3213之Pascal架构TITAN X(P)/GTX1080(Ti)/1070(Ti)/1060(Ti)(含ns/day)统计贴 - 第6页 - Folding@Home 模拟蛋白质折叠 - 中国分布式计算论坛

horst1981 发表于 2016-11-11 23:25:31

本帖最后由 horst1981 于 2016-11-11 23:46 编辑

开始更新TITAN X:

TITANX，P11707，PPD≈990K，TPF=1:47,folding speed=120 ns/day ns/day（有波动），GPU主频≈1710MHz（有波动）
截取部分log如下Completed 225000 out of 7500000 steps (3%)
Performance since last checkpoint: 120 ns/day
Completed 450000 out of 7500000 steps (6%)
Performance since last checkpoint: 119.0082645 ns/day
Completed 975000 out of 7500000 steps (13%)
Performance since last checkpoint: 118.3561644 ns/day
Completed 1500000 out of 7500000 steps (20%)
Performance since last checkpoint: 120 ns/day
Completed 1725000 out of 7500000 steps (23%)
Performance since last checkpoint: 121.0084034 ns/day
Completed 1950000 out of 7500000 steps (26%)
Performance since last checkpoint: 121.3483146 ns/day
Completed 5250000 out of 7500000 steps (70%)
Performance since last checkpoint: 120.3342618 ns/day
P9675（Core18)，TPF＝8.6sec，GPU主频＝1809.5MHz，PPD＝540K部分log 截取如下15:36:03:WU00:FS00:0x18:Completed 1000000 out of 2000000 steps (50%)
15:36:12:WU00:FS00:0x18:Completed 1020000 out of 2000000 steps (51%)
15:36:21:WU00:FS00:0x18:Completed 1040000 out of 2000000 steps (52%)
15:36:29:WU00:FS00:0x18:Completed 1060000 out of 2000000 steps (53%)
15:36:37:WU00:FS00:0x18:Completed 1080000 out of 2000000 steps (54%)
15:36:45:WU00:FS00:0x18:Completed 1100000 out of 2000000 steps (55%)
15:36:55:WU00:FS00:0x18:Completed 1120000 out of 2000000 steps (56%)
15:37:03:WU00:FS00:0x18:Completed 1140000 out of 2000000 steps (57%)
15:37:11:WU00:FS00:0x18:Completed 1160000 out of 2000000 steps (58%)

horst1981 发表于 2016-11-12 01:08:15

本帖最后由 horst1981 于 2016-11-12 19:17 编辑

P13205,TPF=1:27,PPD=950K,频率＝1417.5MHz
部分log截取如下Completed 200000 out of 2000000 steps (10%)
Performance since last checkpoint: 25.3372434 ns/day
Completed 400000 out of 2000000 steps (20%)
Performance since last checkpoint: 24.89913545 ns/day
Completed 600000 out of 2000000 steps (30%)
Performance since last checkpoint: 22.52933507 ns/day
Completed 800000 out of 2000000 steps (40%)
Performance since last checkpoint: 22.21079692 ns/day
Completed 1000000 out of 2000000 steps (50%)
Performance since last checkpoint: 22.21079692 ns/dayP10494,PPD=940K，TPF=2：21，GPU主频1417M到1755M之间波动温度84。部分log截取如下<blockquote>Completed 50000 out of 1000000 steps (5%)P9212,PPD=685K(真低），TPF=3：34，主频1522M到1759M温度84。感觉得不稳定
部分 log截取如下Completed 100000 out of 2500000 steps (4%)
Performance since last checkpoint: 20.35335689 ns/day
Completed 200000 out of 2500000 steps (8%)
Performance since last checkpoint: 20.25791325 ns/day
Completed 300000 out of 2500000 steps (12%)
Performance since last checkpoint: 20.30552291 ns/day
Completed 400000 out of 2500000 steps (16%)
Performance since last checkpoint: 20.18691589 ns/day
Completed 500000 out of 2500000 steps (20%)
Performance since last checkpoint: 20.25791325 ns/day
Completed 600000 out of 2500000 steps (24%)
Performance since last checkpoint: 20.18691589 ns/day
Completed 700000 out of 2500000 steps (28%)
Performance since last checkpoint: 20.13986014 ns/day

Completed 2200000 out of 2500000 steps (88%)
Performance since last checkpoint: 20.11641444 ns/day
Completed 2300000 out of 2500000 steps (92%)
Performance since last checkpoint: 20.13986014 ns/day
Completed 2400000 out of 2500000 steps (96%)
Performance since last checkpoint: 20.16336056 ns/day

金鹏发表于 2016-12-15 21:33:43

本帖最后由金鹏于 2016-12-15 21:36 编辑

horst1981 发表于 2016-11-12 01:08
P13205,TPF=1:27,PPD=950K,频率＝1417.5MHz
部分log截取如下P10494,PPD=940K，TPF=2：21，GPU主频1417M到17 ...
2025/5005MHZ P11712 tpf=1:42Project: 11712 (Run 2, Clone 39, Gen 1)

*************************** Core21 Folding@home Core ***************************
   Type: 33
   Core: Core21
Website: http://folding.stanford.edu/
Copyright: (c) 2009-2014 Stanford University
Author: Yutong Zhao <yutong.zhao@stanford.edu>
   Args: -dir 00 -suffix 01 -version 704 -lifeline 9812 -checkpoint 3 -gpu 0
         -gpu-vendor nvidia
Config: <none>
************************************ Build *************************************
Version: 0.0.17
   Date: Dec 18 2015
   Time: 23:07:48
Compiler: Visual C++ 2008
Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
Platform: win32 7
   Bits: 64
   Mode: Release
************************************ System ************************************
   CPU: Unknown
CPU ID:
   CPUs: 8
Memory: 9.95GiB
Free Memory: 5.46GiB
Threads: WINDOWS_THREADS
OS Version: 6.1
Has Battery: false
On Battery: false
UTC Offset: 8
   PID: 1820
   CWD: C:\FAHClient\FAHClient\work
      OS: Windows 7 Ultimate Service Pack 1
OS Arch: AMD64
   GPUs: 0
   CUDA: Not detected
OpenCL: Not detected
********************************************************************************
Folding@home GPU Core21 Folding@home Core

Version 0.0.17

compatible platform(s):
-- 0 --
PROFILE = FULL_PROFILE
VERSION = OpenCL 1.2 CUDA 8.0.0
NAME = NVIDIA CUDA
VENDOR = NVIDIA Corporation

(2) device(s) found on platform 0:
-- 0 --
DEVICE_NAME = TITAN X (Pascal)
DEVICE_VENDOR = NVIDIA Corporation
DEVICE_VERSION = OpenCL 1.2 CUDA

-- 1 --
DEVICE_NAME = GeForce GTX 1080
DEVICE_VENDOR = NVIDIA Corporation
DEVICE_VERSION = OpenCL 1.2 CUDA

[ Entering Init ]
Launch time: 2016-12-15T12:58:58Z
Arguments passed: -dir 00 -suffix 01 -version 704 -lifeline 9812 -checkpoint 3 -gpu 0 -gpu-vendor nvidia
For testState comparison of CPU and GPU, will use:
forceTolerance: 5 kJ/mol/nm
energyTolerance: 10 kJ/mol
[ LeavingInit ]
[ Entering Main ]
Reading core settings...
Total number of steps: 7500000
XTC write frequency: 250000
Checkpoint write frequency: 250000 (3.3%)
Number of frames per WU: 30
[ Initializing Core Contexts ]
Using platform OpenCL
Looking for vendor: nvidia...found on platformId 0
Setting platform precision to mixed
Deserializing System...
Found MonteCarloBarostat @ 1.01325 (default) Bar, 300 Kelvin, 25 pressure change frequency.
Found: 35515 atoms, 6 forces.
Deserializing State...done.
Ewald error tolerance in force 3 is 0.0005
Ewald parameters: alpha 2.920289872087185 nx 63 ny 63 nz 63
Integrator Type: class OpenMM::LangevinIntegrator
Constraint Tolerance: 1e-005
Time Step in PS: 0.002
Temperature: 300
Friction Coeff: 1
Using CPU platform for reference calculations.
Checking core state against reference...
Checking checkpoint state against reference...
[ Initialized Core Contexts... ]
Using OpenCL on platformId 0 and gpu 0
v(^_^)vMD ready starting from step 0

Completed 0 out of 7500000 steps (0%)

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

Completed 75000 out of 7500000 steps (1%)

Completed 150000 out of 7500000 steps (2%)

Completed 225000 out of 7500000 steps (3%)

Performance since last checkpoint: 124.4956772 ns/day
Completed 300000 out of 7500000 steps (4%)

Completed 375000 out of 7500000 steps (5%)

Completed 450000 out of 7500000 steps (6%)

Performance since last checkpoint: 126.686217 ns/day
Completed 525000 out of 7500000 steps (7%)

Completed 600000 out of 7500000 steps (8%)

Completed 675000 out of 7500000 steps (9%)

Completed 750000 out of 7500000 steps (10%)

Performance since last checkpoint: 126.3157895 ns/day
Completed 825000 out of 7500000 steps (11%)

Completed 900000 out of 7500000 steps (12%)

Completed 975000 out of 7500000 steps (13%)

Performance since last checkpoint: 127.4336283 ns/day

凤如流云 发表于 2017-1-19 09:47:54

本帖最后由凤如流云于 2017-1-20 10:22 编辑

这里补充一个1050ti的数据！平台是i7 4790K+Z97 显卡是技嘉1050tiwf2oc 用MSI AB解锁功耗墙（125% 97℃）并且默认电压超频+150mhz

跑P11710 PPD是19.1万左右显卡驱动是376.33 工作频率1873Mhz
晚上（1.19）回来更新了376.60驱动跑P9179 PPD是14.2万左右工作频率仍然是1873Mhz但是开浏览器的时候会降频到1328Mhz 很呵呵（关闭浏览器硬件加速解决）

1.20早上跑P11710 PPD是20.1万（200730）显卡驱动是376.60 工作频率仍然是1873Mhz

horst1981 发表于 2017-1-19 13:11:53

凤如流云发表于 2017-1-19 09:47
这里补充一个1050ti的数据！平台是i7 4790K+Z97 显卡是技嘉1050tiwf2oc 用MSI AB解锁功耗墙（125% 97℃）并 ...

19万的PPD，那还真是挺强的啊。

金鹏发表于 2017-3-15 15:02:58

本帖最后由金鹏于 2017-3-15 15:06 编辑

GTX1080TI+190MHZ @2062mhz/5508

Project: 13500 (Run 0, Clone 138, Gen 112)

Unit: 0x000000a88ca304f457a358592504288a

CPU: 0x00000000000000000000000000000000

Machine: 1

*************************** Core21 Folding@home Core ***************************
   Type: 33
   Core: Core21
Website: http://folding.stanford.edu/
Copyright: (c) 2009-2014 Stanford University
Author: Yutong Zhao <yutong.zhao@stanford.edu>
   Args: -dir 02 -suffix 01 -version 704 -lifeline 2588 -checkpoint 3 -gpu 0
         -gpu-vendor nvidia
Config: <none>
************************************ Build *************************************
Version: 0.0.18
   Date: Jan 20 2017
   Time: 10:32:43
Compiler: Visual C++ 2008
Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox -arch:SSE2 /MT
Platform: win32 7
   Bits: 64
   Mode: Release
************************************ System ************************************
   CPU: Unknown
CPU ID:
   CPUs: 8
Memory: 9.95GiB
Free Memory: 6.86GiB
Threads: WINDOWS_THREADS
OS Version: 6.1
Has Battery: false
On Battery: false
UTC Offset: 8
   PID: 2672
   CWD: C:\FAHClient\FAHClient\work
      OS: Windows 7 Ultimate Service Pack 1
OS Arch: AMD64
   GPUs: 0
   CUDA: Not detected
OpenCL: Not detected
********************************************************************************
Folding@home GPU Core21 Folding@home Core

Version 0.0.18

compatible platform(s):
-- 0 --
PROFILE = FULL_PROFILE
VERSION = OpenCL 1.2 CUDA 8.0.0
NAME = NVIDIA CUDA
VENDOR = NVIDIA Corporation

(2) device(s) found on platform 0:
-- 0 --
DEVICE_NAME = GeForce GTX 1080 Ti
DEVICE_VENDOR = NVIDIA Corporation
DEVICE_VERSION = OpenCL 1.2 CUDA

-- 1 --
DEVICE_NAME = GeForce GTX 1080
DEVICE_VENDOR = NVIDIA Corporation
DEVICE_VERSION = OpenCL 1.2 CUDA

v(^_^)vMD ready starting from step 0

Completed 0 out of 5000000 steps (0%)

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

Completed 50000 out of 5000000 steps (1%)

Completed 100000 out of 5000000 steps (2%)

Performance since last checkpoint: 126.3157895 ns/day
Completed 150000 out of 5000000 steps (3%)

Completed 200000 out of 5000000 steps (4%)

Completed 250000 out of 5000000 steps (5%)

Performance since last checkpoint: 127.0588235 ns/day
Completed 300000 out of 5000000 steps (6%)

Completed 350000 out of 5000000 steps (7%)

Performance since last checkpoint: 127.0588235 ns/day
Completed 400000 out of 5000000 steps (8%)

Completed 450000 out of 5000000 steps (9%)

Completed 500000 out of 5000000 steps (10%)

Performance since last checkpoint: 126.3157895 ns/day
Completed 550000 out of 5000000 steps (11%)

Completed 600000 out of 5000000 steps (12%)

Performance since last checkpoint: 127.0588235 ns/day
Completed 650000 out of 5000000 steps (13%)

Completed 700000 out of 5000000 steps (14%)

Completed 750000 out of 5000000 steps (15%)

Performance since last checkpoint: 127.0588235 ns/day

horst1981 发表于 2017-3-16 11:40:57

金鹏发表于 2017-3-15 15:02
GTX1080TI+190MHZ @2062mhz/5508

不上win10试试？

freestman 发表于 2017-3-16 12:24:08

horst1981 发表于 2017-3-16 11:40
不上win10试试？

Win10可不是个稳定的系统，各种傲娇替你做主。

金鹏发表于 2017-3-16 14:50:18

horst1981 发表于 2017-3-16 11:40
不上win10试试？

怕折腾，估计跑FAH下还是WIN7给力

horst1981 发表于 2017-4-2 08:26:41

收到金版@金鹏寄回的TTX，耽搁了两天，
拆了散热器和信仰灯，上了个山寨厂的水冷头，保留原风扇。

X99+E5 2683V3 +Ubuntu16.10平台，378.13开源驱动，默认设置，未超频。

P9188，TDF30秒，主频1822MHz，内存5004（NVIDIA X Server Settings显示9026）,

PPD有137万，温度66度上下。

金鹏发表于 2017-4-2 08:48:37

horst1981 发表于 2017-4-2 08:26
收到金版@金鹏寄回的TTX，耽搁了两天，
拆了散热器和信仰灯，上了个山寨厂的水冷头，保留原风扇。

兄弟上水一定要时刻关注安全啊,卡卡怕水水

LINUX下容易出成绩,口水满地那个小超一下跑13200包估计LINUX下可以上探160万PPD

horst1981 发表于 2017-4-2 08:52:07

金鹏发表于 2017-4-2 08:48
兄弟上水一定要时刻关注安全啊,卡卡怕水水

LINUX下容易出成绩,口水满地那个小超一下跑13200 ...

不知是驱动的问题还是其他问题，装上ttX后，NVIDIA X Server Settings 的超频显项居然没有出来。之前1070是可以的。
现在只能按默认设置跑，连风扇转速都调不了。

这两天找时间再折腾折腾。

金鹏发表于 2017-4-2 08:57:50

horst1981 发表于 2017-4-2 08:52
不知是驱动的问题还是其他问题，装上ttX后，NVIDIA X Server Settings 的超频显项居然没有出来。之前1070 ...

把原来驱动卸载掉,再全新安装一次驱动试试

horst1981 发表于 2017-4-2 11:40:43

金鹏发表于 2017-4-2 08:57
把原来驱动卸载掉,再全新安装一次驱动试试

已经重装了驱动还是不行，看来要另想办法

freestman 发表于 2017-4-2 12:19:32

horst1981 发表于 2017-4-2 08:26
收到金版@金鹏寄回的TTX，耽搁了两天，
拆了散热器和信仰灯，上了个山寨厂的水冷头，保留原风扇。

照片拍一下

页: 1 2 3 4 5 [6] 7

中国分布式计算论坛's Archiver