收到vennyfan兄弟的9800GT～～～上图说明情况～～～

eqzero

回复 43# ONLY


    散热、电池都是我拆的，这两块板子都是我之前研究无轴电脑时花客户钱折腾的，目标是小企业录用设备.......

eqzero

回复 44# ONLY


    恩，下次需要电子元件的话，只要不很急、不太偏门的话，我们都可以找供应商要免费样品的......

金鹏

回复 43# ONLY

兄弟牛

3213的神医

金鹏

回复 25# caozhonghua

这两天RP爆发，98GT和96GSO混插总是出现EUE，今天重装了驱动再试试看

[14:31:16] Completed 25%
[14:33:01] Completed 26%
[14:34:45] Completed 27%
[14:36:30] Completed 28%
[14:37:38] Run: exception thrown during GuardedRun
[14:37:38] Run: exception thrown in GuardedRun -- Gromacs cannot continue further.
[14:37:38] Going to send back what have done -- stepsTotalG=15000000
[14:37:38] Work fraction=0.2865 steps=15000000.
[14:37:42] logfile size=15716 infoLength=15716 edr=0 trr=23
[14:37:42] + Opened results file
[14:37:42] - Writing 16252 bytes of core data to disk...
[14:37:42] Done: 15740 -> 4666 (compressed to 29.6 percent)
[14:37:42]   ... Done.
[14:37:42] DeleteFrameFiles: successfully deleted file=work/wudata_03.ckp
[14:37:42]
[14:37:42] Folding@home Core Shutdown: EARLY_UNIT_END
[14:37:44] CoreStatus = 72 (114)
[14:37:44] Sending work to server
[14:37:44] Project: 10511 (Run 5, Clone 945, Gen 5)
[14:37:44] - Read packet limit of 540015616... Set to 524286976.


[14:37:44] + Attempting to send results [July 14 14:37:44 UTC]
[14:37:45] + Results successfully sent
[14:37:45] Thank you for your contribution to Folding@Home.
[14:37:49] - Preparing to get new work unit...
[14:37:49] + Attempting to get work packet
[14:37:49] - Connecting to assignment server
[14:37:51] - Successful: assigned to (171.64.65.61).
[14:37:51] + News From Folding@Home: Welcome to Folding@Home
[14:37:51] Loaded queue successfully.
[14:37:53] + Closed connections
[14:37:58]
[14:37:58] + Processing work unit
[14:37:58] Core required: FahCore_11.exe
[14:37:58] Core found.
[14:37:58] Working on queue slot 04 [July 14 14:37:58 UTC]
[14:37:58] + Working ...
[14:37:58]
[14:37:58] *------------------------------*
[14:37:58] Folding@Home GPU Core
[14:37:58] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[14:37:58]
[14:37:58] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[14:37:58] Build host: amoeba
[14:37:58] Board Type: Nvidia
[14:37:58] Core      :
[14:37:58] Preparing to commence simulation
[14:37:58] - Looking at optimizations...
[14:37:58] DeleteFrameFiles: successfully deleted file=work/wudata_04.ckp
[14:37:58] - Created dyn
[14:37:58] - Files status OK
[14:37:58] - Expanded 73881 -> 383588 (decompressed 519.1 percent)
[14:37:58] Called DecompressByteArray: compressed_data_size=73881 data_size=383588, decompressed_data_size=383588 diff=0
[14:37:58] - Digital signature verified
[14:37:58]
[14:37:58] Project: 6605 (Run 10, Clone 521, Gen 242)
[14:37:58]
[14:37:58] Assembly optimizations on if available.
[14:37:58] Entering M.D.
[14:38:04] Tpr hash work/wudata_04.tpr:  179754323 3409893053 3835613840 2577839537 857494302
[14:38:04]
[14:38:04] Calling fah_main args: 14 usage=100
[14:38:04]
[14:38:04] Working on Protein
[14:38:04] mdrun_gpu returned
[14:38:04] Self-test failure
[14:38:04]
[14:38:04] Folding@home Core Shutdown: UNSTABLE_MACHINE
[14:38:08] CoreStatus = 7A (122)
[14:38:08] Sending work to server
[14:38:08] Project: 6605 (Run 10, Clone 521, Gen 242)
[14:38:08] - Read packet limit of 540015616... Set to 524286976.
[14:38:08] - Error: Could not get length of results file work/wuresults_04.dat
[14:38:08] - Error: Could not read unit 04 file. Removing from queue.
[14:38:08] - Preparing to get new work unit...
[14:38:08] + Attempting to get work packet
[14:38:08] - Connecting to assignment server
[14:38:12] - Successful: assigned to (171.64.65.61).
[14:38:12] + News From Folding@Home: Welcome to Folding@Home
[14:38:12] Loaded queue successfully.
[14:38:14] + Closed connections
[14:38:19]
[14:38:19] + Processing work unit
[14:38:19] Core required: FahCore_11.exe
[14:38:19] Core found.
[14:38:19] Working on queue slot 05 [July 14 14:38:19 UTC]
[14:38:19] + Working ...
[14:38:19]
[14:38:19] *------------------------------*
[14:38:19] Folding@Home GPU Core
[14:38:19] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[14:38:19]
[14:38:19] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[14:38:19] Build host: amoeba
[14:38:19] Board Type: Nvidia
[14:38:19] Core      :
[14:38:19] Preparing to commence simulation
[14:38:19] - Looking at optimizations...
[14:38:19] DeleteFrameFiles: successfully deleted file=work/wudata_05.ckp
[14:38:19] - Created dyn
[14:38:19] - Files status OK
[14:38:19] - Expanded 73750 -> 383588 (decompressed 520.1 percent)
[14:38:19] Called DecompressByteArray: compressed_data_size=73750 data_size=383588, decompressed_data_size=383588 diff=0
[14:38:19] - Digital signature verified
[14:38:19]
[14:38:19] Project: 6603 (Run 6, Clone 618, Gen 214)
[14:38:19]
[14:38:19] Assembly optimizations on if available.
[14:38:19] Entering M.D.
[14:38:25] Tpr hash work/wudata_05.tpr:  2800354173 3406071782 1707120220 961573766 307902076
[14:38:25]
[14:38:25] Calling fah_main args: 14 usage=100
[14:38:25]
[14:38:25] Working on Protein
[14:38:25] mdrun_gpu returned
[14:38:25] Self-test failure
[14:38:25]
[14:38:25] Folding@home Core Shutdown: UNSTABLE_MACHINE
[14:38:29] CoreStatus = 7A (122)
[14:38:29] Sending work to server
[14:38:29] Project: 6603 (Run 6, Clone 618, Gen 214)
[14:38:29] - Read packet limit of 540015616... Set to 524286976.
[14:38:29] - Error: Could not get length of results file work/wuresults_05.dat
[14:38:29] - Error: Could not read unit 05 file. Removing from queue.
[14:38:29] - Preparing to get new work unit...
[14:38:29] + Attempting to get work packet
[14:38:29] - Connecting to assignment server
[14:38:30] - Successful: assigned to (171.64.65.61).
[14:38:30] + News From Folding@Home: Welcome to Folding@Home
[14:38:30] Loaded queue successfully.
[14:38:31] + Closed connections
[14:38:36]
[14:38:36] + Processing work unit
[14:38:36] Core required: FahCore_11.exe
[14:38:36] Core found.
[14:38:37] Working on queue slot 06 [July 14 14:38:37 UTC]
[14:38:37] + Working ...
[14:38:37]
[14:38:37] *------------------------------*
[14:38:37] Folding@Home GPU Core
[14:38:37] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[14:38:37]
[14:38:37] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[14:38:37] Build host: amoeba
[14:38:37] Board Type: Nvidia
[14:38:37] Core      :
[14:38:37] Preparing to commence simulation
[14:38:37] - Looking at optimizations...
[14:38:37] DeleteFrameFiles: successfully deleted file=work/wudata_06.ckp
[14:38:37] - Created dyn
[14:38:37] - Files status OK
[14:38:37] - Expanded 62556 -> 335048 (decompressed 535.5 percent)
[14:38:37] Called DecompressByteArray: compressed_data_size=62556 data_size=335048, decompressed_data_size=335048 diff=0
[14:38:37] - Digital signature verified
[14:38:37]
[14:38:37] Project: 10516 (Run 5, Clone 953, Gen 1)
[14:38:37]
[14:38:37] Assembly optimizations on if available.
[14:38:37] Entering M.D.
[14:38:43] Tpr hash work/wudata_06.tpr:  504870711 492967750 2522943611 314041953 1789844659
[14:38:43]
[14:38:43] Calling fah_main args: 14 usage=100
[14:38:43]
[14:38:43] Working on Protein
[14:38:43] mdrun_gpu returned
[14:38:43] Self-test failure
[14:38:43]
[14:38:43] Folding@home Core Shutdown: UNSTABLE_MACHINE
[14:38:47] CoreStatus = 7A (122)
[14:38:47] Sending work to server
[14:38:47] Project: 10516 (Run 5, Clone 953, Gen 1)
[14:38:47] - Read packet limit of 540015616... Set to 524286976.
[14:38:47] - Error: Could not get length of results file work/wuresults_06.dat
[14:38:47] - Error: Could not read unit 06 file. Removing from queue.
[14:38:47] - Preparing to get new work unit...
[14:38:47] + Attempting to get work packet
[14:38:47] - Connecting to assignment server
[14:38:48] - Successful: assigned to (171.64.65.61).
[14:38:48] + News From Folding@Home: Welcome to Folding@Home
[14:38:48] Loaded queue successfully.
[14:38:50] + Closed connections
[14:38:55]
[14:38:55] + Processing work unit
[14:38:55] Core required: FahCore_11.exe
[14:38:55] Core found.
[14:38:55] Working on queue slot 07 [July 14 14:38:55 UTC]
[14:38:55] + Working ...
[14:38:55]
[14:38:55] *------------------------------*
[14:38:55] Folding@Home GPU Core
[14:38:55] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[14:38:55]
[14:38:55] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[14:38:55] Build host: amoeba
[14:38:55] Board Type: Nvidia
[14:38:55] Core      :
[14:38:55] Preparing to commence simulation
[14:38:55] - Looking at optimizations...
[14:38:55] DeleteFrameFiles: successfully deleted file=work/wudata_07.ckp
[14:38:55] - Created dyn
[14:38:55] - Files status OK
[14:38:55] - Expanded 73805 -> 383588 (decompressed 519.7 percent)
[14:38:55] Called DecompressByteArray: compressed_data_size=73805 data_size=383588, decompressed_data_size=383588 diff=0
[14:38:55] - Digital signature verified
[14:38:55]
[14:38:55] Project: 6605 (Run 0, Clone 43, Gen 193)
[14:38:55]
[14:38:55] Assembly optimizations on if available.
[14:38:55] Entering M.D.
[14:39:01] Tpr hash work/wudata_07.tpr:  3762373386 2638540392 3928433691 3355971754 3700201311
[14:39:01]
[14:39:01] Calling fah_main args: 14 usage=100
[14:39:01]
[14:39:02] Working on Protein
[14:39:02] mdrun_gpu returned
[14:39:02] Self-test failure
[14:39:02]
[14:39:02] Folding@home Core Shutdown: UNSTABLE_MACHINE
[14:39:05] CoreStatus = 7A (122)
[14:39:05] Sending work to server
[14:39:05] Project: 6605 (Run 0, Clone 43, Gen 193)
[14:39:05] - Read packet limit of 540015616... Set to 524286976.
[14:39:05] - Error: Could not get length of results file work/wuresults_07.dat
[14:39:05] - Error: Could not read unit 07 file. Removing from queue.
[14:39:05] - Preparing to get new work unit...
[14:39:05] + Attempting to get work packet
[14:39:05] - Connecting to assignment server
[14:39:09] - Successful: assigned to (171.64.65.61).
[14:39:09] + News From Folding@Home: Welcome to Folding@Home
[14:39:10] Loaded queue successfully.
[14:39:12] + Closed connections
[14:39:17]
[14:39:17] + Processing work unit
[14:39:17] Core required: FahCore_11.exe
[14:39:17] Core found.
[14:39:17] Working on queue slot 08 [July 14 14:39:17 UTC]
[14:39:17] + Working ...
[14:39:17]
[14:39:17] *------------------------------*
[14:39:17] Folding@Home GPU Core

eqzero

回复 49# 金鹏


    最近是很多怪情况，温度、包、驱动一股脑都在变化，累死了.......

    好惨，之前跑bigadv，cpu或者gpu总有个PPD受损，好容易调整出来smp2配多显卡，bigadv不配显卡，这样PPD到了150K，可是好景不长，bigadv接不到包都算了，尽接些6701毒包，然后smp2也不稳定了，现在又得调回来，还是要根据sp数量搭配显卡了，水冷啊！

caozhonghua

回复 49# 金鹏

辛苦兄弟了
昨天中午发现显卡一个包算完就停住了，发现的时候已经停了5小时
关掉程序重开又好了
今天早晨到公司发现断电了

金鹏

回复 50# eqzero

目前CPU和GPU客户端貌似过渡阶段，等V7整合客户端出来估计会稳定

我这些天也是机器接连出问题，主机接连好几天莫名其妙停止响应死机，单位的小三出现混插EUE，一停就是24小时。

PS:兄弟全套水冷折腾的确是高强度的，要淡定

金鹏

回复 51# caozhonghua

估计和最近有些WU有关，两台机器上都出现过兄弟你的情况。

单卡机器倒是正常

eqzero

回复 52# 金鹏


    哎！说实话，我是带研发出身的，非常不看好v7整合类的东西，整合意味着效能损失，不整合意味着生手不便.....

eqzero

当年老黄靠M$整合的dx干掉了3dfx，今天发现需要发挥性能的时候不得已祭出了cuda，而ati现在最想的就是利用通用的dc、opencl摸灭nv的长处.......

金鹏

回复 54# eqzero

V7是通过不同的CORE内核来实现各种计算的，感觉FAH开始学习伯克利的做法了，V7客户端只是个平台，实际计算是通过不同的CORE内核实现，客户端起到的作用甄别硬件类型并下载相关程序文件和内核。

我个人很期待V7，那时候我们论坛的教程就简单多了，有利于扩大参与人数。

PS：现在要参与FAH没有一定基础还真搞不定

eqzero

回复 56# 金鹏


    难度有时候也是成就。

zglloo

回复 57# eqzero


    话说 难度也让这种降低到普及化不方便~ 我仍然觉得boinc做的很好~

eqzero

回复 58# zglloo


    个人喜欢吧，seti就是因为没什么难度，天天看着机器傻跑，所以没有动力继续了，就像我买硬件不喜欢重复一样......

金鹏

回复  金鹏


    难度有时候也是成就。
eqzero 发表于 2010-7-15 12:33

难度=折腾