human driving.</p><p>The VLA, through a unified large model architecture, seamlessly connects perception, decision-making, and execution, forming a closed loop of &quot;image input - semantic understanding - human-like decision-making - action output&quot;. It can simultaneously improve both the upper and lower limits of intelligent driving, achieving the unification of space, behavior, and language.</p><p>In terms of reasoning, the VLA model&#39;s capabilities are significantly superior to &quot;end-to-end + VLM&quot;. VLA integrates the perception capabilities of VLM and the decision-making abilities of the end-to-end model, and also incorporates &quot;thought chain&quot; technology. This endows it with global context understanding and human-like reasoning abilities, enabling it to think and make judgments like a human driver when facing special scenarios such as complex traffic rules, tidal lanes, and long-term reasoning.</p><p>For example, in terms of reasoning duration, traditional rule-based solutions can only reason about road conditions for 1 second and make decision control; the system in the end-to-end 1.0 stage can reason about road conditions for the next 7 seconds, while the VLA model can reason about road conditions for tens of seconds, significantly enhancing the decision-making ability and adaptability of intelligent assisted driving systems.</p><p>For this reason, VLA is considered by the industry as the primary technological form of end-to-end 2.0. Currently, VLA is still in its development stage, including DeepMind&#39;s RT-2, OpenVLA model, Waymo&#39;s EMMA, Wayve&#39;s LINGO-2, NVIDIA&#39;s NaVILA, etc. Among them, Waymo&#39;s EMMA and Wayve&#39;s LINGO-2 are mainly targeted at the automotive field, while RT-2, OpenVLA, and NaVILA are mainly targeted at the robotics field.</p><p>The cornerstone of next-generation ass bping of actions, thereby endowing the model with robust 3D spatial understanding, logical reasoning, and behavior generation capabilities, enabling autonomous driving to perceive, think, and adapt to the environment.</p><p>In the field of intelligent assisted driving, perception technology typically relies on various sensors such as radar, lidar, and cameras for perception. The perception results undergo a series of module processing, including object detection, semantic segmentation, trajectory prediction, and behavior planning, before the controller issues action commands such as steering and accelerator. Although the entire process is clear and logical, there are shortcomings such as error accumulation between modules, complex rule design, and difficulty in covering all extreme scenarios.</p><p>The VLA model emerged against this backdrop. It eliminates the intermediate manual design algorithm and directly employs a unified neural network to learn optimal control strategies from multimodal inputs. Leveraging language models, it comprehends human instructions and generates interpretable decision-making processes. Ultimately, it converts multimodal information into specific driving operation instructions, simplifying the system architecture and enhancing data utilization efficiency.</p><p>Key springboard for intelligent assisted driving</p><p>This year, the battle for intelligent driving in the automotive industry is clearly more intense than in previous years. BYD&#39;s &quot;Eye of the God&quot;, Geely&#39;s &quot;Thousand Miles of Vastness&quot;, Chery&#39;s &quot;Falcon Intelligent Driving&quot;, and GAC&#39;s autonomous driving plan, all indicate that the current automotive industry is entering an era where &quot;he who masters intelligent driving will dominate&quot;.</p><p>Since 2023, following the wave of Bird&#39;s Eye View (BEV) and end-to-end technology in the intelligent driving industry, automakers are gradually integrating AI neural networks into perception, planning, control, and other aspects. Compared to traditional rule-based solutions, the &quot;end-to-end&quot; approach based on AI and data-driven has a higher ceiling of capability.</p><p>However, beyond the end-to-end model, car companies have also supplemented it with external plugins such as large language models and VLM models to provide stronger environmental understanding capabilities, thereby raising the upper limit of intelligent driving capabilities.</p><p>Meanwhile, VLA is becoming an important link. The VLA model possesses higher scene reasoning and generalization capabilities, which is of great significance for the evolution of intelligent assisted driving technology. In the long run, VLA is expected to become a key stepping stone in the technological transition from L2 assisted driving to L4 autonomous driving.</p><p>In terms of enhancing vehicle intelligence, new-energy vehicle companies are the most aggressive. At the NVIDIA GTC 2025 conference, Ideal Automobile unveiled its next-generation autonomous driving architecture, MindVLA. By integrating spatial intelligence, language intelligence, and behavioral intelligence, it endows the autonomous driving system with 3D spatial understanding, logical reasoning, and behavior generation capabilities. It is planned to be mass-produced and applied in 2026.</p><p>Before the VLA, &quot;end-to-end + VLM&quot; had been the mainstream technical solution in the intelligent driving industry. Because a multimodal perception interaction system is required during driving, and the user&#39;s vision, hearing, changes in the surrounding environment, and even personal emotional fluctuations are closely related to driving behavior, in the technical architecture of &quot;end-to-end + VLM&quot;, the end-to-end system is responsible for handling the entire process of perception, decision-making, and execution, while VLM serv
�c || � E 33qq0QQ �1adminadmin2020-06-25 00:00:002020-06-25 00:00:00O 33birthday0生日 �1adminadmin2020-06-25 00:00:002020-06-25 00:00:00J 33sex0性别 �1adminadmin2020-06-25 00:00:002020-06-25 00:00:00
K U � N � �
�
+ � i�7���U� �>�%!!�Gtableay_messageay_messageCREATE TABLE ay_message (id INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL, acode TEXT (20) NOT NULL, conten�(�%!!�tableay_messageay_messageCREATE TABLE ay_message (id INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL, acode TEXT (20) NOT NULL, content TEXT (500), user_ip TEXT (11) NOT NULL, user_os TEXT (30) NOT NULL, user_bs TEXT (30) NOT NULL, recontent TEXT (500) NOT NULL, status TEXT (1) NOT NULL, create_user TEXT (30) NOT NULL, update_user TEXT (30) NOT NULL, create_time TEXT NOT NULL, update_time TEXT NOT NULL, uid INTEGER (10) NOT NULL DEFAULT 0, name TEXT(20) NULL, phone TEXT(20) NULL, company TEXT(20) NULL, email TEXT(20) NULL, cont TEXT(400) NULL)_�&-!indexay_message_acodeay_message
CREATE INDEX ay_message_acode ON ay_message ("acode" ASC)K�O�"))�Ytableay_content_extay_content_extCREATE TABLE ay_content_ext (extid INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL, contentid INTEGER (10) NOT NULL, ext_cpms TEXT(10000) NULL, ext_wenjian TEXT(200) NULL)|�#=)� indexay_content_ext_contentiday_content_extCREATE INDEX ay_content_ext_contentid ON ay_content_ext ("contentid" ASC)�� ##�gtableay_extfielday_extfield CREATE TABLE ay_extfield (id INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL, mcode TEXT (20) NOT NULL, name TEXT (30) NOT NULL, type TEXT (2) NOT NULL, value TEXT (500) NOT NULL, description TEXT (30) NOT NULL, sorting INTEGER (11) NOT NULL) a;;� tablesqlitestudio_temp_tablesqlitestudio_temp_tablerCREATE TABLE sqlitestudio�" d� /#�indexay_extfield_mcodeay_extfield;CREATE INDEX ay_extfield_mcode ON ay_extfield ("mcode" ASC)� � 7+�+indexay_member_group_gcodeay_member_group^CREATE UNIQUE INDEX "ay_member_group_gcode"
ON "ay_member_group" ("gcode" ASC)�N� ++�Stableay_member_groupay_member_groupZCREATE TABLE "ay_member_group" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"gcode" TEXT(20) NOT NULL,
"gname" TEXT(100) NOT NULL,
"description" TEXT(200) NOT NULL,
"status" TEXT(1) NOT NULL,
"lscore" INTEGER NOT NULL DEFAULT 0,
"uscore" INTEGER NOT NULL DEFAULT 0,
"create_user" TEXT(30) NOT NULL,
"update_user" TEXT(30) NOT NULL,
"create_time" TEXT NOT NULL,
"update_time" TEXT NOT NULL
)^�) �indexay_member_wxiday_memberqCREATE INDEX "ay_member_wxid"
ON "ay_member" ("wxid" ASC)^�) �indexay_member_wbiday_memberoCREATE INDEX "ay_member_wbid"
ON "ay_member" ("wbid" ASC)q�1 � indexay_member_usernameay_membernCREATE UNIQUE INDEX "ay_member_username"
ON "ay_member" ("username" ASC)p�5 �indexay_member_usermobileay_membergCREATE INDEX "ay_member_usermobile"
ON "ay_member" ("usermobile" ASC)m�3 �indexay_member_useremailay_memberfCREATE INDEX "ay_member_useremail"
ON "ay_member" ("useremail" ASC)h�+ �indexay_member_ucodeay_membereCREATE UNIQUE INDEX "ay_member_ucode"
ON "ay_member" ("ucode" ASC)^�) �indexay_member_qqiday_member_CREATE INDEX "ay_member_qqid"
ON "ay_member" ("qqid" ASC)Z�' }indexay_member_giday_member]CREATE INDEX "ay_member_gid"
ON "ay_member" ("gid" ASC)�Q� �qtableay_memberay_member[CREATE TABLE "ay_member" (
"id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
"ucode" TEXT(20) NOT NULL,
"username" TEXT(100) NOT NULL,
"usermobile" TEXT(11) NOT NULL DEFAULT '',
"useremail" TEXT(50) NOT NULL DEFAULT '',
"nickname" TEXT(100) NOT NULL,
"password" TEXT(32) NOT NULL,
"headpic" TEXT(200) NOT NULL,
"status" TEXT(1) NOT NULL,
"gid" TEXT(20) NOT NULL,
"wxid" TEXT(50) NOT NULL,
"qqid" TEXT(50) NOT NULL,
"wbid" TEXT(50) NOT NULL,
"score" INTEGER NOT NULL DEFAULT 0,
"register_time" TEXT NOT NULL,
"login_count" INTEGER NOT NULL DEFAULT 0,
"last_login_ip" TEXT(11) NOT NULL,
"last_login_time" TEXT(11) NOT NULL,
"sex" TEXT(2) NOT NULL DEFAULT '',
"birthday" TEXT(20) NOT NULL DEFAULT '',
"qq" TEXT(15) NOT NULL DEFAULT '',
"activation" TEXT(1) NOT NULL DEFAULT 1
)
G Gu� �3T'
�
#
3
�k
�m35
cn9The world's largest chip sets AI speed record, surpassing NVIDIA#333333AB模板网本站2025-06-04 12:13:25<p>Today, I personally held the world&#39;s largest computer chip. Although its volume is impressive, its speed is even more astonishing and important. Most computer chips are very small, only the size of stamps or even smaller. In contrast, the Cerebras WSE (Wafer Scale Engine) is a huge square with each side measuring 8.5 inches or 22 centimeters, and the latest model has an astonishing 4 billion transistors on a single chip. It is precisely because of these billions of transistors that WSE has set a world speed record for AI inference operations, about 2.5 times faster than an NVIDIA cluster with roughly equivalent performance.</p><p>Th ��
V'
�]
#
3
�S
�m35
cn9The Ministry of National Security: Beware of "electronic spies" and prevent and resist the theft of secrets through hidden "technological backdoors"#333333AB模板网本站2025-05-22 18:17:30<p>In today&#39;s highly digitized era, the importance of network security is becoming increasingly prominent, not only related to personal privacy and corporate secrets, but even affecting national security. It should be note ��U'
�
#
3
�+
�e35
cn9CCTV cites foreign media: US chip industry collectively 'turns around', AMD will restart exporting AI chips to China#333333AB模板网本站2025-05-30 17:20:08<p>According to reports from multiple media outlets such as Consumer and Business Channel and Bloomberg, American chip company AMD has announced that it will soon resume delivering its artificial intelligence chip MI308 to the Chinese market. Previously, the United States made a similar decision regarding a chip from Nvidia.</p><p>According to reports, a spokesperson for Supermicro Semiconductor stated on July 15th that the US Department of Commerce has informed the company that the license application for MI308 products will enter the review process.</p><p>According to reports, the US chip industry has recently collectively called on the US government to relax export controls and warned that restrictions on China will result in revenue losses.</p><p>In April of this year, Supermicro Semiconductor announced that export restrictions on MI308 chips would result in a loss of approximately $800 million for the company.</p><p>On the 15th of this month, Huang Renxun, founder and CEO of Nvidia, revealed in an interview with CCTV reporters that the United States has approved the sale of H20 chips to China, and the company will also launch a new RTX PRO graphics processor (GPU) for the Chinese market.</p><p>Nvidia&#39;s H20 chip and AMD&#39;s MI308 chip are both developed for the Chinese market to comply with US export control regulations. Earlier this year, the Trump administration restricted the export of artificial intelligence chips to China citing concerns about national security.</p><p><br/></p>According to reports from multiple media outlets such as Consumer and Business Channel and Bloomberg, American chip company AMD has announced that it will soon resume delivering its artificial intelligence chip MI308 to the Chinese market. Previously, the United States made a similar decision regard �1000adminadmin2025-07-30 15:18:22 2025-07-30 15:29:3140
- c �T��E
�
�
X
� V � � 8
�
�
2 � � .��*�~(�|&�x �p�V��+�) � c `�w Q !3info修改数据内容栏目4成功！245066568Windows 10Chromeadmin2025-07-29 17:50:50`�v Q !3info修改数据内容栏目1成功！245066568Windows 10Chromeadmin2025-07-29 17:49:14S�u 7 !3info清理缓存成功！245066568Windows 10Chromeadmin2025-07-29 17:19:17S�t 7 !3info清理缓存成功！245066568Windows 10Chromeadmin2025-07-29 17:19:15S�s 7 !3info用户资料成功！245066568Windows 10Chromeadmin2025-07-29 17:18:46[�r G !3info修改单页内容47成功！245066568Windows 10Chromeadmin2025-07-29 17:16:02`�q Q !3info修改数据内容栏目4成功！245066568Windows 10Chromeadmin2025-07-29 17:12:40g�p _ !3info修改数据内容栏目4状态1成功！245066568Windows 10Chromeadmin2025-07-29 17:11:36g�o _ !3info修改数据内容栏目4状态0成功！245066568Windows 10Chromeadmin2025-07-29 17:11:14U�n ; !3info修改文章51成功！245066568Windows 10Chromeadmin2025-07-29 17:09:57U�m ; !3info修改文章65成功！245066568Windows 10Chromeadmin2025-07-29 17:07:52U�l ; !3info修改文章71成功！245066568Windows 10Chromeadmin2025-07-29 17:06:15U�k ; !3info修改文章72成功！245066568Windows 10Chromeadmin2025-07-29 17:04:19U�j ; !3info修改文章73成功！245066568Windows 10Chromeadmin2025-07-29 17:03:24U�i ; !3info修改文章73成功！245066568Windows 10Chromeadmin2025-07-29 17:02:22S�h 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 17:01:09S�g 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:59:18S�f 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:57:33S�e 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:56:33S�d 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:55:03S�c 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:53:59S�b 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:52:33S�a 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:50:52U�` ; !3info修改文章65成功！245066568Windows 10Chromeadmin2025-07-29 16:49:10S�_ 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:48:33S�^ 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:46:38S�] 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:44:56U�\ ; !3info修改文章61成功！245066568Windows 10Chromeadmin2025-07-29 16:43:58U�[ ; !3info修改文章62成功！245066568Windows 10Chromeadmin2025-07-29 16:43:25S�Z 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:42:58U�Y ; !3info修改文章61成功！245066568Windows 10Chromeadmin2025-07-29 16:42:13S�X 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:41:13a�W S !3info删除数据内容栏目11成功！245066568Windows 10Chromeadmin2025-07-29 16:37:28a�V S !3info删除数据内容栏目12成功！245066568Windows 10Chromeadmin2025-07-29 16:36:08S�U 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:34:42S�T 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:31:36S�S 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 16:29:11K�R ' !3info登录成功!245066568Windows 10Chromeadmin2025-07-29 16:28:16F�Q ' !
3info登录失败!245066568Windows 10Chrome2025-07-29 16:27:53S�P 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 07:48:32S�O 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 07:21:15S�N 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 07:12:48`�M Q !3info修改数据内容栏目4成功！245066568Windows 10Chromeadmin2025-07-29 07:06:37S�L 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 07:04:58S�K 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-29 07:02:02
X �j �S 4<p>describe</p><p>8 Darlington arrays</p><p>Brand Name</p><p>ST (STMicroelectronics)</p><p>marque</p><p>ULN2804A</p><p>item code</p><p>C31415</p><p>Product packaging</p><p>DIP-18</p><p>Packaging method</p><p>Pipe installation</p><p>Gross weight of the product</p><p>2.409 grams (g)</p><p><br/></p> ybut also implies a value leap from &quot;specific scenario assistance&quot; to &quot;intelligent empowerment throughout the entire journey&quot; in high-level intelligent driving. Its core driving force stems from the ternary synergy breakthrough of algorithms, data, and computing power.</p><p>The computing power side achieves a quantum leap through a three-tier architecture, while the data side relies on both real and synthetic data to drive breakthroughs in long-tail scenarios. The algorithm side evolves from &quot;rule engine + module stacking&quot; to &quot;data engine + end-to-end integration&quot;, completing the full-chain optimization of &quot;perception-decision-control&quot;.</p><p>In 2025, the trend of &quot;technological equality&quot; in high-level intelligent driving accelerated significantly. BYD extended the high-speed NOA function to models priced at 80,000 yuan, while XPeng extended the city NOA coverage to the 150,000 yuan market segment. Other mainstream car companies also lowered the price of city NOA function to models priced between 100,000 and 200,000 yuan, highlighting the increasing technological equality in high-level intelligent assisted driving.</p><p>The realization of equal rights for high-level intelligent assisted driving technology is the result of the resonance of three forces: software efficiency enhancement, hardware cost reduction, and scale expansion. Driven by the equal rights for technology, the market size and penetration rate of L2 and L2+ high-level intelligent assisted driving have experienced explosive growth. It is expected that the penetration rate of L2+ will jump from 8% in 2024 to 15% in 2025, and the high-level intelligent driving market for passenger vehicles is ushering in an explosive growth cycle.</p><p>Data closed loop promotes technology application to &quot;reduce costs&quot;</p><p>Currently, the intelligent driving assistance system is undergoing a paradigm shift from &quot;hardware redundancy stacking&quot; to &quot;algorithmic perception definition&quot;, with the data closed-loop ecosystem becoming the core carrier of this transformation. From the moment the multi-modal sensors on the vehicle are activated to collect data, the gears of the entire &quot;data machine&quot; begin to rotate, mesh, and operate in coordination.</p><p>Firstly, the data collected by sensors is standardized on the vehicle side, preprocessed through caching, and then automatically annotated with metadata based on driving behavior, environmental parameters, and real-time labels of target objects. When the labels meet specific conditions, event packaging is triggered.</p><p>After the data is transmitted to the edge, the rule engine and lightweight model will filter it based on different levels of accuracy, ultimately obtaining high-value data related to intelligent driving. According to the algorithm, such data will be compressed and optimized for hierarchical storage, and categorized into upload queues with different priorities based on urgency. During this process, the data will also undergo desensitization, and the transmission process will be monitored to achieve the highest level of compliance and security.</p><p>Ultimately, a data closed loop will be formed in the cloud. Here, multi-source data transmitted from the edge will complete the steps of ingestion, labeling, and index establishment, and undergo automated cleaning based on clustering algorithms to eliminate conflicting data.</p><p>Next, the 4D annotation toolchain and data synthesis tool will annotate and enhance the retained valid data. Afterwards, through distributed model training, simulation verification and deployment, value assessment, and compliance audit, the data will officially complete closed-loop feedback and iteration.</p><p>With the continuous enhancement of computing power at the edge and in the cloud, the de vde, the computing power of intelligent assisted driving systems has achieved a leap from quantitative change to qualitative change, with its core support lying in the deep implementation of the three-tier architecture of &quot;cloud training-edge inference-vehicle execution&quot;. By transferring the computational load from resource-constrained vehicle-end to the cloud, it has unleashed an exponential growth in computing power potential.</p><p>The cloud, serving as the &quot;super brain&quot; of intelligent assisted driving, has become the core carrier for model training, data closed-loop, and algorithm iteration. In the field of end-to-end model training and corner case scenario mining, the scale of cloud computing power directly determines the evolution speed of intelligent assisted driving systems.</p><p>Globally, Tesla&#39;s cloud computing power has surpassed 88.5 EFLOPS; within the domestic camp, Geely&#39;s Starry Smart Computing Center leads the way with 23.5 EFLOPS, Huawei has broken through 10 EFLOPS, Ideal and Xiaomi both reach 8.1 EFLOPS, and Xiaopeng plans to increase its cloud computing power from 2.51 EFLOPS to 10 EFLOPS by 2025, building a more powerful training foundation.</p><p>As the &quot;edge terminal&quot; for real-time decision-making, the computing power of vehicle terminals is surging towards the thousand TOPS level at an exponential rate. Level 2 tasks such as autonomous parking and urban NOA require tens to hundreds of TOPS support, while Level 3 and above tasks need to break through the 1000 TOPS threshold to meet the real-time inference requirements of end-to-end models.</p><p>In the current chip matrix, NVIDIA&#39;s Thor-X-Super chip is poised to launch, with a computing power of 2000 TOPS that will redefine performance benchmarks; NIO&#39;s NX9031 has achieved a breakthrough in computing power of 1000 TOPS, followed closely by Xiaopeng&#39;s Turing chip with 750 TOPS. Huawei&#39;s Ascend 910 and Horizon&#39;s J6P respectively establish mid-range computing power barriers with 512 TOPS and 560 TOPS.</p><p>In the popular intelligent assisted driving solutions, NIO Pilot boasts a computing power of 1016 TOPS on the vehicle side; Tesla FSD follows closely with a computing power of 720 TOPS; Xiaopeng XNGP, Li Xiang AD PRO, Xiaomi Pilot Max, BYD Tianchenzhi Yan A, and Zeekr Qianli Haohan H7 are all on par, with a computing power of 508 TOPS; in comparison, Huawei ADS3.0&#39;s computing power on the vehicle side is slightly inferior, at around 200 TOPS.</p><p>Edge computing, as the &quot;neural node&quot; of vehicle-cloud collaboration, bears the responsibility of real-time collaboration among vehicle-side, roadside, and cloud data. Its standardization process directly affects the implementation efficiency of intelligent driving above Level 3. By enhancing the accuracy of local environmental perception and the speed of traffic optimization response, edge computing is becoming an indispensable technical foundation for high-level intelligent driving.</p><p>High-level intelligent assisted driving is moving towards &quot;technological equality&quot;</p><p>China&#39;s intelligent assisted driving is continuously making breakthroughs along the gradient of scene complexity: from the implementation of basic functions such as automatic lane changing and ramp passing on structured roads in highway NOA, to overcoming complex urban scenes such as traffic light recognition and unprotected left turns in city NOA, and ultimately achieving full-scene door-to-door (D2D) connectivity, building a &quot;garage-urban area-highway&quot; full-chain intelligent driving system, and integrating ultimate scenes such as automatic charging and cross-floor parking.</p><p>This evolutionary trajectory not only signifies a fundamental shift in technological paradigm from &quot;rule-driven&quot; to &quot;data-driven&quot;,
i i � � � � 5<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #fff; --tw-ring-color: rgba(59,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; � zep integration of real-world data and synthetic data has become a key path to breakthroughs in high-level autonomous driving. The maturity of dynamic data distillation technology and multi-modal feature alignment algorithms is gradually reducing the system&#39;s dependence on hardware redundancy, leading to a significant improvement in scene generalization ability.</p><p>This trend directly drives the structural decline in the cost of core perception components. Over the past year, the average price of LiDAR has dropped from 2,500 yuan to 1,200 yuan, a decrease of 52%; the prices of millimeter-wave radar, on-board cameras, and ultrasonic radar have decreased by 25%-31%, clearing the cost hurdle for the large-scale implementation of advanced intelligent driving.</p><p>Positioning technology is also evolving simultaneously, moving from the traditional model relying on LiDAR point cloud matching and high-precision maps to a new stage of mapless and end-to-end models.</p><p>By generating bird&#39;s-eye views through multi-camera fusion instead of pre-configured high-precision maps, the cost of map maintenance is significantly reduced. The cloud simulation environment constructed from massive data on the vehicle side continuously enhances the model&#39;s generalization ability for dynamic scenes. The increase in visual perception weight further reduces the dependence on high-cost hardware such as lidar.</p><p>In the long run, the image-free and end-to-end models are merely transitional forms. When vehicle-road-cloud collaboration and self-evolutionary learning become the core of technology, and breakthroughs are made in chip-level integration, global dynamic positioning based on Ambient Intelligence and generative AI will become the ultimate form.</p><p>The leap from &quot;modularity&quot; to &quot;end-to-end&quot;</p><p>Intelligent driving algorithms are undergoing a revolutionary evolution from modular stacking to end-to-end integration, with architectural innovation becoming the core engine of technological breakthroughs.</p><p>The end-to-end (E2E) architecture relies on a single neural network to achieve the entire process of &quot;perception-decision-control&quot; computation, directly outputting vehicle control commands. It eliminates the need for manual rule intervention and, through extensive data training, can approximate human intuitive decision-making. Its response efficiency and adaptability to complex conditions are significantly improved, but the system&#39;s generalization ability is limited by the threshold of the data scale provided by car companies.</p><p>To overcome this limitation, the Vision-Language Model (VLM) emerged as an enhancement module. It integrates visual perception and natural language understanding capabilities, analyzes complex traffic semantics, and generates decisions through chain reasoning. Together with the end-to-end architecture, it forms a &quot;fast and slow collaborative&quot; decision-making system, enhancing the safety of long-tail scenarios and corner cases.</p><p>However, VLMs consume a significant amount of computing power and cannot provide intuitive responses like end-to-end responses quickly. In addition, the high dependence on high-precision maps increases maintenance costs and compromises generalization ability.</p><p>To this end, the developers of intelligent driving algorithms have taken a further step on the multi-segment end-to-end architecture consisting of &quot;VLM+end-to-end&quot;, designing a more powerful technical combination - Vision-Language-Action (VLA) model.</p><p>VLA achieves advancement based on VLM, integrating action modalities to construct a unified &quot;vision-language-action&quot; model, completing an end-to-end closed loop of &quot;perception-decision-execution&quot;.</p><p>Compared to VLM and &quot;VLM+end-to-end&quot;, VL �A takes into account the action modality, forming a unified model that integrates vision, language, and action. It directly embeds multimodal information into the driving decision-making chain, enhancing generalization ability while significantly reducing dependence on data.</p><p>Of particular importance is that its reasoning process is fully derivable, allowing for an intuitive presentation of the reasoning process and working logic to users through a human-computer interaction interface, thereby enhancing users&#39; confidence in its use.</p><p>Despite its significant advantages, VLA still faces multiple challenges. The requirement for VLA to integrate visual, language, and action encoders within the same architecture significantly increases the difficulty of engineering development.</p><p>Meanwhile, the high data demand and acquisition cost of such architectures lead to excessively high initial training costs, and multimodal perception provides insufficient support for embodied abilities such as force feedback and physical interaction.</p><p>In addition, there is a contradiction between the real-time computing requirements and the current computing power of chips in the automotive domain, while the risk of black-box decision-making reduces the interpretability of decisions and increases the difficulty of debugging, among other issues.</p><p>Therefore, until the research and development threshold and implementation costs of VLA are effectively controlled, the multi-segment end-to-end architecture consisting of &quot;VLM+E2E&quot; remains a more inclusive mainstream choice.</p><p>The world model may become an engine for &quot;human-like intelligent driving&quot;</p><p>To a large extent, more advanced cloud-based algorithms will help reduce the training difficulty of the vehicle-end VLA architecture and rapidly enhance its generalization ability. Currently, cloud-based algorithms themselves are also undergoing significant technological innovation.</p><p>From data-driven imitation learning to generative world models capable of modeling the physical world, the underlying logic of cloud algorithm development is very clear: based on generative AI technology, a mileage closed-loop simulation system with a scale of billions of kilometers is constructed through the synthesis of virtual scenes. The latter not only encompasses long-tail scenarios that occur in the real world, but also utilizes spatiotemporal evolution prediction capabilities to simulate environmental changes triggered by specific actions of vehicles in unknown scenarios (such as extreme weather, traffic accidents, etc.).</p><p>This move will significantly reduce the reliance on real-world driving data during the training process of algorithm models. Furthermore, it can continuously generate multimodal training data while incorporating an enhanced self-learning mechanism, aiming to achieve ultimate optimization of driving strategies with a goal of human-like behavior.</p><p>On the other hand, the continuously evolving vehicle-cloud collaborative distillation mechanism will effectively enhance the generalization ability of the vehicle-side VLA algorithm architecture. Meanwhile, the real-time data flowing back from the vehicle side can nourish the cloud-side model, helping it integrate long-tail scenarios, drive iteration, and ultimately form a bidirectionally enhanced cognitive evolution loop.</p><p>What cannot be overlooked is that the cloud-based world model highly integrates intelligent driving elements such as sensor data, traffic rules, and practical experience. Therefore, it is essentially an AI framework capable of understanding, reasoning, and predicting the driving environment.</p><p>Leveraging its strengths in perception and decision optimization, the cloud-based world model can provide semantic information to enhance the environmental recognition capabilities of intelligent driving system
��xJ(
�
#
3q
��
�e35
cn9The profit of a few cents' worth of chip is catching up with NVIDIA#333333AB模板网本站2025-07-30 11:35:44/static/upload/image/20250730/1753847059486723.jpg<p>The growth potential for these small chip and edge chip manufacturers may not be as attractive as that of AMD, but they have a secure job.</p><p>Author | Yang Zhichao</p><p>In 2010, Apple first equipped its own A-series self-developed chips on the iPhone 4. The core processor SOC is the jewel in the crown of human manufacturing. However, Apple, which is capable of designing SOC, was troubled by a small chip: the signal baseband chip.</p><p>This is a very small chip. According to third-party teardown reports, its cost is only $15, whereas the cost of an A-series SOC can easily exceed $100.</p><p>Such a thing costing just a few dozen yuan has become Apple&#39;s Achilles heel.</p><p>Initially, the signal of the iPhone 4 suffered due to significant technical issues with its supplier. Subsequently, Apple switched suppliers to Qualcomm, which not only made a substantial profit from Apple each year but also threatened to cut off supply at one point. Over the years, Apple has invested heavily in attempting to develop its own baseband chips, but has always ended up with failure.</p><p>In the world of chips, there are numerous such &quot;gadgets&quot;. From Apple&#39;s mobile phones to Microsoft&#39;s computing centers, there are countless power management chips, signal chips, network chips, and so on. Their unit prices are extremely low, with the most expensive ones costing just tens of dollars and the cheapest ones only a few cents. Their process requirements are also far lower than CPUs, which often require 3nm technology. Many of them use outdated processes above 100nm.</p><p>But behind it, there is often a terrifying profit that surpasses Intel, MediaTek, Qualcomm, and even rivals NVIDIA.</p><p>The fundamental reason for this contrast is that, to some extent, these small chips play the role of WeChat in the hardware field.</p><p>01 The price of scrap products, Qualcomm&#39;s profits</p><p>Open any mobile phone or computer, and you will find dozens or even hundreds of chips. Besides the two cores of CPU and GPU, there are countless places that require chips: USB ports have USB chips, power supply has power management chips, WIFI requires WIFI chips, and there are also networ ��hK'
�
#
3q
�}
�m35
cn9The ranking of the top ten global packaging and testing factories has been released, with a total annual revenue increase of 3%#333333AB模板网本站2025-07-28 11:46:08/static/upload/image/20250730/1753847190112880.png<p>Revenue ranking of the world&#39;s top ten packaging and testing companies in 2024</p><p>The first-ranked company, Advanced Semiconductor Engineering, had a revenue of US$18.54 billion last year, accoun �
� � �PL'
�?
#
3q
�!
�e35
cn9Hailo announces the launch of the Hailo-10H chip, which supports edge-based generative AI#333333AB模板网本站2025-07-24 09:45:43/static/upload/image/20250730/1753847281944200.png<p>Hailo Technologies Ltd., a startup specializing in artificial intelligence chips, announced today the official launch of Hailo-10H, its second-generation AI accelerator, featuring powerful generative AI capabilities.</p><p>Hailo-10H introduces brand-new device-side capabilities, including large language models, visual-language models, and other generative architectures, without relying on cloud connectivity.</p><p>This new AI chip is based on the successful experience of the company&#39;s previous Hailo-8 processor, which focused on visual AI tasks. While maintaining a high energy-efficiency design optimized for edge environments, the new chip adds support for generative AI. The typical power consumption of Hailo-10H is only 2.5 watts, enabling it to support complex AI applications in various products such as personal devices, smart home systems, automobiles, and telecommunications infrastructure.</p><p>&quot;With the Hailo-10H now available for ordering, we have taken another significant step towards our mission of enabling everyone to access AI,&quot; said Orr Danon, CEO and co-founder of Hailo. &quot;This is the first standalone AI processor to bring true generative AI performance to the edge, combining high efficiency, cost-effectiveness, and a robust software ecosystem.&quot;</p><p>The Hailo-10H is fully compatible with Hailo&#39;s existing software stack and supported by a global developer community with over 10,000 monthly active users. This chip enables developers to run cutting-edge visual and generative AI models directly on edge devices, providing ultra-low-latency real-time response capabilities.</p><p>The chip also supports ultra-low-latency real-time deployment of AI models, capable of generating the first Token in less than a second and processing over 10 Tokens per second on a 2 billion-parameter model. In video application scenarios, it can achieve state-of-the-art performance in 4K object detection using models such as YOLOv11m.</p><p>According to Hailo, by enabling local processing, the chip enhances data privacy and reduces reliance on cloud-based AI services, while also contributing to lower operational costs. It meets the automotive-grade AEC-Q100 Grade 2 standard certification and is targeted for mass production in automotive systems such as cockpit displays and driver monitoring in 2026.</p><p>Hailo is a venture-backed startup that has raised $564 million in nine rounds of funding, including $136 million in October 2021. The company&#39;s investors include SKC Co. Ltd., 5AM Ventures LP, Versant Ventures LP, GPB Capital Holdings, Reedland Capital Partners, and Mitsubishi UFJ Capital Corporation.</p><p><br/></p>Hailo Technologies Ltd., a startup specializing in artificial intelligence chips, announced today the official launch of Hailo-10H, its second-generation AI accelerator, featuring powerful generative AI capabilities.Hailo-10H introduces brand-new device-side capabilities, including large language mo �10005adminadmin2025-07-30 11:48:48 2025-07-30 15:25:5340 s. It can also predict the behavioral intentions of surrounding traffic participants, thereby assisting in decision-making planning and vehicle control. To a certain extent, its emergence is expected to replace on-board visual language models, using cloud computing power to help on-board systems develop more accurate scene reasoning capabilities.</p><p>Currently, many industry players have begun to invest in cloud computing power. For example, Huawei ADS 4.0 publicly claims to have completed up to 600 million kilometers of L3-level intelligent assisted driving simulation verification through the cloud; Xiaopeng Automobile has disclosed their plan to achieve lightweight vehicle deployment by distilling and compressing cloud model parameters.</p><p>In addition, third-party intelligent driving solution providers like Momenta and Horizon Robotics have integrated the world model with Reinforcement Learning, optimized the design of the Reward Function, and significantly improved decision-making efficiency.</p><p>Although the world model has gradually become an important development trend for cloud-based algorithms in intelligent driving systems, it still demands high data quality, diversity, and real-time computing power on the vehicle side. Furthermore, the ability to create 3D models based on physical rules poses challenges for its large-scale commercial application.</p><p>The large-scale implementation of full-scenario D2D functionality in 2025, coupled with the upcoming mass production and deployment of L3 functionality from the end of the year to 2026, has prompted third-party suppliers of intelligent assisted driving, LiDAR manufacturers, steering system enterprises, intelligent assisted driving map providers, data annotation and simulation enterprises, etc., to increasingly become key ecosystem partners for OEMs in enhancing functional safety. Simultaneously, the complex application scenarios brought by full-scenario D2D functionality have also made intelligent assisted driving chip suppliers pay more attention to the business layout of perception algorithms.</p><p>With the continuous development of intelligent assisted driving functions, the market competition among participating enterprises is becoming more and more concentrated. In the future, the market concentration will further increase, and enterprises that can quickly respond to the needs of original equipment manufacturers (OEMs) and possess AI capabilities will stand out.</p><p>When the door to L3 officially opens, what we may see is not just a brief &quot;letting go&quot; of the steering wheel, but also an ultimate inquiry about the &quot;boundaries of intelligence&quot;.</p><p>The three-tier computing power architecture has redefined the thinking speed of machines, and the data loop has removed the cost barriers to technology implementation. VLA and the world model are bringing algorithms closer and closer to human cognitive logic. However, when intelligent assisted driving can handle 99.99% of scenarios, should the remaining 0.01% of unknowns be left to code or human intuition?</p><p>Perhaps, in the new decade of intelligent assisted driving, the real test question is never about the decimal points of technical parameters, but rather whether, in every iteration of code, we remember that what enables machines to learn driving is not just data and computing power, but also humanity&#39;s most fundamental expectations for driving and travel - not the efficiency of reaching the destination, but the calmness and peace of mind at every moment on the road.</p><p><br/></p>At the inception of any technology, debates surrounding its development path always arise, and intelligent assisted driving technology is no exception.Should we opt for a gradual progression from L2 to L4, or leapfrog directly to the L4 endpoint? Should we pursue the ultimate simplification of a pur �1000-adminadmin2025-07-30 14:16:48 2025-07-30 15:26:4440
� � �YQ'
�/
#
3
�
�s35
cn9The State Council Taiwan Affairs Office responds to the Taiwan authorities' inclusion of Huawei, SMIC, and others on the export control list#333333AB模板网本站2025-06-27 11:30:49<p>On June 25th, the Taiwan Affairs Office of the State Council held a regular press conference. A reporter asked, recently, Taiwan&#39;s economic authorities claimed that 601 entities, including mainland chip R&amp;D enterprises such as Huawei and Semiconductor Manufacturing International Corporation (Smic), had been included in the &quot;export control list&quot;. Some investors in Taiwan are worried that the relevant &quot;export ban&quot; will impact Taiwan&#39;s semiconductor enterprises. What is your comment on this?</p><p>Spokesperson Zhu Fenglian stated that the Democratic Progressive Party (DPP) authorities have no bottom line and are groveling and begging for mercy towards tariff bullying that harms Taiwan. However, when it comes to mutually beneficial cross-strait economic cooperation, they have repeatedly resorted to underhanded tactics and malicious tricks, even cooperating with the United States to suppress mainland enterprises as a &quot;tribute&quot; to show sincerity to the US and &quot;rely on foreign support to seek independence&quot;. Such despicable acts are despicable. We will take effective measures to resolutely maintain the normal order of cross-strait economic and trade exchanges and cooperation, and safeguard the interests and well-being of compatriots on both sides of the strait. We solemnly warn the DPP authorities that technological blockades cannot stop the pace of technological innovation in the mainland; attempts to &quot;decouple and sever links&quot; cannot delay the process of industrial upgrading in the mainland. Any actions by the DPP authorities that undermine cross-strait economic cooperation will only damage the competitiveness of Taiwanese enterprises themselves and limit the development space of Taiwan&#39;s economy. Facts have repeatedly proved that it is the right path for mankind to work together across the strait to strengthen the economy of the Chinese nation.</p><p><br/></p>On June 25th, the Taiwan Affairs Office of the State Council held a regular press conference. A reporter asked, recently, Taiwan&#39;s economic authorities claimed that 601 entities, including mainland chip R&amp;D enterprises such as Huawei and Semiconductor Manufacturing International Corporation �1000
adminadmin2025-07-30 14:37:45 2025-07-30 15:27:5840�SS(
�U
#
3
��u
�c35
cn9Computex Insight: With Higher IPC and Stronger Performance, Is Arm Going to Make a Big Splash Again?#333333AB模板网本站2025-06-12 17:15:18<p>At the 2025 Taipei International Computer Show, Arm unveiled its new generation Armv9 flagship CPU, codenamed Travis, for the first time, attracting a lot of attention and high market expectations. At this moment, I realized that almost a year had passed since the release of the previous gener ��zR'
�
#
3
�s
�m35
cn9Huang Renxun concludes his second trip to China within the year: at a critical moment, he reaffirms his commitment to the Chinese market#333333AB模板网本站2025-06-17 10:43:34<p>Renxun Huang, CEO of NVIDIA, the leading AI (Artificial Intelligence) chip company, has concluded his second trip to China within the year.</p><p>On April 19th, informed sources told The Paper that NVIDIA CEO Jen-Hsun Huang had concluded his trip � �ting for nearly 45% of the top ten. It is worth noting that its performance in 2024 was basically flat compared to 2023, mainly due to the weak recovery of mobile phones, consumer electronics, automotive, and industrial applications, which limited the rebound of related packaging orders. In the testing business, it faced challenges such as competition from rivals and some customers pushing for test automation.</p><p>The second-ranked company, Amkor, achieved revenue of US$6.32 billion last year, a decrease of 2.8% year-on-year. The reason for the performance change is that automotive electronics were constrained by the inventory reduction effect and sluggish vehicle sales in 2024, resulting in a lower-than-expected recovery in related packaging demand. Although orders for consumer components have shown signs of recovery, the intensified price competition in the Asian market has affected the momentum of revenue growth.</p><p>JCET Technology achieved a revenue of $5 billion in 2024, representing a 19.3% annual increase and ranking third. According to its annual report, the company has continuously increased its investment in advanced packaging, Chiplet, SiP, high-computing power computing, automotive electronics, industrial electronics, and other fields. The company invested 49.66 billion yuan to focus on the construction of an annual production capacity of 3.6 billion high-density packaging modules and 10 billion communication high-density packaging modules, which is expected to be completed by the end of 2025.</p><p>The fourth-ranked company, Tongfu Microelectronics, achieved a 5.6% annual revenue growth to USD 3.32 billion in 2024, with net profit attributable to shareholders of the parent company surging by 299.90% year-on-year. Its ongoing advanced packaging projects include expanding the bumping (bump packaging) production line and advanced packaging production line at its Penang factory. These projects include the industrialization of high-performance computing product packaging and testing (RMB 8.28 billion), MCU (microcontroller) product packaging and testing (RMB 7.8 billion), and power device packaging and testing (RMB 6.365 billion). It is expected that some of these projects will be completed by the end of 2025, further enhancing the advanced packaging capacity.</p><p>The fifth-ranked Powertech Technology had a revenue of US$2.28 billion last year, with an annual growth rate of only about 1%. The reasons include the lack of explosive growth in its main memory packaging and testing business, as well as the advanced packaging still being in a transitional period.</p><p>The sixth-ranked company, Tianshui Huatian, achieved a revenue of US$2.01 billion last year, representing a 26% annual increase. The performance growth was primarily attributed to its high-value customers. It is noteworthy that the company has already mass-produced low-end and mid-end packaging, and high-end technology has been developed. Furthermore, its advanced packaging technology has been deployed in areas such as AI, high-performance computing, automotive electronics, and memory.</p><p>The seventh-ranked Zhilu Packaging and Testing achieved a revenue of US$1.56 billion in 2024, representing a 5% annual increase. The performance growth was primarily driven by the recovery of semiconductor demand, technological upgrades, and strategic adjustments made by some enterprises. It is worth noting that Zhilu Packaging and Testing comprises five companies. In January 2021, Zhilu Capital acquired Licheng Technology&#39;s packaging and testing enterprise in Singapore, and in 2022, it acquired four packaging and testing factories of ASE Group located in mainland China, specifically in Suzhou, Kunshan, Weihai, and Shanghai.</p><p>The eighth-ranked company, Hanyawei, had a revenue of US$920 million last year, with an annual growth rate of 23.7%. Its main business is semiconductor packaging services. The performance growth in 2024 was mainly benefited from memory cus
'� �R��N
�
�
H � � B � � <
�
�
4 � � ,�|$�t �r$��.�z#�w � >�K '
3info登录成功!LinuxFirefoxadmin2026-08-18 18:20:26T�J 7!!3info清理缓存成功！3683779653Windows 10Chromeadmin2026-07-22 16:25:14Z�I C!!3info修改站点信息成功！3683779653Windows 10Chromeadmin2026-07-22 16:25:10L�H '!!3info登录成功!3683779653Windows 10Chromeadmin2026-07-22 16:25:00T�G 7!!3info清理缓存成功！1738043738Windows 10Chromeadmin2026-06-06 20:17:27T�F 7!!3info清理缓存成功！1738043738Windows 10Chromeadmin2026-06-06 20:17:26Z�E C!!3info修改站点信息成功！1738043738Windows 10Chromeadmin2026-06-06 20:17:26Z�D C!!3info修改参数配置成功！1738043738Windows 10Chromeadmin2026-06-06 20:17:25L�C '!!3info登录成功!1738043738Windows 10Chromeadmin2026-06-06 20:17:24G�B '!
3info登录失败!2936383695Windows 7Firefox2026-06-06 15:52:04K�A '!!3info登录成功!2028228850Windows 10Otheradmin2026-06-06 05:47:48F�@ '
3info登录失败!707875479Windows 7Firefox2026-06-02 22:31:20K�? ' !3info登录成功!245047321Windows 10Chromeadmin2025-07-31 09:12:50h�> _!!3info修改数据内容栏目2状态0成功！1728159372Windows 10Chromeadmin2025-07-30 18:37:28U�= ; !3info删除文章49成功！245066568Windows 10Chromeadmin2025-07-30 17:12:13U�< ; !3info修改文章86成功！245066568Windows 10Chromeadmin2025-07-30 15:29:51U�; ; !3info修改文章85成功！245066568Windows 10Chromeadmin2025-07-30 15:29:31U�: ; !3info修改文章84成功！245066568Windows 10Chromeadmin2025-07-30 15:29:10U�9 ; !3info修改文章83成功！245066568Windows 10Chromeadmin2025-07-30 15:28:46U�8 ; !3info修改文章82成功！245066568Windows 10Chromeadmin2025-07-30 15:28:21U�7 ; !3info修改文章81成功！245066568Windows 10Chromeadmin2025-07-30 15:27:58U�6 ; !3info修改文章80成功！245066568Windows 10Chromeadmin2025-07-30 15:27:36U�5 ; !3info修改文章79成功！245066568Windows 10Chromeadmin2025-07-30 15:27:08U�4 ; !3info修改文章78成功！245066568Windows 10Chromeadmin2025-07-30 15:26:44U�3 ; !3info修改文章77成功！245066568Windows 10Chromeadmin2025-07-30 15:26:18U�2 ; !3info修改文章76成功！245066568Windows 10Chromeadmin2025-07-30 15:25:54U�1 ; !3info修改文章75成功！245066568Windows 10Chromeadmin2025-07-30 15:25:30S�0 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 15:19:44U�/ ; !3info删除文章61成功！245066568Windows 10Chromeadmin2025-07-30 15:19:28S�. 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 15:18:22U�- ; !3info删除文章62成功！245066568Windows 10Chromeadmin2025-07-30 15:18:07S�, 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 15:15:41U�+ ; !3info删除文章63成功！245066568Windows 10Chromeadmin2025-07-30 15:15:25U�* ; !3info删除文章64成功！245066568Windows 10Chromeadmin2025-07-30 15:14:22S�) 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 15:13:34S�( 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 14:53:27U�' ; !3info删除文章66成功！245066568Windows 10Chromeadmin2025-07-30 14:48:32S�& 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 14:37:45U�% ; !3info删除文章67成功！245066568Windows 10Chromeadmin2025-07-30 14:30:47 tomers.</p><p>The ninth-ranked company, Jingyuan Electronics, reported a revenue of US$910 million in 2024, a decrease of 14.5% year-on-year. The performance change in 2024 was primarily influenced by the sale of Suzhou Jinglong Electronics.</p><p>The tenth-ranked company, Nanmao Technology, achieved a revenue of US$710 million, representing an annual growth of 3.1%. Its business covers testing, assembly, wafer bumping, and other fields such as flat panel display driver semiconductors (LCD driver ICs). It is worth noting that the performance growth in 2024 was primarily driven by the driver IC business.</p><p>According to the report, the OSAT market is undergoing value chain restructuring in 2024. Whether it is heterogeneous integration, wafer-level packaging (WLP), wafer stacking, the introduction of advanced testing equipment, or the urgent demand for high-frequency, high-density packaging from AI and edge computing, all pose higher requirements for OSAT providers. The packaging and testing industry has shifted from traditional manufacturing to a highly technology-integrated and R&amp;D-oriented strategic core.</p><p>The global packaging and testing market is expected to reach a value of $86.2 billion by 2025</p><p>According to the &quot;Research Report on Market Prospect Forecast and Development Trend of Global and Chinese Integrated Circuit Packaging and Testing Industry from 2025 to 2030&quot; released by the China Commerce Industry Research Institute, the global integrated circuit packaging and testing market will reach a scale of 82.1 billion US dollars in 2024, with a year-on-year growth of 5%. It is expected that the global integrated circuit packaging and testing market will reach a scale of 86.2 billion US dollars in 2025. Focusing on the Chinese market, the sales revenue of the integrated circuit packaging and testing industry in mainland China reached 314.6 billion yuan in 2024, with an increase of 7.14% compared to 2023. It is expected that the sales revenue of the integrated circuit packaging and testing industry in mainland China will reach 330.33 billion yuan in 2025.</p><p>With the continuous improvement of chip performance and the continuous shrinkage of system size, the requirements for packaging technology are also increasing. The rise of advanced packaging technologies such as Flip Chip (FC), Wafer-Level Packaging (WLP), 2.5D packaging, 3D packaging, and Chiplet can achieve chip miniaturization, thinning, high efficiency, and multi-integration, optimize chip performance, and reduce costs. They will become the mainstream in the packaging and testing market in the future.</p><p>Currently, domestic packaging and testing manufacturers are also actively deploying advanced packaging technologies. The growth potential for advanced packaging is vast, and its penetration rate is expected to increase. According to predictions, the penetration rate of advanced packaging in China will rise to 41% by 2025.</p><p>*Declaration: This article is created by the original author. The content of the article represents his personal views. Our reprint is solely for sharing and discussion purposes and does not represent our endorsement or agreement. If there are any objections, please contact the back-end.</p><p><br style="text-wrap-mode: wrap;"/></p><p><br/></p>Revenue ranking of the world&#39;s top ten packaging and testing companies in 2024The first-ranked company, Advanced Semiconductor Engineering, had a revenue of US$18.54 billion last year, accounting for nearly 45% of the top ten. It is worth noting that its performance in 2024 was basically flat co �10003adminadmin2025-07-30 11:46:32 2025-07-30 15:25:3040
� � ��UP(
�E
#
3
��
�m35
cn9From L2 to L3, VLA becomes the "key springboard" for intelligent assisted driving?#333333AB模板网本站2025-07-02 10:19:18<p>Recently, the Gemini family welcomed a new member - Gemini Robotics On-Device. This is Google DeepMind&#39;s first Vision-Language-Action (VLA) model that can be directly deployed on robots. The model does not rely on data networks during operation and can run entirely offline on the robot device locally. It has demonstrated strong visual, semantic, and behavioral generalization capabilities in various test scenarios, capable of understanding natural language instructions and completing tasks requiring high dexterity such as unzipping and folding clothes.</p><p>As a pivotal new paradigm in the field of machine intelligence, VLA not only endows robots with the ability to weave language intention, visual perception, and physical actions into a continuous decision-making flow, but also serves as one of the key technologies for enhancing the intelligence level of assisted driving.</p><p>Since last year, overseas companies such as Google, Microsoft, and Figure AI, as well as domestic embodied intelligence and automotive companies like Galaxy General, Zhiyuan Robot, Ideal Automobile, and Xiaopeng Automobile, have all released their own VLA models. This technology has become one of the key research areas in the fields of embodied intelligence and intelligent assisted driving.</p><p>VLM is moving towards VLA</p><p>The VLA model was first proposed by DeepMind and applied in the field of robotics, aiming to solve the intelligent agent control problem of the coordination among &quot;vision-language-action&quot;.</p><p>In July 2023, Google DeepMind introduced RT-2, the world&#39;s first VLA model for controlling robots. Unlike traditional robot models that can only support a limited set of specific instructions, RT-2 leverages the powerful language comprehension capabilities of large language models to directly interact with users through language. Upon receiving raw data from cameras and language commands, it can directly output control signals to complete various complex operations and tasks.</p><p>The success of VLA in the field of robotics was soon applied to the field of autonomous driving. At the end of October 2024, Waymo, a self-driving car company under Google, launched EMMA, an end-to-end multimodal model for autonomous driving.</p><p>EMMA is built upon Gemini, a multimodal large language model, which directly maps raw camera sensor data to various driving-specific outputs, including planner trajectories, perception targets, and road map elements. By representing all non-sensor inputs (such as navigation instructions and self-driving vehicle status) and outputs (such as trajectories and 3D positions) as natural language text, it maximizes the utilization of world knowledge from pre-trained large language models.</p><p>From a technological perspective, VLA represents an evolutionary step following VLM (Vision-Language Model) and is regarded as the end-to-end large model 2.0 - a multimodal machine learning model. VLA integrates three capabilities: vision, language, and action, unifying them within a single model. By inputting only into the machine, it can perform end-to-end map c 246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Model Number</span></p></li><li><p>LIS2DW12TR</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Part Number</span></p></li><li><p>C189624</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package</span></p></li><li><p>LGA-12(2x2)</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tape &amp; Reel</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>0.064g</p></li></ul><p><br/></p>
� � � �p
�_ 3<ul class=" list-paddingleft-2"><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Brand Name</span></p></li><li><p><a class="ml-[30px] text-[#000000] max-w-[350px] lc-text-hidden hover:text-lcBlue" href="https://list.szlcsc.com/brand/74.html?spm=sc.it.pd.pp___sc.gbn.xh1.zy.n&c=SO&qhclickid=2867deb651d3e7ba&lcsc_vid=RVVXAl1QFVMPVgUDTlheAlVQQwNdUVACE1BfUF0DE1QxVlNTRFZZVlFeQVJZVztW" target="_blank" data-spm="pp" data-spm-reset="sc.it.pd.pp" data-anchor-id="pp" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; --tw-rotate: 0; --tw-skew-x: 0; --tw-skew-y: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-pan-x: ; --tw-pan-y: ; --tw-pinch-zoom: ; --tw-scroll-snap-strictness: proximity; --tw-gradient-from-position: ; --tw-gradient-via-position: ; --tw-gradient-to-position: ; --tw-ordinal: ; --tw-slashed-zero: ; --tw-numeric-figure: ; --tw-numeric-spacing: ; --tw-numeric-fraction: ; --tw-ring-inset: ; --tw-ring-offset-width: 0px; --tw-ring-offset-color: #fff; --tw-ring-color: rgba(59,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max- � f; --tw-ring-color: rgba(59,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Model</span></p></li><li><p>STM32F072RBT6</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Number</span></p></li><li><p>C46046</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package Type</span></p></li><li><p>LQFP-64(10x10)</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tray</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>1.913g</p></li></ul><p><br/></p> --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Model Number</span></p></li><li><p>STM32F401RET6</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Part Number</span></p></li><li><p>C116978</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package</span></p></li><li><p>LQFP-64(10x10)</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"></span><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tray</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>1.913g</p></li></ul> ff; --tw-ring-color: rgba(59,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Model</span></p></li><li><p>STM8S003K3T6C</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Number</span></p></li><li><p>C27142</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package Type</span></p></li><li><p>LQFP-32(7x7)</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tray</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>1.269g</p></li></ul> drop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Model</span></p></li><li><p>LIS3DHTR</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Number</span></p></li><li><p>C15134</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package Type</span></p></li><li><p>LGA-16(3x3)</p></li><li><p></p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></li><li><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tape &amp; Reel</span><br/></span></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>0.235g</p></li></ul><p><br/></p> width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST<span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">(STMicroelectronics)</span></a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Model Number</span></p></li><li><p>STM32F103RDT6</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Part Number</span></p></li><li><p>C88435</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);"><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package</span></span></p></li><li><p>LQFP-64(10x10)</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tray</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>1.913g</p></li></ul><ul class=" list-paddingleft-2"></ul> �to China. However, Huang&#39;s trip to China this time did not include a meeting with DeepSeek.</p><p>Previously, foreign media reported that Huang Renxun met with NVIDIA&#39;s customers after his visit to Beijing on the 17th, including Liang Wenfeng, the founder of AI startup DeepSeek.</p><p>Ren Huang&#39;s whirlwind visit to China began on the 17th. According to CCTV News, on April 17, Ren Hongbin, President of the China Council for the Promotion of International Trade, held talks with Ren Huang, CEO of NVIDIA, in Beijing. During the talks, Ren Huang stated that China is a very important market for NVIDIA and he hopes to continue to cooperate with China.</p><p>During the meeting, regarding the US government&#39;s implementation of indefinite export controls on NVIDIA&#39;s H20 chips exported to China, Huang Renxun stated that the US government&#39;s tightening of chip export controls has had a significant impact on NVIDIA&#39;s business. As a company that has been deeply rooted in the Chinese market for three decades, NVIDIA has grown and achieved mutual success with the Chinese market. China is not only one of the largest consumer markets globally, but its thriving industrial ecosystem and leading software capabilities have also become important driving forces for their continuous innovation. NVIDIA will continue to spare no effort in optimizing its product system that meets regulatory requirements and unswervingly serve the Chinese market.</p><p>On the evening of April 17th, He Lifeng, member of the Political Bureau of the CPC Central Committee and Vice Premier of the State Council, met with Huang Renxun at the Great Hall of the People. According to Xinhua News Agency, He Lifeng stated that the Chinese market boasts immense investment and consumption potential, with accelerated industrial transformation and upgrading. It represents the optimal application scenario for the new round of technological revolution and industrial transformation, and has always been a fertile ground for foreign-funded enterprises to invest and trade. He welcomed more US-funded enterprises, including NVIDIA, to deeply explore the Chinese market, fully demonstrate their industrial advantages and capabilities in China, and thereby gain an advantage in global competition.</p><p>Huang Renxun expressed his optimism about China&#39;s economic prospects and his willingness to continue to deeply engage in the Chinese market, playing a positive role in promoting economic and trade cooperation between the United States and China.</p><p>Renxun Huang&#39;s second visit to China within the year comes at a crucial time. According to CCTV Financial Commentary, just before Renxun Huang&#39;s visit, the US government included NVIDIA&#39;s AI accelerator H20, which is &quot;specially supplied&quot; for the Chinese market, in its export control list, requiring a license for export to China. On April 16th local time, NVIDIA&#39;s stock price plummeted by 6.9%, after having already fallen sharply previously.</p><p>CCTV News pointed out that recently, the US government has &quot;stabbed in the back&quot; of NVIDIA by imposing indefinite restrictions on NVIDIA&#39;s exports of H20 chips to China. At this juncture, Huang Renxun visited Beijing and stated, &quot;NVIDIA will continue to optimize products that meet regulatory requirements and continue to serve the Chinese market.&quot; It can be seen that in the face of the tariff wall built by the US government, both American consumers and US-funded enterprises are resisting and striving to break through. This also once again demonstrates that using tariffs to &quot;build a wall&quot; cannot change market rules or stop the pace of cooperation.</p><p>After visiting Beijing, Huang Renxun arrived in Shanghai, NVIDIA&#39;s R&amp;D hub in China.</p><p>On April 18th, Gong Zheng, the mayor of Shanghai, met with Huang Renxun , the president and CEO of NVIDIA, and his delegation.</p><p>Gong Zheng stated that Shanghai, as China&#39;s economic center city, is focusing on the &quot;five centers&quot; as its main direction, accelerating the establishment of a socialist modern international metropolis with global influence. Integrated circuits and artificial intelligence are the leading industries that Shanghai prioritizes for development, and have achieved rapid development in recent years. Shanghai has become the city with the strongest comprehensive technological strength, the largest industrial scale, and the most complete industrial chain in China&#39;s integrated circuit industry. We welcome globally renowned high-tech enterprises such as NVIDIA to seize the opportunities presented by China&#39;s ultra-large market and Shanghai&#39;s expanded high-level opening-up, gather global resource elements, strengthen in-depth cooperation with local enterprises, and jointly create a bright future for technological development. Shanghai will, as always, create a market-oriented, law-based, and international first-class business environment to help various enterprises achieve better and greater development.</p><p>Huang Renxun stated that artificial intelligence holds immense potential for development, and countries should strengthen cooperation and sharing to promote mutual progress and global prosperity. The Chinese market is of great importance to NVIDIA, not only because of its vast size, but also due to its high-quality customer base and favorable industrial ecosystem. Over the past 30 years, NVIDIA&#39;s business has thrived precisely because of its deep commitment to the Chinese market. Shanghai serves as an important research and development base for NVIDIA, boasting a favorable business and industrial environment. NVIDIA adheres to the laws of all countries, upholds its commitment to the Chinese market, and strives to provide services to the Chinese market in the best possible way.</p><p>Foreign media analysis points out that during such a turbulent period for NVIDIA, Huang Renxun&#39;s visit to China can be regarded as a globally notable event. Huang Renxun&#39;s visit to China this time indicates that the company is unwilling to give up the Chinese market and may be designing other chips for the Chinese market that are not subject to US export controls.</p><p>On April 14th, not long ago, NVIDIA pledged to invest $500 billion in the United States to build AI infrastructure. However, a day later, according to NVIDIA&#39;s disclosure, a new round of US government restrictions on chip exports to China may cause the company a loss of $5.5 billion.</p><p>In the previous fiscal year, NVIDIA&#39;s global revenue reached $130.5 billion, marking an 114% year-on-year increase. According to data from Bernstein Research, NVIDIA&#39;s revenue in the Chinese market for the same period amounted to $17 billion.</p><p>On the evening of January 19th, three months ago, NVIDIA held a Spring Festival Gala in Beijing, and Jen-Hsun Huang attended and delivered a speech. When talking about NVIDIA&#39;s development in China, Jen-Hsun Huang said that NVIDIA has a history of 25 years in China, and through Geforce graphics cards, the company entered China earlier than many technology companies. Jen-Hsun Huang also pointed out that NVIDIA has nearly 4,000 employees in Beijing, Shanghai, and Shenzhen, and has grown by nearly 60% over the past few years.</p><p><br/></p>Renxun Huang, CEO of NVIDIA, the leading AI (Artificial Intelligence) chip company, has concluded his second trip to China within the year.On April 19th, informed sources told The Paper that NVIDIA CEO Jen-Hsun Huang had concluded his trip to China. However, Huang&#39;s trip to China this time did n �1000adminadmin2025-07-30 14:53:27 2025-07-30 15:28:2140 d that some intentionally designed or maliciously implanted technological backdoors may become the trigger for leaks.</p><p>Invisible &#39;electronic spies&#39; and&#39; time bombs&#39;</p><p>Technical backdoors typically refer to methods that bypass normal security checks and gain access to programs or systems. The original intention of designing a technical backdoor is to facilitate developers in debugging and modifying vulnerabilities. However, if it is not removed in a timely manner and exploited by malicious attackers, it can become a security risk, allowing unauthorized access to the system and sensitive information.</p><p>——Malicious self contained: the &quot;insider&quot; in the device. Some chips, smart devices, or software produced overseas may have intentionally embedded &quot;backdoors&quot; during the design and manufacturing stage. Manufacturers can remotely control the devices through specific signals, such as automatically turning on cameras, microphones, or commanding the background to automatically collect specified data and transmit it back.</p><p>——Post cracking: &quot;Eyes in the Dark&quot;. Some manufacturers set up a &quot;backdoor&quot; that allows remote access when leaving the factory for the convenience of later maintenance and repair. This was originally an after-sales service function, but if poorly managed or maliciously cracked by a third party, this&#39; backdoor &#39;will peek into and steal sensitive information data in a dark corner.</p><p>——Secretly implanted: &quot;poisoning&quot; in the supply chain. Individual criminals may use software update channels, pollute open source code repositories, or tamper with code in the supply chain to implant &quot;backdoors&quot; in the device usage process, which can also achieve the purpose of illegally manipulating devices and stealing secrets.</p><p>Reminder from national security agencies</p><p>The security of smart devices and information systems is closely related to each and every one of us, and even more so to national security. The broad masses of the people must keep their eyes open, remain vigilant, continuously enhance their national security awareness and literacy, and prevent and resist the theft of secrets through hidden &quot;technological backdoors&quot;. Key classified positions can avoid the risk of backdoors in overseas software and hardware by using independently controllable chips and domestically produced operating systems. Potential technical backdoor security risks can also be reduced by strengthening technical protection measures, such as developing patch strategies, regularly updating operating systems, regularly checking device logs, and monitoring abnormal traffic.</p><p>National security is everyone&#39;s responsibility. Citizens and organizations should cooperate with the national security organs to do a good job in the security prevention, investigation and disposal of cyber espionage. If any suspicious behavior is found, please report it to the local national security organs through the 12339 national security organ&#39;s report acceptance telephone, the online report platform (www.12339. gov.cn), the WeChat official account of the Ministry of National Security or directly to the local national security organs.</p><p><br/></p>In today&#39;s highly digitized era, the importance of network security is becoming increasingly prominent, not only related to personal privacy and corporate secrets, but even affecting national security. It should be noted that some intentionally designed or maliciously implanted technological bac �1000adminadmin2025-07-30 15:19:44 2025-07-30 15:29:5140 lation has once again made their attempts successful, and the market will usher in another performance explosion.</p><p>As Travis becomes the foundation of the next generation flagship chip, it can be foreseen that a batch of flagship devices in the future will experience significant performance and experience improvements.</p><p>Arm makes everything smarter</p><p>As the influence of the Arm ecosystem continues to expand, the changes brought by the integration of Travis&#39;s new generation mobile computing platform will also affect everyone. In Arm&#39;s keynote speech at Computex, Chris Bergey, Senior Vice President and General Manager of the Terminal Business Unit, mentioned a data point: 99% of smartphones worldwide use the Arm architecture. Considering the penetration rate of smartphones in modern society, Arm has become one of the core components of the entire intelligent ecosystem.</p><p>Therefore, in recent years, the demand for Arm in the PC and tablet markets has been rapidly growing, and it is expected to account for 40% of the overall shipment volume by 2025. From smartphones to laptops, from tablets to IoT devices, Arm is becoming a common &quot;computing base&quot; for various terminal devices, helping the entire smart ecosystem become more efficient and intelligent.</p><p>At the end of this year, with the launch of the Dimensity 9500 flagship phone integrated with Arm Travis CPU, the impact of Travis on the market will gradually become apparent. As a bridge between the intelligent ecosystem and users, smartphones with higher performance, stronger energy efficiency, and more comprehensive AI capabilities will reshape the user&#39;s intelligent experience.</p><p>Imagine that when you want to purchase commonly used daily necessities for your home, all you need to do is talk to the AI assistant on your phone and complete a series of operations such as &quot;search confirm order&quot;. At the same time, it can also help you manage your daily schedule and arrangements, just like a personal butler. In the past, in order to achieve a similar functional experience, mobile phones often needed to allocate more resources to AI computing, which would inevitably affect other functions and battery life. However, for Travis and Dimensity 9500, which were specifically designed for the AI era, this was not difficult.</p><p>With the further development of AI, Arm will also expand their influence. From MediaTek to Nvidia, from Amazon to Google, Arm and its partners are facing unprecedented opportunities. As emphasized in the keynote speech, the AI era is a once-in-a-lifetime opportunity that will redefine how technology affects the world. However, to truly unleash this potential, deep collaboration across the entire Arm ecosystem is needed.</p><p>In my opinion, &#39;everything powered by Arm&#39; is not just a slogan, but a blueprint for the future of intelligent interconnection of all things. By utilizing the efficient and powerful computing platform designed by Arm, all terminals can become a part of the AI ecosystem, making every device smarter and ultimately changing the work and life of each and every one of us.</p><p><br/></p>At the 2025 Taipei International Computer Show, Arm unveiled its new generation Armv9 flagship CPU, codenamed Travis, for the first time, attracting a lot of attention and high market expectations. At this moment, I realized that almost a year had passed since the release of the previous generation �1000adminadmin2025-07-30 15:13:34 2025-07-30 15:28:4640 �her than playing a word game to benchmark against the previous two generations of products.</p><p>You should know that the previous generation Cortex-X925 achieved an IPC improvement of about 15% on the Geekbench 6.2 benchmark based on the previous generation Cortex-X4, and Travis will go further on this basis. It can be said that the CPU architecture efficiency of smartphones is experiencing a rare continuous breakthrough in the past. More importantly, Travis not only improves IPC, but also integrates the next generation SME (Scalable Matrix Expansion), which is an instruction set extension for AI acceleration that can significantly enhance CPU throughput in matrix operations, machine learning inference, and other areas.</p><p>You can understand it this way: SME is like configuring an AI assistant for the CPU, which can efficiently complete AI related tasks without the need for a dedicated NPU accelerator. This is crucial in the future mobile market. As the AI application ecosystem expands, more and more apps are incorporating AI functions, such as real-time image recognition for cameras, local recognition for voice assistants, augmented reality, etc. The addition of SMEs will enable Travis to make a qualitative leap in tasks such as AI and computer vision recognition.</p><p>From the currently released information, the improvement of Travis is undoubtedly comprehensive. In terms of traditional performance, Arm has made all daily use smoother and more energy-efficient through higher IPC. In terms of AI applications, it has also expanded the versatility of CPU through SME, giving phones a more efficient &quot;brain&quot;.</p><p>Travis (code name) has made many optimizations based on user needs, while also providing platform vendors with more efficient solutions. As the new flagship CPU of Arm, it will form a new high-performance computing platform together with Arm&#39;s new generation GPU codenamed Drage, providing more assistance to chip manufacturers.</p><p>Previously, MediaTek&#39;s Dimensity 9400, with its excellent architecture design of Cortex-X925, coupled with a doubling cache design and Dimensity scheduling engine, has significantly improved its performance, energy efficiency, and AI applications with up to 29.1 billion transistors. From the test results, it can be seen that Cortex-X925 can maintain a gaming experience of nearly 60fps for &quot;Mingchao&quot; throughout the entire process without exceeding a cluster clock frequency of 3GHz.</p><p>In daily use, the Dimensity 9400 performs well in large-scale mobile gaming and other application scenarios, which has also earned high praise for flagship phones equipped with this chip in the market. It can be seen that the number of flagship devices using the Dimensity 9400 in the market has also increased significantly, which undoubtedly gives more confidence to the market and terminal manufacturers.</p><p>It is inferred that MediaTek&#39;s next-generation flagship chip, the Dimensity 9500, will adopt new core solutions such as Travis. If nothing unexpected happens, the Dimensity 9500 will still choose the &quot;full core&quot; design scheme, and it will also be MediaTek&#39;s third-generation full core architecture. Thanks to the significant improvements in IPC and other aspects of Arm CPUs in recent years, even with high-performance large cores, power consumption and heat generation can still be effectively controlled, allowing phones to have more extreme performance while maintaining good energy efficiency.</p><p>From Cortex-X925 to Travis, it is not difficult to see that Arm is not simply stacking &quot;higher GHz&quot;, but rather achieving performance breakthroughs by continuously improving IPC, empowering AI, and bringing innovative synergy through overall solutions, while finding the best balance between performance and energy efficiency. We all know it&#39;s difficult, but Arm&#39;s deep technical accumu �ation Arm Cortex-X925 without realizing it.</p><p>Recalling the Cortex-X925 released in 2024, with outstanding IPC improvements and MediaTek&#39;s advanced second-generation full core architecture design, flagship processors such as the Dimensity 9400 based on it have significantly improved performance and experience, while also breaking multiple Arm chip performance records. So many people are curious whether Travis, who took the baton from Cortex-X925, can once again establish a new performance dynasty?</p><p>To answer this question, we first need to clarify two questions: 1. What is the key to improving CPU performance; 2. How to measure the performance of a CPU? After clarifying these issues, we may be able to gain a glimpse into the details of future flagship chips, and also better understand why Arm executive Chris&#39;s keynote speech at the Taipei Electronics Show caused a sensation.</p><p>What is IPC and why is it important?</p><p>Have you ever thought about why processors with different architectures have significant performance differences at the same clock frequency? The answer to this question is a performance parameter that is not well-known to everyone - IPC, which stands for Instructions Per Clock Cycle. It refers to the number of instructions that a CPU can execute in each clock cycle and is one of the key indicators for evaluating the actual performance of a CPU. It is also a key parameter mentioned in Arm&#39;s keynote speech.</p><p>Simply put, the higher the IPC, the stronger the CPU&#39;s ability to work at the same clock frequency. You can think of CPU frequency and IPC as the speed and torque of a car: the highest frequency is like the top speed of a car, representing extreme performance, but rarely used in daily driving, while IPC is similar to the torque of an engine, determining the acceleration and response ability of the vehicle in daily use.</p><p>In other words, even without reaching the maximum speed, the powerful torque can still make the car start quickly and accelerate rapidly. Similarly, the high IPC architecture allows the CPU to quickly complete tasks even at lower frequencies, bringing a smooth user experience. This makes IPC performance crucial in mobile device chips, as mobile devices often operate at their highest frequency for very long periods of time. Most of the time, in order to control heat generation and extend battery life, they will work at mid to low frequencies.</p><p>A CPU with high IPC can enable mobile phones to run high load apps smoothly without having to desperately increase the clock speed, thereby avoiding the huge power consumption and heat caused by simply increasing the frequency for performance. On the contrary, if a processor has a high clock frequency but insufficient IPC, it can only barely stack performance through &quot;burning electricity&quot;, resulting in skyrocketing power consumption and severe heating. In situations where it is difficult to maintain high clock frequency on small devices such as mobile phones, the phone will experience lagging due to overheating and frequency reduction.</p><p>It can be said that on small devices like mobile phones, IPC is the key factor that affects the actual experience. High IPC can bring higher performance per watt, longer battery life, lower operating temperature, and faster response speed, allowing the phone to visibly improve in usage scenarios such as app launch, multitasking switching, and game frame rates.</p><p>Travis appears, Arm is going to flip the table again</p><p>The importance of IPC should be clear to everyone, so it is not difficult to explain why Travis has attracted so much attention and expectation. Because it has been revealed in Arm&#39;s official information that Travis will once again experience a &quot;double-digit&quot; IPC improvement, and it is a breakthrough based on the already double-digit IPC improvement of Cortex-X925, rat � � � � � � d � � � � � � � � � � � r � { | ~ �
- s �9��D
�
�
O � � * � w
�
o
� g �_�W��C��7��1��+�}'�y! � s S�$ 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 14:27:32U�# ; !3info删除文章68成功！245066568Windows 10Chromeadmin2025-07-30 14:19:16U�" ; !3info删除文章69成功！245066568Windows 10Chromeadmin2025-07-30 14:18:22S�! 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 14:18:15U� ; !3info删除文章70成功！245066568Windows 10Chromeadmin2025-07-30 14:16:58S� 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 14:16:48U� ; !3info删除文章71成功！245066568Windows 10Chromeadmin2025-07-30 14:09:43S� 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 14:08:37U� ; !3info删除文章72成功！245066568Windows 10Chromeadmin2025-07-30 14:05:23S� 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 11:48:48U� ; !3info删除文章65成功！245066568Windows 10Chromeadmin2025-07-30 11:46:52S� 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 11:46:32U� ; !3info删除文章73成功！245066568Windows 10Chromeadmin2025-07-30 11:45:39U� ; !3info修改文章74成功！245066568Windows 10Chromeadmin2025-07-30 11:44:21S� 7 !3info新增文章成功！245066568Windows 10Chromeadmin2025-07-30 11:41:29U� ; !3info修改文章65成功！245066568Windows 10Chromeadmin2025-07-30 11:30:50[� G !3info修改单页内容47成功！245066568Windows 10Chromeadmin2025-07-30 11:23:13[� G !3info修改单页内容43成功！245066568Windows 10Chromeadmin2025-07-30 11:19:05[� G !3info修改单页内容48成功！245066568Windows 10Chromeadmin2025-07-30 11:18:21U� ; !3info修改文章58成功！245066568Windows 10Chromeadmin2025-07-30 11:11:57U� ; !3info修改文章59成功！245066568Windows 10Chromeadmin2025-07-30 11:10:14U� ; !3info修改文章57成功！245066568Windows 10Chromeadmin2025-07-30 11:08:37U� ; !3info修改文章56成功！245066568Windows 10Chromeadmin2025-07-30 10:31:00U�
; !3info修改文章55成功！245066568Windows 10Chromeadmin2025-07-30 10:29:05U� ; !3info修改文章53成功！245066568Windows 10Chromeadmin2025-07-30 10:27:25U� ; !3info修改文章54成功！245066568Windows 10Chromeadmin2025-07-30 10:26:57U�
; !3info修改文章53成功！245066568Windows 10Chromeadmin2025-07-30 10:25:20U� ; !3info修改文章52成功！245066568Windows 10Chromeadmin2025-07-30 10:23:14U� ; !3info修改文章51成功！245066568Windows 10Chromeadmin2025-07-30 10:16:01U� ; !3info修改文章50成功！245066568Windows 10Chromeadmin2025-07-30 09:49:54U� ; !3info修改文章49成功！245066568Windows 10Chromeadmin2025-07-30 09:45:31U� ; !3info修改文章60成功！245066568Windows 10Chromeadmin2025-07-30 09:39:52a� Q!!3info修改数据内容栏目4成功！1728159372Windows 10Chromeadmin2025-07-30 09:25:33L� '!!3info登录成功!1728159372Windows 10Chromeadmin2025-07-30 09:25:04[� G !3info修改单页内容47成功！245066568Windows 10Chromeadmin2025-07-30 09:21:03`� Q !3info修改数据内容栏目2成功！245066568Windows 10Chromeadmin2025-07-30 09:17:18a� S !3info修改数据内容栏目15成功！245066568Windows 10Chromeadmin2025-07-30 09:16:16[� G !3info修改单页内容46成功！245066568Windows 10Chromeadmin2025-07-30 09:15:55K�~ ' !3info登录成功!245066568Windows 10Chromeadmin2025-07-30 09:13:30F�} ' !
3info登录失败!245066568Windows 10Chrome2025-07-30 09:13:00F�| ' !
3info登录失败!245066568Windows 10Chrome2025-07-30 09:12:06F�{ ' !
3info登录失败!245066568Windows 10Chrome2025-07-30 09:11:52`�z Q !3info修改数据内容栏目2成功！245066568Windows 10Chromeadmin2025-07-29 17:57:26a�y S !3info修改数据内容栏目13成功！245066568Windows 10Chromeadmin2025-07-29 17:55:19`�x Q !3info修改数据内容栏目9成功！245066568Windows 10Chromeadmin2025-07-29 17:54:52 f; text-wrap-mode: wrap; background-color: rgb(255, 255, 255); margin-top: 0px !important;"><span style="font-weight: 600;">Shenzhen Pengcheng Wanli Semiconductor Co., Ltd</span>&nbsp;is an innovative enterprise specializing in the design and R&amp;D of high-performance semiconductor chips. Headquartered in Shenzhen, China, the company has established R&amp;D centers in Shanghai, Beijing, and Silicon Valley, USA. With the mission of &quot;Driving the Internet of Everything, Empowering Artificial Intelligence,&quot; the company is committed to providing global customers with cutting-edge chip solutions characterized by low power consumption and high computing power, covering fields such as consumer electronics, automotive electronics, the Internet of Things (IoT), 5G communications, and AI acceleration.</p><p class="ds-markdown-paragraph" style="margin: calc(var(--ds-md-zoom)*12px)0; font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Additionally, the company offers comprehensive solutions, adhering to a customer-first philosophy and providing customized products and personalized services based on its strengths in independent R&amp;D, unified production, and flexible sales. The management philosophy of &quot;putting people first and relying on talent&quot; has enabled the company to assemble a team of professional technical experts, implement standardized manufacturing management practices, and build a skilled marketing and service team.</p><p class="ds-markdown-paragraph" style="font-size: 16.002px; line-height: var(--ds-md-line-height); color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; text-wrap-mode: wrap; background-color: rgb(255, 255, 255); margin-bottom: 0px !important;">Guided by the business principles of &quot;integrity, innovation, collaboration, and win-win,&quot; the company continuously enhances its production capabilities and scales operations, laying a solid foundation for a new phase of development. &quot;Faster, More Precise, and More Stable&quot; is not only our promise to customers but also the relentless pursuit of&nbsp;<span style="font-weight: 600;">Shenzhen Pengcheng Wanli Semiconductor Co., Ltd</span>.</p><p><br/></p>Shenzhen Pengcheng Wanli Semiconductor Co., Ltd is an innovative enterprise specializing in the design and R&D of high-performance semiconductor chips. Headquartered in Shenzhen, China, the company has established R&D centers in Shanghai, Beijing, and Silicon Valley, USA. With the mission of "Driving the Internet of Everything, Empowering Artificial Intelligence," the company is committed to providing global customers with cutting-edge chip solutions characterized by low power consumption and high computing power, covering fields such as consumer electronics, automotive electronics, the Internet of Things (IoT), 5G communications, and AI acceleration. �1000iadminadmin2022-11-07 16:55:38 2025-07-30 11:23:1340 ba(59,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Model Number</span></p></li><li><p>STM32F103C8T6</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Part Number</span></p></li><li><p>C8734</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package</span></p></li><li><p>LQFP-48(7x7)</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tray</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Gross Weight</span></p></li><li><p>1.420667g</p></li></ul><p><br/></p> ,130,246,.5); --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-shadow: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-shadow-colored: 0 0 #0000; --tw-blur: ; --tw-brightness: ; --tw-contrast: ; --tw-grayscale: ; --tw-hue-rotate: ; --tw-invert: ; --tw-saturate: ; --tw-sepia: ; --tw-drop-shadow: ; --tw-backdrop-blur: ; --tw-backdrop-brightness: ; --tw-backdrop-contrast: ; --tw-backdrop-grayscale: ; --tw-backdrop-hue-rotate: ; --tw-backdrop-invert: ; --tw-backdrop-opacity: ; --tw-backdrop-saturate: ; --tw-backdrop-sepia: ; --tw-contain-size: ; --tw-contain-layout: ; --tw-contain-paint: ; --tw-contain-style: ; box-sizing: border-box; border: 0px solid rgb(229, 231, 235); color: rgb(0 0 0/var(--tw-text-opacity,1)); text-decoration: inherit; background-color: transparent; outline: none; cursor: pointer; transition: color 0.3s; touch-action: manipulation; margin-left: 30px; max-width: 350px; --tw-text-opacity: 1; text-overflow: ellipsis; text-wrap-mode: nowrap; overflow: hidden;">ST</a></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Model</span></p></li><li><p>TDA7850</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Product Number</span></p></li><li><p>C91679</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Package Type</span></p></li><li><p>FLEXIWATT-25</p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p><span style="color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Tube</span></p></li><li><p><span style="font-weight: 600; color: rgb(64, 64, 64); font-family: quote-cjk-patch, Inter, system-ui, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Noto Sans&quot;, Ubuntu, Cantarell, &quot;Helvetica Neue&quot;, Oxygen, &quot;Open Sans&quot;, sans-serif; font-size: 16.002px; text-wrap-mode: wrap; background-color: rgb(255, 255, 255);">Packaging Method</span></p></li><li><p>10.92g</p></li></ul><p><br/></p>
� � �-O'
�U
#
3
�#
�k35
cn9Siemens claims that the United States has lifted export restrictions on Chinese chip design software#333333AB模板网本站2025-07-07 08:30:05<p>On July 2nd local time, a reporter from CMG learned that Siemens AG received a notice from the US government stating that the US has lifted export restrictions on Chinese chip design software. According to the company&#39;s statement, the German supplier has restored full access to its software and technology for Chinese customers.</p><p>(Original title: &quot;Siemens Says U.S. Has Lifted Export Restrictions on Chinese Chip Design Software&quot;)</p><p><br/></p>On July 2nd local time, a reporter from CMG learned that Siemens AG received a notice from the US government stating that the US has lifted export restrictions on Chinese chip design software. According to the company&#39;s statement, the German supplier has restored full access to its software and �1000adminadmin2025-07-30 14:18:15 2025-07-30 15:27:0840 is is the fastest inference speed in the world, &quot;Naor Penso, Chief Information Security Officer of Cerebras, told me at the Vancouver Web Summit today. Last week, NVIDIA announced that it had achieved a speed of 1000 tokens per second on Llama 4, which was impressive. And the benchmark test we just released today reached 2500 tokens per second. ”</p><p>If you are unfamiliar with these concepts, you can understand &quot;reasoning&quot; as thinking or acting: constructing sentences, images, or videos based on your input or prompts. Consider &#39;token&#39; as the fundamental unit of thought: a word, a character, or a symbol.</p><p>The more tokens an AI engine can process per second, the faster it will return results. And speed is crucial. Perhaps it doesn&#39;t mean much to you personally, but when corporate customers want to add an AI engine to their shopping cart to instantly tell you that they only need to add one more ingredient to make the perfect Korean style grilled beef Taco, they must be able to quickly serve thousands of people.</p><p>Interestingly, speed is about to become even more crucial.</p><p>We are entering an era of agency, where AI will be able to perform complex multi-step tasks for us, such as planning and booking weekend trips to Austin to watch Formula One races. Agents are not magic: they approach big tasks in the same way as you... step by step. That is to say, it is necessary to break down a large task into 40, 50, or even 100 sub tasks, which means the workload will increase significantly.</p><p>AI agents need to handle more work, and various tasks need to communicate with each other, &quot;Penso told me. You can&#39;t tolerate slow reasoning speed</p><p>The 4 billion transistors on WSE are a key component in achieving this high-speed performance. As a comparison, the Intel Core i9 has 3.35 billion transistors, while the Apple M2 Max chip only provides 6.7 billion transistors. But the key to building a high-speed computing monster lies not only in the number of transistors, but also in the coordinated layout of the system: integrating all components on the same chip and equipping it with 44GB of the fastest RAM (memory).</p><p>AI computing requires a large amount of memory, &quot;Penso said. NVIDIA requires external storage, while with Cerebras you don&#39;t need external storage. ”</p><p>The independent institution Artificial Analysis verified this speed data, stating that when testing the chip on Llama 4, it reached 2522 tokens per second, while NVIDIA Blackwell&#39;s speed was only 1038 tokens per second.</p><p>We tested dozens of suppliers, and Cerebros is the only solution that outperforms Blackwell in Meta flagship model inference, &quot;said Micah Hill Smith, CEO of Artificial Analysis.</p><p>The WSE chip showcases an interesting evolution in computer chip design.</p><p>Since the 1950s when integrated circuits were manufactured and the 1960s when microprocessors were manufactured, CPUs have long dominated the field of computing. Until recently, GPUs (Graphics Processing Units) have transformed from auxiliary tools in the fields of graphics and gaming to critical processing components in AI development. Julie Shin, Chief Marketing Officer of Cerebras, told me that WSE is not based on x86 or ARM architecture, but on a completely new architecture designed to accelerate GPU computing.</p><p>This is not a progressive technology, &quot;she added. This is another leap forward in the field of chips</p><p><br/></p>Today, I personally held the world&#39;s largest computer chip. Although its volume is impressive, its speed is even more astonishing and important. Most computer chips are very small, only the size of stamps or even smaller. In contrast, the Cerebras WSE (Wafer Scale Engine) is a huge square with e �1000adminadmin2025-07-30 15:15:41 2025-07-30 15:29:1040 �k interface controller (NIC) chips behind the network cable ...</p><p>Compared to the well-known Nvidia or Qualcomm, these &quot;small chips&quot; are relatively unknown and very cheap. A USB chip costs as low as 1 yuan, and a Broadcom WIFI chip may only cost 10 yuan. Dozens of small chips combined may not be as valuable as a CPU.</p><p>However, they are indispensable, as they are active throughout the entire digital industry, from smartphones to Microsoft&#39;s computing clusters, all of which are built with a small number of core chips and a larger number of smaller chips.</p><p>Among them, the core chip market is dominated by MediaTek, Apple, NVIDIA, AMD, Intel, and Qualcomm, which are the world&#39;s most renowned chip manufacturers and the largest enterprises in terms of market capitalization in the entire industry.</p><p>It seems that the remaining chips are all for the taking.</p><p>But the reality is just the opposite. The profits they earn are no less than those from CPUs and GPUs.</p><p>Infineon is a typical &quot;small chip manufacturer&quot;. As a chip company spun off from Siemens, Infineon does not possess the SOC design capabilities of Samsung, MediaTek, or Qualcomm. Its expertise lies in microcontroller processors, such as power management chips in mobile phones, motor drive chips in automobiles, and sensors in headphones.</p><p>Putting aside these technical terms, a more intuitive explanation is: Infineon produces small chips worth a few cents or a few dollars that are used in various electrical devices.</p><p>Searching for Infineon on 1688 is like entering a bargain basement, where most chips are priced at less than 5 yuan, and some are only a few cents. In the second-hand trading realm, such chips generally don&#39;t deserve to enter the Xianyu market, and most of them are traded at waste recycling stations.</p><p>But the profitability of this business is unexpected.</p><p>In 2024, Infineon&#39;s net profit margin reached a high of 20.8%. This was still a minor dip in Infineon&#39;s performance, as its profit margin reached 26% in the third quarter of 2023. This was higher than AMD&#39;s profit margin during the same period, and significantly higher than MediaTek, which has the largest shipment volume in the mobile phone chip industry.</p><p>Another frequent visitor to the &quot;junkyard&quot; is Broadcom.</p><p>Broadcom&#39;s business is diverse and complex, but its true traditional core business is only one: communications. From small chips in household routers, mobile phone WIFI chips, and GPS chips, to large network chips in data centers and PCIe switching chips, Broadcom dominates them all. Due to Broadcom&#39;s strength in the field of communication chips, there has been a popular saying on the Internet: 99% of the world&#39;s network signals pass through at least one Broadcom chip.</p><p>Of course, in the eyes of the bosses of the scrap metal recycling station, Broadcom&#39;s strength is meaningless. They only care about how much copper they can extract from the chips.</p><p>Like Infineon, Broadcom chips are also extremely affordable. Searching for Broadcom&#39;s English name on 1688, what comes into view are still cheap chips ranging from a few yuan to tens of yuan. Of course, the overall unit price of Broadcom is still higher than that of Infineon. In data centers and enterprise-level switches, Broadcom also has some products that cost thousands of dollars, but compared to enterprise-level CPUs and GPUs that often cost tens of thousands of dollars, their prices are still considered &quot;marginal&quot;.</p><p>Compared to manufacturers like Qualcomm that produce SOC, Broadcom&#39;s business seems very &quot;dumb and crude&quot;. But in fact, Broadcom once came close to acquiring Qualcomm. Its profitability is not inferior to th �at of glamorous companies like Intel and Qualcomm.</p><p>Broadcom&#39;s net profit in 2024 was affected by its AI business and a large acquisition, making the data relatively complex. Let&#39;s directly look at the data for 2023: In fiscal year 2023, Broadcom&#39;s net profit margin reached 39.31%. In terms of gross profit margin, Broadcom has consistently maintained a rate above 60% throughout the years, with some years exceeding 70%.</p><p>Long before the advent of the AI boom, Broadcom was a renowned advanced asset and profit machine on the Nasdaq, with profit margins second only to NVIDIA but significantly stronger than other so-called &quot;AI chip manufacturers&quot;.</p><p>For comparison, the net profit margin in Qualcomm&#39;s latest financial report is only 25.77%. Among today&#39;s leading chip manufacturers, perhaps only NVIDIA&#39;s profit margin can surpass Broadcom&#39;s.</p><p>Qualcomm, which produces high-end chips for mobile phones, and Intel, which produces enterprise-level CPUs, surprisingly cannot earn as much as Broadcom, which produces cheaper chips with larger manufacturing processes. Could it be that there is some special core technology in those cheap chips?</p><p>02 WeChat in the chip industry</p><p>In startups, the most profitable ones are often not the technical partners, but the sales partners. The sales partners are also often the real top bosses.</p><p>The secret to making money from small chips does not lie in superb technology, but rather in the &quot;networking&quot; in the silicon-based world and the network effect that emerges from it.</p><p>Taking the baseband chip as an example, it is not just a technical game of computing power. It requires connection and adaptation with different operators. After all, the network frequency bands, parameters, and rules of operators vary from place to place. In layman&#39;s terms: signals, like dialects, have their own &quot;accent&quot;. No matter how standard Apple&#39;s London accent is, it cannot be understood by people from northern Shaanxi.</p><p>This requirement for compatibility with telecom operators has transformed chip manufacturing, a standardized industrial process, into a &quot;social&quot; game at the silicon layer.</p><p>Therefore, a truly globalized baseband chipset must be compatible with hundreds of operators worldwide, and the testing cycle can easily take up to a year. Many issues have no shortcuts and can only be resolved through experience. If the compatibility is not good, the phone will experience poor signal and severe heating issues. This is also one of the reasons why Apple has failed in the field of baseband chipsets: Apple possesses the technical capability to develop top-notch SOC in-house, but as a newcomer, it is difficult for it to establish good relationships with hundreds of operators worldwide in a short period of time.</p><p>This extremely high threshold makes it difficult for Qualcomm, as an incumbent, to be shaken, and even Apple has to continue to pay an expensive premium. Its 5G baseband chip market share is around 50% all year round, while in the mobile phone SOC market, Qualcomm&#39;s share is only 23%, and the top player MediaTek has a share of only 36%. Clearly, this moat formed by &quot;social networks&quot; is stronger than the latest technology.</p><p>However, in the &quot;social networking&quot; field, the most powerful chip manufacturer is not Qualcomm, but Broadcom. After all, 99% of the signals in silicon-based life are transmitted through its chips.</p><p>This has naturally become Broadcom&#39;s competitive advantage.</p><p>In the enterprise hardware market, Broadcom plays a role in connecting various devices. For example, its PCIe switching chips are responsible for high-speed interconnection between CPUs and GPUs. NVIDIA&#39;s super �computing clusters utilize a large number of Broadcom chips for interconnection.</p><p>In other words, various hardware components need to communicate with each other, and Broadcom serves as their &quot;WeChat&quot;.</p><p>This type of connectivity is extremely complex in data centers, where the sources of various devices are extremely diverse, some may be from brand A, and some may be from brand B. In order to connect different devices, a Broadcom Ethernet switching chip may need to support more than a hundred protocols to ensure seamless docking between motherboards, storage, servers, and other devices.</p><p>If communication is compared to social interaction, Broadcom is a perfect mediator who can dance gracefully between all parties. It understands the needs of each party, knows how to speak to people in their own language and to ghosts in their own language, and understands how to balance the interests of all parties.</p><p>This kind of competitive advantage is more difficult to shake than the competitive advantage of technology. Even if other manufacturers can surpass Broadcom in manufacturing technology, how can they coordinate the communication tasks of so many parties?</p><p>Meanwhile, Broadcom offers a wide range of products, with a single data center potentially utilizing up to 5-10 different Broadcom chips. One chip is considered a component, while 10 chips constitute a platform. Once customers adopt a platform, it becomes challenging for them to switch to another, as it involves a shift in the entire ecosystem, akin to asking a group of Northeasterners to switch their communication language to Cantonese, which is a difficult task.</p><p>In fact, as a peripheral chip manufacturer, Broadcom should have been a &quot;subordinate&quot; of CPUs and GPUs, adapting to the rules set by Intel and NVIDIA, the core of servers. However, in reality, it has become a rule-setter to some extent, illustrating the phrase &quot;power flows from the bottom up&quot; from one perspective.</p><p>Beyond communication chips, even the field of electronic control chips exhibits a certain network effect: electronic control requires power supply, so it is often tied to the main control, and easy replacement requires re-adaptation.</p><p>This ecological effect actually applies to the downstream of chips: the terminal industry.</p><p>Xiaomi Group&#39;s products span a price range from a few yuan to hundreds of thousands of yuan per unit. However, AIOT, which has the lowest overall unit price, is Xiaomi&#39;s most profitable segment: in the first quarter of 2025, Xiaomi&#39;s AIOT segment achieved a gross profit margin of 25.2%, higher than that of mobile phones and even cars. Furthermore, across the entire industry, the profit margin of &quot;small devices&quot; such as smart wristbands is generally higher than that of &quot;large devices&quot; such as smartphones and PCs.</p><p>This also reflects a pattern in the entire digital economy industry: the smaller and more fragmented the demand, the more capable it is of building an ecological network.</p><p>03 Inherently secure job</p><p>Another key to a comfortable life in the small chip business is: not getting involved.</p><p>Football players earning tens of millions of yuan annually are undoubtedly in the limelight, but as soon as their performance declines, they will be fired by their bosses. In elite teams, those who can truly work until retirement are often the front-desk staff, logistics personnel, and cleaners.</p><p>The role of small chips in the system is similar to that of administrative positions.</p><p>Gaming enthusiasts all know that the majority of a computer&#39;s performance is determined by the CPU and graphics card (GPU). As for peripheral chips such as power management chips and USB chips, their impact on gaming performance is almost negl igible. Players&#39; requirements for them can be summed up in two words:</p><p>stable.</p><p>As a result, CPU and GPU manufacturers have embarked on fierce technological competition, undergoing a revolution in manufacturing processes every few years and constantly pushing the boundaries of performance. This has led to MediaTek, the largest shipper of mobile phone CPUs, achieving operating profit margins last year comparable to those of food and beverage brands.</p><p>In the demand for small chips represented by Broadcom, there is not much pursuit of performance. Until 2020, Broadcom&#39;s Bluetooth audio chip was still using 55-nanometer technology, which was a process that Snapdragon had eliminated 10 years ago. Of course, for such chips, using the latest process is unreasonable. The market demand for them is for the stability of the entire system, and for them to assist SOC and GPU well.</p><p>As for the field of electronic control chips, which companies like Infineon are involved in, it&#39;s even more of a stable game.</p><p>Their task is often just to supply power. What customers demand is solid technology, long-term proven feasibility, reliability under special conditions, compatibility with equipment, and stable supply capacity. As for performance? That&#39;s the job of a car engine. It&#39;s not the task of an electronic control chip.</p><p>Upon comparing financial report data, it can be observed that manufacturers of small chips have consistently lower research and development expenses compared to those of large chip manufacturers. For instance, in 2024, Infineon&#39;s research and development expenses accounted for 13% of its revenue, whereas in the same year, Intel&#39;s research and development expenses exceeded 30% of its revenue.</p><p>Stability is precisely the advantage of &quot;vested interests&quot;. Company bosses may hire a programmer with stronger technical skills, but they will not easily replace their well-coordinated secretary. Their products have been tested for many years, and even if newcomers surpass them in certain parameters, it is difficult for them to catch up in terms of adaptability and stability. This is exactly what happened to AMD and Intel in the past few years. AMD&#39;s Ryzen series surpassed Intel in performance, but when it comes to reliability, gamers still prefer Intel&#39;s Core series.</p><p>While the game rules do not involve performance, the market size of small chips is also very limited.</p><p>For example, the market size of microcontroller processors, where Infineon is located, is only a fraction of the SOC chip market. Even if a giant company invests in research and development to capture this market, what significance can it have?</p><p>The meat is substantial, yet the bones are tough to crack. The market for &quot;small chips&quot; is extremely fragmented, with numerous SKUs and diverse fields, making it exceptionally complex and challenging to penetrate. Just as NVIDIA faced strong resistance from Google, AMD, and Broadcom in the AI field, when it comes to the field of embodied robots, which are cheaper, have a smaller market scale, and have more complex demands, NVIDIA has no rivals at all.</p><p>Over the past few decades, Intel has declined, NVIDIA has faced its low points, and AMD has experienced its downfalls, yet these small chip and edge chip manufacturers have consistently maintained stability. Their growth potential may not be as alluring as AMD&#39;s, but they have a secure job.</p><p><br/></p>The growth potential for these small chip and edge chip manufacturers may not be as attractive as that of AMD, but they have a secure job.Author | Yang ZhichaoIn 2010, Apple first equipped its own A-series self-developed chips on the iPhone 4. The core processor SOC is the jewel in the crown of huma �10005adminadmin2025-07-30 11:41:29 2025-07-30 11:44:2140