收藏本站

 找回密码
 新注册用户
搜索
中国分布式计算论坛»论坛 › 非 BOINC 平台的主流项目 › Folding@Home 模拟蛋白质折叠 › 【发现】core21内置性能监测功能
12下一页
返回列表 发新帖
查看: 9151|回复: 29

【发现】core21内置性能监测功能

[复制链接]
wpf999
发表于 2016-1-30 17:05:16 | 显示全部楼层 |阅读模式
当运行core21时，在 work/{wu-id}/01 目录下，有一个log.txt文件，其中可以看到该GPU运行该算例的算力性能。

例如C:\ProgramData\FAHClient\work\01\01\log.txt
  1. Project: 9205 (Run 0, Clone 39, Gen 100)

  3. Unit: 0x000000f6664f2dd055d4c817a79c054e

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 0

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 01 -suffix 01 -version 704 -lifeline 6160 -checkpoint 3 -gpu 0
  26.              -gpu-vendor nvidia
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 2
  41.      Memory: 1.81GiB
  42. Free Memory: 828.47MiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 5780
  49.         CWD: C:\ProgramData\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [2] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.2
  64.   NAME = Intel(R) OpenCL
  65.   VENDOR = Intel(R) Corporation
  66.   -- 1 --
  67.   PROFILE = FULL_PROFILE
  68.   VERSION = OpenCL 1.2 CUDA 7.5.0
  69.   NAME = NVIDIA CUDA
  70.   VENDOR = NVIDIA Corporation

  72. (2) device(s) found on platform 0:
  73.   -- 0 --
  74.   DEVICE_NAME =         Intel(R) Pentium(R) CPU G2030 @ 3.00GHz
  75.   DEVICE_VENDOR = Intel(R) Corporation
  76.   DEVICE_VERSION = OpenCL 1.2 (Build 76427)

  78.   -- 1 --
  79.   DEVICE_NAME = Intel(R) HD Graphics
  80.   DEVICE_VENDOR = Intel(R) Corporation
  81.   DEVICE_VERSION = OpenCL 1.2

  83. (1) device(s) found on platform 1:
  84.   -- 0 --
  85.   DEVICE_NAME = GeForce GTX 960
  86.   DEVICE_VENDOR = NVIDIA Corporation
  87.   DEVICE_VERSION = OpenCL 1.2 CUDA

  89. [ Entering Init ]
  90.   Launch time: 2016-01-30T06:49:04Z
  91.   Arguments passed: -dir 01 -suffix 01 -version 704 -lifeline 6160 -checkpoint 3 -gpu 0 -gpu-vendor nvidia
  92.   For testState comparison of CPU and GPU, will use:
  93.     forceTolerance: 5 kJ/mol/nm
  94.     energyTolerance: 10 kJ/mol
  95. [ Leaving  Init ]
  96. [ Entering Main ]
  97.   Reading core settings...
  98.   Total number of steps: 2500000
  99.   XTC write frequency: 100000
  100.   Checkpoint write frequency: 100000 (4%)
  101.   Number of frames per WU: 25
  102. [ Initializing Core Contexts ]
  103.   Using platform OpenCL
  104.   Looking for vendor: nvidia...found on platformId 1
  105.   Setting platform precision to single
  106.   Deserializing System...
  107.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 25 pressure change frequency.
  108.     Found: 116418 atoms, 6 forces.
  109.   Deserializing State...  done.
  110.     Ewald error tolerance in force 3 is 0.0005
  111.     Ewald parameters: alpha 2.628260884878466 nx 90 ny 90 nz 147
  112.     Integrator Type: class OpenMM::LangevinIntegrator
  113.     Constraint Tolerance: 1e-005
  114.     Time Step in PS: 0.002
  115.     Temperature: 300
  116.     Friction Coeff: 1
  117.     Using CPU platform for reference calculations.
  118.   Checking core state against reference...
  119.   Checking checkpoint state against reference...
  120. [ Initialized Core Contexts... ]
  121.   Using OpenCL on platformId 1 and gpu 0
  122.   v(^_^)v  MD ready starting from step 0

  124. Completed 0 out of 2500000 steps (0%)

  126. Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900

  128. Completed 25000 out of 2500000 steps (1%)

  130. Completed 50000 out of 2500000 steps (2%)

  132. Completed 75000 out of 2500000 steps (3%)

  134. Completed 100000 out of 2500000 steps (4%)

  136. Performance since last checkpoint: 7.143447706 ns/day
  137. Completed 125000 out of 2500000 steps (5%)

  139. Completed 150000 out of 2500000 steps (6%)

  141. Completed 175000 out of 2500000 steps (7%)

  143. Completed 200000 out of 2500000 steps (8%)

  145. Performance since last checkpoint: 7.152317881 ns/day
  146. Completed 225000 out of 2500000 steps (9%)

  148. Completed 250000 out of 2500000 steps (10%)

  150. Completed 275000 out of 2500000 steps (11%)

  152. Completed 300000 out of 2500000 steps (12%)

  154. Performance since last checkpoint: 7.152317881 ns/day
  155. Completed 325000 out of 2500000 steps (13%)
复制代码

评分

参与人数 2基本分 +35 收起 理由
金鹏 + 20 牛!这也能发现
CCCP0081 + 15 辛苦了！

查看全部评分

[2020.03.10] BOINC 版本 7.16.5 客户端已正式发布，推荐大家更新！
回复

使用道具 举报

wpf999
 楼主| 发表于 2016-1-30 17:11:27 | 显示全部楼层
可以看到有 7.143447706 ns/day 之类的输出

这表明960在运行9205任务包的性能为 7.14纳秒/天，含义为在960上一天可以模拟7.14纳秒的蛋白质折叠过程。 该指标反映了此GPU的科学计算性能，比PPD更有价值。
【优化程序】【SETI@home】Windows 系统的一体化优化程序安装器 v0.41，自助式简便安装各种版本优化程序，推荐使用。
回复

使用道具 举报

zhouxiaobo
发表于 2016-1-30 17:41:50 | 显示全部楼层
刚刚送走一个CORE21的包马上又来CORE18的，晚了一步了。看不到ns/day的参数。
[建议]上传图片前先进来了解下相关经验
回复

使用道具 举报

金鹏
发表于 2016-1-30 19:32:51 | 显示全部楼层
本帖最后由 金鹏 于 2016-1-30 19:33 编辑

超频1521的980TI的45.18828452 ns/day
  1. Project: 10493 (Run 2, Clone 1, Gen 31)

  3. Unit: 0x0000002d8ca304f555d6166f2056870c

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 00 -suffix 01 -version 704 -lifeline 2472 -checkpoint 3 -gpu 0
  26.              -gpu-vendor nvidia -tmax=86 -twait=900
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 12
  41.      Memory: 9.99GiB
  42. Free Memory: 5.97GiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 6904
  49.         CWD: C:\FAHClient\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [1] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.2 CUDA 8.0.5
  64.   NAME = NVIDIA CUDA
  65.   VENDOR = NVIDIA Corporation

  67. (1) device(s) found on platform 0:
  68.   -- 0 --
  69.   DEVICE_NAME = GeForce GTX 980 Ti
  70.   DEVICE_VENDOR = NVIDIA Corporation
  71.   DEVICE_VERSION = OpenCL 1.2 CUDA

  73. [ Entering Init ]
  74.   Launch time: 2016-01-30T07:44:29Z
  75.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 2472 -checkpoint 3 -gpu 0 -gpu-vendor nvidia -tmax=86 -twait=900
  76.   For testState comparison of CPU and GPU, will use:
  77.     forceTolerance: 5 kJ/mol/nm
  78.     energyTolerance: 10 kJ/mol
  79. [ Leaving  Init ]
  80. [ Entering Main ]
  81.   Reading core settings...
  82.   Total number of steps: 5000000
  83.   XTC write frequency: 125000
  84.   Checkpoint write frequency: 125000 (2.5%)
  85.   Number of frames per WU: 40
  86. [ Initializing Core Contexts ]
  87.   Using platform OpenCL
  88.   Looking for vendor: nvidia...found on platformId 0
  89.   Setting platform precision to single
  90.   Deserializing System...
  91.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  92.     Found: 80983 atoms, 6 forces.
  93.   Deserializing State...  done.
  94.     Ewald error tolerance in force 3 is 0.0005
  95.     Ewald parameters: alpha 2.628260884878466 nx 75 ny 75 nz 75
  96.     Integrator Type: class OpenMM::LangevinIntegrator
  97.     Constraint Tolerance: 1e-005
  98.     Time Step in PS: 0.002
  99.     Temperature: 300
  100.     Friction Coeff: 1
  101.     Using CPU platform for reference calculations.
  102.   Checking core state against reference...
  103.   Checking checkpoint state against reference...
  104. [ Initialized Core Contexts... ]
  105.   Using OpenCL on platformId 0 and gpu 0
  106.   v(^_^)v  MD ready starting from step 0

  108. Completed 0 out of 5000000 steps (0%)

  110. Core21 Folding@home Core: single GPU Temperature Control enabled, tmax: 86 twait: 900

  112. Completed 50000 out of 5000000 steps (1%)

  114. Completed 100000 out of 5000000 steps (2%)

  116. Performance since last checkpoint: 44.72049689 ns/day
  117. Completed 150000 out of 5000000 steps (3%)

  119. Completed 200000 out of 5000000 steps (4%)

  121. Completed 250000 out of 5000000 steps (5%)

  123. Performance since last checkpoint: 44.72049689 ns/day
  124. Completed 300000 out of 5000000 steps (6%)

  126. Completed 350000 out of 5000000 steps (7%)

  128. Performance since last checkpoint: 44.81327801 ns/day
  129. Completed 400000 out of 5000000 steps (8%)

  131. Completed 450000 out of 5000000 steps (9%)

  133. Completed 500000 out of 5000000 steps (10%)

  135. Performance since last checkpoint: 44.90644491 ns/day
  136. Completed 550000 out of 5000000 steps (11%)

  138. WARNING:Console control signal 1 on PID 6904

  140. Exiting, please wait. . .

  142. 2016-01-30T08:20:26Z
  143. [ Leaving  Main ]
  144. Folding@home Core Shutdown: INTERRUPTED

  146. Project: 10493 (Run 2, Clone 1, Gen 31)

  148. Unit: 0x0000002d8ca304f555d6166f2056870c

  150. CPU: 0x00000000000000000000000000000000

  152. Machine: 1

  154. Digital signatures verified

  156. *************************** Core21 Folding@home Core ***************************
  157.        Type: 33
  158.        Core: Core21
  159.     Website: http://folding.stanford.edu/
  160.   Copyright: (c) 2009-2014 Stanford University
  161.      Author: Yutong Zhao <[email protected]>
  162.        Args: -dir 00 -suffix 01 -version 704 -lifeline 5916 -checkpoint 3 -gpu 0
  163.              -gpu-vendor nvidia -tmax=86 -twait=900
  164.      Config: <none>
  165. ************************************ Build *************************************
  166.     Version: 0.0.17
  167.        Date: Dec 18 2015
  168.        Time: 23:07:48
  169.    Compiler: Visual C++ 2008
  170.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  171.    Platform: win32 7
  172.        Bits: 64
  173.        Mode: Release
  174. ************************************ System ************************************
  175.         CPU: Unknown
  176.      CPU ID:
  177.        CPUs: 12
  178.      Memory: 9.99GiB
  179. Free Memory: 6.00GiB
  180.     Threads: WINDOWS_THREADS
  181. OS Version: 6.1
  182. Has Battery: false
  183. On Battery: false
  184. UTC Offset: 8
  185.         PID: 8760
  186.         CWD: C:\FAHClient\FAHClient\work
  187.          OS: Windows 7 Ultimate Service Pack 1
  188.     OS Arch: AMD64
  189.        GPUs: 0
  190.        CUDA: Not detected
  191.      OpenCL: Not detected
  192. ********************************************************************************
  193. Folding@home GPU Core21 Folding@home Core

  195. Version 0.0.17

  197. [1] compatible platform(s):
  198.   -- 0 --
  199.   PROFILE = FULL_PROFILE
  200.   VERSION = OpenCL 1.2 CUDA 8.0.5
  201.   NAME = NVIDIA CUDA
  202.   VENDOR = NVIDIA Corporation

  204. (1) device(s) found on platform 0:
  205.   -- 0 --
  206.   DEVICE_NAME = GeForce GTX 980 Ti
  207.   DEVICE_VENDOR = NVIDIA Corporation
  208.   DEVICE_VERSION = OpenCL 1.2 CUDA

  210. [ Entering Init ]
  211.   Launch time: 2016-01-30T08:20:31Z
  212.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 5916 -checkpoint 3 -gpu 0 -gpu-vendor nvidia -tmax=86 -twait=900
  213.   Found a checkpoint file

  215.   For testState comparison of CPU and GPU, will use:
  216.     forceTolerance: 5 kJ/mol/nm
  217.     energyTolerance: 10 kJ/mol
  218. [ Leaving  Init ]
  219. [ Entering Main ]
  220.   Reading core settings...
  221.   Total number of steps: 5000000
  222.   XTC write frequency: 125000
  223.   Checkpoint write frequency: 125000 (2.5%)
  224.   Number of frames per WU: 40
  225. [ Initializing Core Contexts ]
  226.   Using platform OpenCL
  227.   Looking for vendor: nvidia...found on platformId 0
  228.   Setting platform precision to single
  229.   Deserializing System...
  230.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  231.     Found: 80983 atoms, 6 forces.
  232.   Deserializing State...  done.
  233.     Ewald error tolerance in force 3 is 0.0005
  234.     Ewald parameters: alpha 2.628260884878466 nx 75 ny 75 nz 75
  235.     Integrator Type: class OpenMM::LangevinIntegrator
  236.     Constraint Tolerance: 1e-005
  237.     Time Step in PS: 0.002
  238.     Temperature: 300
  239.     Friction Coeff: 1
  240.     Using CPU platform for reference calculations.
  241.   Checking core state against reference...
  242.   Checking checkpoint state against reference...
  243. [ Initialized Core Contexts... ]
  244.   Using OpenCL on platformId 0 and gpu 0
  245.   v(^_^)v  MD ready starting from step 500000

  247. Completed 500000 out of 5000000 steps (10%)

  249. Core21 Folding@home Core: single GPU Temperature Control enabled, tmax: 86 twait: 900

  251. Completed 550000 out of 5000000 steps (11%)

  253. Completed 600000 out of 5000000 steps (12%)

  255. Performance since last checkpoint: 45.18828452 ns/day
  256. Completed 650000 out of 5000000 steps (13%)

  258. Completed 700000 out of 5000000 steps (14%)

  260. Completed 750000 out of 5000000 steps (15%)

  262. Performance since last checkpoint: 45.18828452 ns/day
  263. Completed 800000 out of 5000000 steps (16%)

  265. Completed 850000 out of 5000000 steps (17%)

  267. Performance since last checkpoint: 45.18828452 ns/day
  268. Completed 900000 out of 5000000 steps (18%)

  270. Completed 950000 out of 5000000 steps (19%)

  272. Completed 1000000 out of 5000000 steps (20%)

  274. Performance since last checkpoint: 45.18828452 ns/day
  275. Completed 1050000 out of 5000000 steps (21%)

  277. Completed 1100000 out of 5000000 steps (22%)

  279. Performance since last checkpoint: 45.09394572 ns/day
  280. Completed 1150000 out of 5000000 steps (23%)

  282. Completed 1200000 out of 5000000 steps (24%)

  284. Completed 1250000 out of 5000000 steps (25%)

  286. Performance since last checkpoint: 44.81327801 ns/day
  287. Completed 1300000 out of 5000000 steps (26%)

  289. Completed 1350000 out of 5000000 steps (27%)

  291. Performance since last checkpoint: 45.09394572 ns/day
  292. Completed 1400000 out of 5000000 steps (28%)

  294. Completed 1450000 out of 5000000 steps (29%)

  296. Completed 1500000 out of 5000000 steps (30%)

  298. Performance since last checkpoint: 45.09394572 ns/day
  299. Completed 1550000 out of 5000000 steps (31%)

  301. Completed 1600000 out of 5000000 steps (32%)

  303. Performance since last checkpoint: 45.09394572 ns/day
  304. Completed 1650000 out of 5000000 steps (33%)

  306. Completed 1700000 out of 5000000 steps (34%)

  308. Completed 1750000 out of 5000000 steps (35%)

  310. Performance since last checkpoint: 45.09394572 ns/day
  311. Completed 1800000 out of 5000000 steps (36%)

  313. Completed 1850000 out of 5000000 steps (37%)

  315. Performance since last checkpoint: 45.09394572 ns/day
  316. Completed 1900000 out of 5000000 steps (38%)

  318. Completed 1950000 out of 5000000 steps (39%)

  320. Completed 2000000 out of 5000000 steps (40%)

  322. Performance since last checkpoint: 45.18828452 ns/day
  323. Completed 2050000 out of 5000000 steps (41%)

  325. Completed 2100000 out of 5000000 steps (42%)

  327. Performance since last checkpoint: 45.18828452 ns/day
  328. Completed 2150000 out of 5000000 steps (43%)

  330. Completed 2200000 out of 5000000 steps (44%)

  332. Completed 2250000 out of 5000000 steps (45%)

  334. Performance since last checkpoint: 45.18828452 ns/day
  335. Completed 2300000 out of 5000000 steps (46%)

  337. Completed 2350000 out of 5000000 steps (47%)

  339. Performance since last checkpoint: 45.18828452 ns/day
  340. Completed 2400000 out of 5000000 steps (48%)

  342. Completed 2450000 out of 5000000 steps (49%)

  344. Completed 2500000 out of 5000000 steps (50%)

  346. Performance since last checkpoint: 45.18828452 ns/day
  347. Completed 2550000 out of 5000000 steps (51%)

  349. Completed 2600000 out of 5000000 steps (52%)

  351. Performance since last checkpoint: 45.28301887 ns/day
  352. Completed 2650000 out of 5000000 steps (53%)

  354. Completed 2700000 out of 5000000 steps (54%)

  356. Completed 2750000 out of 5000000 steps (55%)

  358. Performance since last checkpoint: 45.09394572 ns/day
  359. Completed 2800000 out of 5000000 steps (56%)

  361. Completed 2850000 out of 5000000 steps (57%)

  363. Performance since last checkpoint: 44.90644491 ns/day
  364. Completed 2900000 out of 5000000 steps (58%)

  366. Completed 2950000 out of 5000000 steps (59%)

  368. Completed 3000000 out of 5000000 steps (60%)

  370. Performance since last checkpoint: 44.72049689 ns/day
  371. Completed 3050000 out of 5000000 steps (61%)

  373. Completed 3100000 out of 5000000 steps (62%)

  375. Performance since last checkpoint: 44.72049689 ns/day
  376. Completed 3150000 out of 5000000 steps (63%)

  378. Completed 3200000 out of 5000000 steps (64%)

  380. Completed 3250000 out of 5000000 steps (65%)

  382. Performance since last checkpoint: 44.81327801 ns/day
  383. Completed 3300000 out of 5000000 steps (66%)

  385. Completed 3350000 out of 5000000 steps (67%)

  387. Performance since last checkpoint: 45 ns/day
  388. Completed 3400000 out of 5000000 steps (68%)
复制代码
[2020.04.17] Folding@home V7.6.8 客户端已正式发布，推荐大家更新！
回复

使用道具 举报

金鹏
发表于 2016-1-30 19:46:47 | 显示全部楼层
相同的卡在不同的包下应该有不同的模拟表现,
否则最甜的最容易爆包的11402用980跑50.46728972 ns/day居然比980TI还高就无法解释了
  1. Project: 11402 (Run 6, Clone 12, Gen 40)

  3. Unit: 0x000000328ca304f255ed4e85aaebae88

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 02 -suffix 01 -version 704 -lifeline 17680 -checkpoint 3 -gpu
  26.              0 -gpu-vendor nvidia -tmax=83 -twait=900
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 8
  41.      Memory: 7.95GiB
  42. Free Memory: 4.99GiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 16756
  49.         CWD: d:\FAHClient\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [1] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.1 CUDA 7.0.29
  64.   NAME = NVIDIA CUDA
  65.   VENDOR = NVIDIA Corporation

  67. (1) device(s) found on platform 0:
  68.   -- 0 --
  69.   DEVICE_NAME = GeForce GTX 980
  70.   DEVICE_VENDOR = NVIDIA Corporation
  71.   DEVICE_VERSION = OpenCL 1.1 CUDA

  73. [ Entering Init ]
  74.   Launch time: 2016-01-30T11:24:12Z
  75.   Arguments passed: -dir 02 -suffix 01 -version 704 -lifeline 17680 -checkpoint 3 -gpu 0 -gpu-vendor nvidia -tmax=83 -twait=900
  76.   For testState comparison of CPU and GPU, will use:
  77.     forceTolerance: 5 kJ/mol/nm
  78.     energyTolerance: 10 kJ/mol
  79. [ Leaving  Init ]
  80. [ Entering Main ]
  81.   Reading core settings...
  82.   Total number of steps: 5000000
  83.   XTC write frequency: 125000
  84.   Checkpoint write frequency: 125000 (2.5%)
  85.   Number of frames per WU: 40
  86. [ Initializing Core Contexts ]
  87.   Using platform OpenCL
  88.   Looking for vendor: nvidia...found on platformId 0
  89.   Setting platform precision to single
  90.   Deserializing System...
  91.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  92.     Found: 54377 atoms, 5 forces.
  93.   Deserializing State...  done.
  94.     Ewald error tolerance in force 3 is 0.0005
  95.     Ewald parameters: alpha 2.628260884878466 nx 70 ny 70 nz 70
  96.     Integrator Type: class OpenMM::LangevinIntegrator
  97.     Constraint Tolerance: 1e-005
  98.     Time Step in PS: 0.002
  99.     Temperature: 300
  100.     Friction Coeff: 1
  101.     Using CPU platform for reference calculations.
  102.   Checking core state against reference...
  103.   Checking checkpoint state against reference...
  104. [ Initialized Core Contexts... ]
  105.   Using OpenCL on platformId 0 and gpu 0
  106.   v(^_^)v  MD ready starting from step 0

  108. Completed 0 out of 5000000 steps (0%)

  110. Core21 Folding@home Core: single GPU Temperature Control enabled, tmax: 83 twait: 900

  112. Completed 50000 out of 5000000 steps (1%)

  114. Completed 100000 out of 5000000 steps (2%)

  116. Performance since last checkpoint: 50.46728972 ns/day
  117. Completed 150000 out of 5000000 steps (3%)

  119. Completed 200000 out of 5000000 steps (4%)

  121. Completed 250000 out of 5000000 steps (5%)

  123. Performance since last checkpoint: 50.46728972 ns/day
  124. Completed 300000 out of 5000000 steps (6%)
复制代码
【优化程序】【SETI@home】Windows 系统的一体化优化程序安装器 v0.41，自助式简便安装各种版本优化程序，推荐使用。
回复

使用道具 举报

Lynt
发表于 2016-1-30 19:49:48 | 显示全部楼层
  1. Project: 11703 (Run 0, Clone 32, Gen 45)
  2. Unit: 0x0000002e8ca304f3568833fb694304c9
  3. CPU: 0x00000000000000000000000000000000
  4. Machine: 1
  5. Reading tar file core.xml
  6. Reading tar file system.xml
  7. Reading tar file integrator.xml
  8. Reading tar file state.xml
  9. Digital signatures verified
  10. *************************** Core21 Folding@home Core ***************************
  11.        Type: 33
  12.        Core: Core21
  13.     Website: http://folding.stanford.edu/
  14.   Copyright: (c) 2009-2014 Stanford University
  15.      Author: Yutong Zhao <[email protected]>
  16.        Args: -dir 00 -suffix 01 -version 704 -lifeline 4146 -checkpoint 3 -gpu 0
  17.              -gpu-vendor nvidia
  18.      Config: <none>
  19. ************************************ Build *************************************
  20.     Version: 0.0.17
  21.        Date: Dec 18 2015
  22.        Time: 22:48:55
  23. Repository: Git
  24.    Revision: 99a324d4be70589b0a127080bb7fdfe05cc4597f
  25.      Branch: HEAD
  26.    Compiler: GNU 4.4.7 20120313 (Red Hat 4.4.7-16)
  27.     Options: -std=gnu++98 -O3 -funroll-loops -mfpmath=sse -ffast-math
  28.              -fno-unsafe-math-optimizations -msse2
  29.    Platform: linux2 4.1.13-boot2docker
  30.        Bits: 64
  31.        Mode: Release
  32. ************************************ System ************************************
  33.         CPU: Intel(R) Core(TM) i5-4590 CPU @ 3.30GHz
  34.      CPU ID: GenuineIntel Family 6 Model 60 Stepping 3
  35.        CPUs: 4
  36.      Memory: 7.72GiB
  37. Free Memory: 5.86GiB
  38.     Threads: POSIX_THREADS
  39. OS Version: 3.16
  40. Has Battery: false
  41. On Battery: false
  42. UTC Offset: 8
  43.         PID: 4150
  44.         CWD: /var/lib/fahclient/work
  45.          OS: Linux 3.16.0-31-generic x86_64
  46.     OS Arch: AMD64
  47.        GPUs: 0
  48.        CUDA: Not detected
  49.      OpenCL: Not detected
  50. ********************************************************************************
  51. Folding@home GPU Core21 Folding@home Core
  52. Version 0.0.17
  53. [1] compatible platform(s):
  54.   -- 0 --
  55.   PROFILE = FULL_PROFILE
  56.   VERSION = OpenCL 1.1 CUDA 7.0.65
  57.   NAME = NVIDIA CUDA
  58.   VENDOR = NVIDIA Corporation

  60. (2) device(s) found on platform 0:
  61.   -- 0 --
  62.   DEVICE_NAME = GeForce GTX 980 Ti
  63.   DEVICE_VENDOR = NVIDIA Corporation
  64.   DEVICE_VERSION = OpenCL 1.1 CUDA

  66.   -- 1 --
  67.   DEVICE_NAME = GeForce GTX 970
  68.   DEVICE_VENDOR = NVIDIA Corporation
  69.   DEVICE_VERSION = OpenCL 1.1 CUDA

  71. [ Entering Init ]
  72.   Launch time: 2016-01-30T11:26:05Z
  73.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 4146 -checkpoint 3 -gpu 0 -gpu-vendor nvidia
  74.   For testState comparison of CPU and GPU, will use:
  75.     forceTolerance: 5 kJ/mol/nm
  76.     energyTolerance: 10 kJ/mol
  77. [ Leaving  Init ]
  78. [ Entering Main ]
  79.   Reading core settings...
  80.   Total number of steps: 5000000
  81.   XTC write frequency: 250000
  82.   Checkpoint write frequency: 250000 (5%)
  83.   Number of frames per WU: 20
  84. [ Initializing Core Contexts ]
  85.   Using platform OpenCL
  86.   Looking for vendor: nvidia...found on platformId 0
  87.   Setting platform precision to mixed
  88.   Deserializing System...
  89.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 25 pressure change frequency.
  90.     Found: 68556 atoms, 6 forces.
  91.   Deserializing State...  done.
  92.     Ewald error tolerance in force 3 is 0.0005
  93.     Ewald parameters: alpha 2.766590405135227 nx 75 ny 75 nz 75
  94.     Integrator Type: N6OpenMM18LangevinIntegratorE
  95.     Constraint Tolerance: 1e-05
  96.     Time Step in PS: 0.002
  97.     Temperature: 300
  98.     Friction Coeff: 0.25
  99.     Using CPU platform for reference calculations.
  100.   Checking core state against reference...
  101.   Checking checkpoint state against reference...
  102. [ Initialized Core Contexts... ]
  103.   Using OpenCL on platformId 0 and gpu 0
  104.   v(^_^)v  MD ready starting from step 0

  106. Completed 0 out of 5000000 steps (0%)
  107. Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
  108. Completed 50000 out of 5000000 steps (1%)
  109. Completed 100000 out of 5000000 steps (2%)
  110. Completed 150000 out of 5000000 steps (3%)
  111. Completed 200000 out of 5000000 steps (4%)
  112. Completed 250000 out of 5000000 steps (5%)
  113. Performance since last checkpoint: 50.46728972 ns/day
  114. Completed 300000 out of 5000000 steps (6%)
复制代码
Linux下980Ti@1493MHz
【优化程序】【SETI@home】Windows 系统的一体化优化程序安装器 v0.41，自助式简便安装各种版本优化程序，推荐使用。
回复

使用道具 举报

faner
发表于 2016-1-30 20:09:59 | 显示全部楼层
这屏幕刷的好。又有新发现。
[建议]上传图片前先进来了解下相关经验
回复

使用道具 举报

cuda
发表于 2016-1-30 21:12:33 | 显示全部楼层
不同atom数的包之间直接比较ns/day恐怕没有意义，相同atom数的可以比一比。
以前v6客户端会给出GFlops为单位的Performance，感觉更直观一些。

【Collatz Conjecture】优化程序，支持 x86 SSE CPU 指令集，以及 Windows 系统下 AMD GPU，和 Linux 系统下 AMD & Nvidia GPU
回复

使用道具 举报

金鹏
发表于 2016-1-30 22:50:10 | 显示全部楼层
本帖最后由 金鹏 于 2016-1-30 22:54 编辑
cuda 发表于 2016-1-30 21:12
不同atom数的包之间直接比较ns/day恐怕没有意义，相同atom数的可以比一比。
以前v6客户端会给出GFlops为单 ...

是的,兄弟精辟不过相比17/18老内核新内核计算效率和密度都高了不少

1485频率下980TI跑的11402的包,62.60869565 ns/day
1513频率下980跑的11402的包,50.46728972 ns/day
1471频率下970跑的11402包,39.488117 ns/day

log970.txt (7.13 KB, 下载次数: 0)
log980.txt (3.77 KB, 下载次数: 0)
log980ti.txt (7.51 KB, 下载次数: 0)
  1. Project: 11402 (Run 1, Clone 24, Gen 49)

  3. Unit: 0x000000448ca304f255ed4e4f93c36a39

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 00 -suffix 01 -version 704 -lifeline 9672 -checkpoint 3 -gpu 0
  26.              -gpu-vendor nvidia
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 32
  41.      Memory: 31.95GiB
  42. Free Memory: 25.30GiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 8948
  49.         CWD: D:\FAHClient\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [1] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.2 CUDA 8.0.5
  64.   NAME = NVIDIA CUDA
  65.   VENDOR = NVIDIA Corporation

  67. (2) device(s) found on platform 0:
  68.   -- 0 --
  69.   DEVICE_NAME = GeForce GTX 980 Ti
  70.   DEVICE_VENDOR = NVIDIA Corporation
  71.   DEVICE_VERSION = OpenCL 1.2 CUDA

  73.   -- 1 --
  74.   DEVICE_NAME = GeForce GTX 980 Ti
  75.   DEVICE_VENDOR = NVIDIA Corporation
  76.   DEVICE_VERSION = OpenCL 1.2 CUDA

  78. [ Entering Init ]
  79.   Launch time: 2016-01-30T12:16:39Z
  80.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 9672 -checkpoint 3 -gpu 0 -gpu-vendor nvidia
  81.   For testState comparison of CPU and GPU, will use:
  82.     forceTolerance: 5 kJ/mol/nm
  83.     energyTolerance: 10 kJ/mol
  84. [ Leaving  Init ]
  85. [ Entering Main ]
  86.   Reading core settings...
  87.   Total number of steps: 5000000
  88.   XTC write frequency: 125000
  89.   Checkpoint write frequency: 125000 (2.5%)
  90.   Number of frames per WU: 40
  91. [ Initializing Core Contexts ]
  92.   Using platform OpenCL
  93.   Looking for vendor: nvidia...found on platformId 0
  94.   Setting platform precision to single
  95.   Deserializing System...
  96.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  97.     Found: 54377 atoms, 5 forces.
  98.   Deserializing State...  done.
  99.     Ewald error tolerance in force 3 is 0.0005
  100.     Ewald parameters: alpha 2.628260884878466 nx 70 ny 70 nz 70
  101.     Integrator Type: class OpenMM::LangevinIntegrator
  102.     Constraint Tolerance: 1e-005
  103.     Time Step in PS: 0.002
  104.     Temperature: 300
  105.     Friction Coeff: 1
  106.     Using CPU platform for reference calculations.
  107.   Checking core state against reference...
  108.   Checking checkpoint state against reference...
  109. [ Initialized Core Contexts... ]
  110.   Using OpenCL on platformId 0 and gpu 0
  111.   v(^_^)v  MD ready starting from step 0

  113. Completed 0 out of 5000000 steps (0%)

  115. Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900

  117. Completed 50000 out of 5000000 steps (1%)

  119. Completed 100000 out of 5000000 steps (2%)

  121. Performance since last checkpoint: 62.60869565 ns/day
  122. Completed 150000 out of 5000000 steps (3%)

  124. Completed 200000 out of 5000000 steps (4%)

  126. Completed 250000 out of 5000000 steps (5%)

  128. Performance since last checkpoint: 62.60869565 ns/day
复制代码
  1. Project: 11402 (Run 6, Clone 12, Gen 40)

  3. Unit: 0x000000328ca304f255ed4e85aaebae88

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 02 -suffix 01 -version 704 -lifeline 17680 -checkpoint 3 -gpu
  26.              0 -gpu-vendor nvidia -tmax=83 -twait=900
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 8
  41.      Memory: 7.95GiB
  42. Free Memory: 4.99GiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 16756
  49.         CWD: d:\FAHClient\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [1] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.1 CUDA 7.0.29
  64.   NAME = NVIDIA CUDA
  65.   VENDOR = NVIDIA Corporation

  67. (1) device(s) found on platform 0:
  68.   -- 0 --
  69.   DEVICE_NAME = GeForce GTX 980
  70.   DEVICE_VENDOR = NVIDIA Corporation
  71.   DEVICE_VERSION = OpenCL 1.1 CUDA

  73. [ Entering Init ]
  74.   Launch time: 2016-01-30T11:24:12Z
  75.   Arguments passed: -dir 02 -suffix 01 -version 704 -lifeline 17680 -checkpoint 3 -gpu 0 -gpu-vendor nvidia -tmax=83 -twait=900
  76.   For testState comparison of CPU and GPU, will use:
  77.     forceTolerance: 5 kJ/mol/nm
  78.     energyTolerance: 10 kJ/mol
  79. [ Leaving  Init ]
  80. [ Entering Main ]
  81.   Reading core settings...
  82.   Total number of steps: 5000000
  83.   XTC write frequency: 125000
  84.   Checkpoint write frequency: 125000 (2.5%)
  85.   Number of frames per WU: 40
  86. [ Initializing Core Contexts ]
  87.   Using platform OpenCL
  88.   Looking for vendor: nvidia...found on platformId 0
  89.   Setting platform precision to single
  90.   Deserializing System...
  91.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  92.     Found: 54377 atoms, 5 forces.
  93.   Deserializing State...  done.
  94.     Ewald error tolerance in force 3 is 0.0005
  95.     Ewald parameters: alpha 2.628260884878466 nx 70 ny 70 nz 70
  96.     Integrator Type: class OpenMM::LangevinIntegrator
  97.     Constraint Tolerance: 1e-005
  98.     Time Step in PS: 0.002
  99.     Temperature: 300
  100.     Friction Coeff: 1
  101.     Using CPU platform for reference calculations.
  102.   Checking core state against reference...
  103.   Checking checkpoint state against reference...
  104. [ Initialized Core Contexts... ]
  105.   Using OpenCL on platformId 0 and gpu 0
  106.   v(^_^)v  MD ready starting from step 0

  108. Completed 0 out of 5000000 steps (0%)

  110. Core21 Folding@home Core: single GPU Temperature Control enabled, tmax: 83 twait: 900

  112. Completed 50000 out of 5000000 steps (1%)

  114. Completed 100000 out of 5000000 steps (2%)

  116. Performance since last checkpoint: 50.46728972 ns/day
  117. Completed 150000 out of 5000000 steps (3%)

  119. Completed 200000 out of 5000000 steps (4%)

  121. Completed 250000 out of 5000000 steps (5%)

  123. Performance since last checkpoint: 50.46728972 ns/day
复制代码
  1. Project: 11402 (Run 5, Clone 45, Gen 55)

  3. Unit: 0x0000004b8ca304f255ed4e8574d71ef0

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 00 -suffix 01 -version 704 -lifeline 7428 -checkpoint 3 -gpu 0
  26.              -gpu-vendor nvidia
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 8
  41.      Memory: 7.99GiB
  42. Free Memory: 5.23GiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 7208
  49.         CWD: E:\FAHClient\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [1] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.1 CUDA 7.0.29
  64.   NAME = NVIDIA CUDA
  65.   VENDOR = NVIDIA Corporation

  67. (2) device(s) found on platform 0:
  68.   -- 0 --
  69.   DEVICE_NAME = GeForce GTX 970
  70.   DEVICE_VENDOR = NVIDIA Corporation
  71.   DEVICE_VERSION = OpenCL 1.1 CUDA

  73.   -- 1 --
  74.   DEVICE_NAME = GeForce GTX 970
  75.   DEVICE_VENDOR = NVIDIA Corporation
  76.   DEVICE_VERSION = OpenCL 1.1 CUDA

  78. [ Entering Init ]
  79.   Launch time: 2016-01-30T11:15:54Z
  80.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 7428 -checkpoint 3 -gpu 0 -gpu-vendor nvidia
  81.   For testState comparison of CPU and GPU, will use:
  82.     forceTolerance: 5 kJ/mol/nm
  83.     energyTolerance: 10 kJ/mol
  84. [ Leaving  Init ]
  85. [ Entering Main ]
  86.   Reading core settings...
  87.   Total number of steps: 5000000
  88.   XTC write frequency: 125000
  89.   Checkpoint write frequency: 125000 (2.5%)
  90.   Number of frames per WU: 40
  91. [ Initializing Core Contexts ]
  92.   Using platform OpenCL
  93.   Looking for vendor: nvidia...found on platformId 0
  94.   Setting platform precision to single
  95.   Deserializing System...
  96.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  97.     Found: 54377 atoms, 5 forces.
  98.   Deserializing State...  done.
  99.     Ewald error tolerance in force 3 is 0.0005
  100.     Ewald parameters: alpha 2.628260884878466 nx 70 ny 70 nz 70
  101.     Integrator Type: class OpenMM::LangevinIntegrator
  102.     Constraint Tolerance: 1e-005
  103.     Time Step in PS: 0.002
  104.     Temperature: 300
  105.     Friction Coeff: 1
  106.     Using CPU platform for reference calculations.
  107.   Checking core state against reference...
  108.   Checking checkpoint state against reference...
  109. [ Initialized Core Contexts... ]
  110.   Using OpenCL on platformId 0 and gpu 0
  111.   v(^_^)v  MD ready starting from step 0

  113. Completed 0 out of 5000000 steps (0%)

  115. Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900

  117. Completed 50000 out of 5000000 steps (1%)

  119. Completed 100000 out of 5000000 steps (2%)

  121. Performance since last checkpoint: 39.488117 ns/day
  122. Completed 150000 out of 5000000 steps (3%)

  124. Completed 200000 out of 5000000 steps (4%)

  126. Completed 250000 out of 5000000 steps (5%)

  128. Performance since last checkpoint: 39.488117 ns/day
复制代码
[2020.03.10] BOINC 版本 7.16.5 客户端已正式发布，推荐大家更新！
回复

使用道具 举报

金鹏
发表于 2016-1-31 17:30:55 | 显示全部楼层
11407下貌似模拟时间最长了77.69784173 ns/day
  1. Project: 11407 (Run 0, Clone 14, Gen 33)

  3. Unit: 0x000000288ca304f25686b2425fe1bc9b

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Reading tar file core.xml

  11. Reading tar file system.xml

  13. Reading tar file integrator.xml

  15. Reading tar file state.xml

  17. Digital signatures verified

  19. *************************** Core21 Folding@home Core ***************************
  20.        Type: 33
  21.        Core: Core21
  22.     Website: http://folding.stanford.edu/
  23.   Copyright: (c) 2009-2014 Stanford University
  24.      Author: Yutong Zhao <[email protected]>
  25.        Args: -dir 00 -suffix 01 -version 704 -lifeline 5416 -checkpoint 3 -gpu 0
  26.              -gpu-vendor nvidia -tmax=86 -twait=900
  27.      Config: <none>
  28. ************************************ Build *************************************
  29.     Version: 0.0.17
  30.        Date: Dec 18 2015
  31.        Time: 23:07:48
  32.    Compiler: Visual C++ 2008
  33.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  34.    Platform: win32 7
  35.        Bits: 64
  36.        Mode: Release
  37. ************************************ System ************************************
  38.         CPU: Unknown
  39.      CPU ID:
  40.        CPUs: 12
  41.      Memory: 9.99GiB
  42. Free Memory: 6.16GiB
  43.     Threads: WINDOWS_THREADS
  44. OS Version: 6.1
  45. Has Battery: false
  46. On Battery: false
  47. UTC Offset: 8
  48.         PID: 10216
  49.         CWD: C:\FAHClient\FAHClient\work
  50.          OS: Windows 7 Ultimate Service Pack 1
  51.     OS Arch: AMD64
  52.        GPUs: 0
  53.        CUDA: Not detected
  54.      OpenCL: Not detected
  55. ********************************************************************************
  56. Folding@home GPU Core21 Folding@home Core

  58. Version 0.0.17

  60. [1] compatible platform(s):
  61.   -- 0 --
  62.   PROFILE = FULL_PROFILE
  63.   VERSION = OpenCL 1.2 CUDA 8.0.5
  64.   NAME = NVIDIA CUDA
  65.   VENDOR = NVIDIA Corporation

  67. (1) device(s) found on platform 0:
  68.   -- 0 --
  69.   DEVICE_NAME = GeForce GTX 980 Ti
  70.   DEVICE_VENDOR = NVIDIA Corporation
  71.   DEVICE_VERSION = OpenCL 1.2 CUDA

  73. [ Entering Init ]
  74.   Launch time: 2016-01-31T07:37:41Z
  75.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 5416 -checkpoint 3 -gpu 0 -gpu-vendor nvidia -tmax=86 -twait=900
  76.   For testState comparison of CPU and GPU, will use:
  77.     forceTolerance: 5 kJ/mol/nm
  78.     energyTolerance: 10 kJ/mol
  79. [ Leaving  Init ]
  80. [ Entering Main ]
  81.   Reading core settings...
  82.   Total number of steps: 5000000
  83.   XTC write frequency: 125000
  84.   Checkpoint write frequency: 125000 (2.5%)
  85.   Number of frames per WU: 40
  86. [ Initializing Core Contexts ]
  87.   Using platform OpenCL
  88.   Looking for vendor: nvidia...found on platformId 0
  89.   Setting platform precision to single
  90.   Deserializing System...
  91.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  92.     Found: 45368 atoms, 5 forces.
  93.   Deserializing State...  done.
  94.     Ewald error tolerance in force 3 is 0.0005
  95.     Ewald parameters: alpha 2.628260884878466 nx 63 ny 63 nz 63
  96.     Integrator Type: class OpenMM::LangevinIntegrator
  97.     Constraint Tolerance: 1e-005
  98.     Time Step in PS: 0.002
  99.     Temperature: 300
  100.     Friction Coeff: 1
  101.     Using CPU platform for reference calculations.
  102.   Checking core state against reference...
  103.   Checking checkpoint state against reference...
  104. [ Initialized Core Contexts... ]
  105.   Using OpenCL on platformId 0 and gpu 0
  106.   v(^_^)v  MD ready starting from step 0

  108. Completed 0 out of 5000000 steps (0%)

  110. Core21 Folding@home Core: single GPU Temperature Control enabled, tmax: 86 twait: 900

  112. Completed 50000 out of 5000000 steps (1%)

  114. Completed 100000 out of 5000000 steps (2%)

  116. Performance since last checkpoint: 77.69784173 ns/day
  117. Completed 150000 out of 5000000 steps (3%)

  119. Completed 200000 out of 5000000 steps (4%)

  121. Completed 250000 out of 5000000 steps (5%)

  123. Performance since last checkpoint: 77.41935484 ns/day
  124. Completed 300000 out of 5000000 steps (6%)

  126. Completed 350000 out of 5000000 steps (7%)

  128. Performance since last checkpoint: 77.14285714 ns/day
复制代码
[2020.04.17] Folding@home V7.6.8 客户端已正式发布，推荐大家更新！
回复

使用道具 举报

wpf999
 楼主| 发表于 2016-2-5 01:09:13 | 显示全部楼层
Project: 11401 (Run 0, Clone 15, Gen 86)
GTX970
蛋白质折叠性能：37.8ns/day


【Collatz Conjecture】优化程序，支持 x86 SSE CPU 指令集，以及 Windows 系统下 AMD GPU，和 Linux 系统下 AMD & Nvidia GPU
回复

使用道具 举报

金鹏
发表于 2016-2-12 20:01:45 | 显示全部楼层
wpf999 发表于 2016-2-5 01:09
Project: 11401 (Run 0, Clone 15, Gen 86)
GTX970
蛋白质折叠性能：37.8ns/day

赶脚10495是目前模拟时间最长了97.2972973 ns/day
  1. Project: 10495 (Run 1, Clone 45, Gen 4)

  3. Unit: 0x000000048ca304f556ba62e7d8c2ac5d

  5. CPU: 0x00000000000000000000000000000000

  7. Machine: 1

  9. Digital signatures verified

  11. *************************** Core21 Folding@home Core ***************************
  12.        Type: 33
  13.        Core: Core21
  14.     Website: http://folding.stanford.edu/
  15.   Copyright: (c) 2009-2014 Stanford University
  16.      Author: Yutong Zhao <[email protected]>
  17.        Args: -dir 00 -suffix 01 -version 704 -lifeline 10600 -checkpoint 3 -gpu
  18.              0 -gpu-vendor nvidia -tmax=86 -twait=900
  19.      Config: <none>
  20. ************************************ Build *************************************
  21.     Version: 0.0.17
  22.        Date: Dec 18 2015
  23.        Time: 23:07:48
  24.    Compiler: Visual C++ 2008
  25.     Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
  26.    Platform: win32 7
  27.        Bits: 64
  28.        Mode: Release
  29. ************************************ System ************************************
  30.         CPU: Unknown
  31.      CPU ID:
  32.        CPUs: 12
  33.      Memory: 9.99GiB
  34. Free Memory: 6.81GiB
  35.     Threads: WINDOWS_THREADS
  36. OS Version: 6.1
  37. Has Battery: false
  38. On Battery: false
  39. UTC Offset: 8
  40.         PID: 9548
  41.         CWD: C:\FAHClient\FAHClient\work
  42.          OS: Windows 7 Ultimate Service Pack 1
  43.     OS Arch: AMD64
  44.        GPUs: 0
  45.        CUDA: Not detected
  46.      OpenCL: Not detected
  47. ********************************************************************************
  48. Folding@home GPU Core21 Folding@home Core

  50. Version 0.0.17

  52. [1] compatible platform(s):
  53.   -- 0 --
  54.   PROFILE = FULL_PROFILE
  55.   VERSION = OpenCL 1.2 CUDA 8.0.5
  56.   NAME = NVIDIA CUDA
  57.   VENDOR = NVIDIA Corporation

  59. (1) device(s) found on platform 0:
  60.   -- 0 --
  61.   DEVICE_NAME = GeForce GTX 980 Ti
  62.   DEVICE_VENDOR = NVIDIA Corporation
  63.   DEVICE_VERSION = OpenCL 1.2 CUDA

  65. [ Entering Init ]
  66.   Launch time: 2016-02-12T11:38:24Z
  67.   Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 10600 -checkpoint 3 -gpu 0 -gpu-vendor nvidia -tmax=86 -twait=900
  68.   Found a checkpoint file

  70.   For testState comparison of CPU and GPU, will use:
  71.     forceTolerance: 5 kJ/mol/nm
  72.     energyTolerance: 10 kJ/mol
  73. [ Leaving  Init ]
  74. [ Entering Main ]
  75.   Reading core settings...
  76.   Total number of steps: 5000000
  77.   XTC write frequency: 250000
  78.   Checkpoint write frequency: 250000 (5%)
  79.   Number of frames per WU: 20
  80. [ Initializing Core Contexts ]
  81.   Using platform OpenCL
  82.   Looking for vendor: nvidia...found on platformId 0
  83.   Setting platform precision to single
  84.   Deserializing System...
  85.   Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 50 pressure change frequency.
  86.     Found: 67896 atoms, 6 forces.
  87.   Deserializing State...  done.
  88.     Ewald error tolerance in force 3 is 0.0005
  89.     Integrator Type: class OpenMM::LangevinIntegrator
  90.     Constraint Tolerance: 1e-005
  91.     Time Step in PS: 0.002
  92.     Temperature: 300
  93.     Friction Coeff: 5
  94.     Using CPU platform for reference calculations.
  95.   Checking core state against reference...
  96.   Checking checkpoint state against reference...
  97. [ Initialized Core Contexts... ]
  98.   Using OpenCL on platformId 0 and gpu 0
  99.   v(^_^)v  MD ready starting from step 250000

  101. Completed 250000 out of 5000000 steps (5%)

  103. Core21 Folding@home Core: single GPU Temperature Control enabled, tmax: 86 twait: 900

  105. Completed 300000 out of 5000000 steps (6%)

  107. Completed 350000 out of 5000000 steps (7%)

  109. Completed 400000 out of 5000000 steps (8%)

  111. Completed 450000 out of 5000000 steps (9%)

  113. Completed 500000 out of 5000000 steps (10%)

  115. Performance since last checkpoint: 97.2972973 ns/day
  116. Completed 550000 out of 5000000 steps (11%)

  118. Completed 600000 out of 5000000 steps (12%)

  120. Completed 650000 out of 5000000 steps (13%)

  122. Completed 700000 out of 5000000 steps (14%)

  124. Completed 750000 out of 5000000 steps (15%)

  126. Performance since last checkpoint: 97.2972973 ns/day
  127. Completed 800000 out of 5000000 steps (16%)

  129. Completed 850000 out of 5000000 steps (17%)

  131. Completed 900000 out of 5000000 steps (18%)
复制代码
【MilkyWay@Home】优化程序，支持多种操作系统多种 CPU 指令集，以及 Windows 系统下部分 AMD GPU
回复

使用道具 举报

horst1981
发表于 2016-2-12 22:12:38 | 显示全部楼层
那以后大家以ns/day论英雄
[2020.03.10] BOINC 版本 7.16.5 客户端已正式发布，推荐大家更新！
回复

使用道具 举报

wpf999
 楼主| 发表于 2016-2-12 22:17:07 | 显示全部楼层
horst1981 发表于 2016-2-12 22:12
那以后大家以ns/day论英雄

我觉得ns/day比point/day更有硬件性能评价意义
【Collatz Conjecture】优化程序，支持 x86 SSE CPU 指令集，以及 Windows 系统下 AMD GPU，和 Linux 系统下 AMD & Nvidia GPU
回复

使用道具 举报

horst1981
发表于 2016-2-12 22:22:12 | 显示全部楼层
wpf999 发表于 2016-2-12 22:17
我觉得ns/day比point/day更有硬件性能评价意义

但是数值听起来PPD更屌啊
我ns/day七点几几,和我PPD二十几万，一般观感来讲后者听起来好像更牛逼
[2020.03.10] BOINC 版本 7.16.5 客户端已正式发布，推荐大家更新！
回复

使用道具 举报

下一页 »
12下一页
返回列表 发新帖
高级模式
B Color Image Link Quote Code Smilies
您需要登录后才可以回帖 登录 | 新注册用户

本版积分规则

论坛官方淘宝店开业啦～
欢迎大家多多支持基金会～

Archiver|手机版|小黑屋|中国分布式计算总站 ( 沪ICP备05042587号 )

GMT+8, 2025-5-12 04:41

Powered by Discuz! X3.5

© 2001-2024 Discuz! Team.

快速回复 返回顶部 返回列表