Core 21 System Requirments?

[ChrisSpeaker]

Hello everyone.
viewtopic.php?f=24&t=28010 Indicates that one of core 21 new project:

p9704
Desc: Voltage gated sodium channel
Atoms: 175k
Base credit: 9000
Extra restriction: 64 bit operating systems only, >2gb system ram (more is recommended)

Exactly how much more RAM does it need? I have 2 gigs but it seems I'm getting:

Code: Select all

17:41:10:WU02:FS00:0x21:ERROR:exception: Error creating array forceBuffers: clCreateBuffer (-61)
17:41:22:WARNING:WU02:FS00:FahCore returned: BAD_WORK_UNIT (114 = 0x72)

Thanks in advance!

[7im]

AFAIK, 64 bit Windows (newer than XP)
2 GB of free system memory (not shared with OS), 4 GB would be safer (as WUs keep getting bigger year after year) (aka 4 GB total system memory or more)
NV 350.xx drivers or newer, AMD 14.9 drivers or newer
Fermi NV or 5xxx AMD or newer (will be tight against the deadline, and may not finish if not folding 24/7) a generation newer is recommended
Subject to correction

[Joe_H]

2 GB is the minimum, as mentioned more RAM is recommended. Depending on the WU you can expect startup to need to page to virtual memory in excess of that. Do you have enough disk space for swap space? Once started, the RAM usage will be lower.

Beyond that, what version of video drivers do you have installed?

[ChrisSpeaker]

AFAIK, 64 bit Windows (newer than XP)
NV 350.xx drivers or newer, AMD 14.9 drivers or newer

That may be the problem. Running Windows Vista 64bit with AMD version 13.4 right now with version 13.12 the latest availiable. Don't know why but AMD seems to have stopped supporting Vista but is still supporting XP 64bit!

Have plenty of disk space (~130Gb free) with 4 Gb swap partation set. Will be adding RAM soon,

[artoar_11]

AFAIK, 64 bit Windows (newer than XP)
2 GB of free system memory (not shared with OS), 4 GB would be safer (as WUs keep getting bigger year after year) (aka 4 GB total system memory or more)
NV 350.xx drivers or newer, AMD 14.9 drivers or newer
Fermi NV or 5xxx AMD or newer (will be tight against the deadline, and may not finish if not folding 24/7) a generation newer is recommended
Subject to correction

3-4 WUs I have received from the core_21. They all ended with failure. Or at the beginning the driver crashed (very long swap) or after few hours.
I understand that 4.0 GB system memory sometimes is not enough. But for my video card GTX 650 Ti (Kepler) the recommended driver is NV 327.23 (for core_17/18). If I installed newer (e.g. NV 350.xx), will not be increase TPF on core_17/18?
I'm a little confused what to do with the driver?

My PC: CPU - 2500K; RAM - 4.0 GB (2.20 - 2.50 GB free); Win7 Pro/64b; GPU - GTX 650Ti (1024 MB VRAM); 4.50 GB paging file.

Yesterday, WU on core_21 failed after the crash the driver.

Code: Select all

08:03:59:WU01:FS00:Starting
08:03:59:WU01:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Arto/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 01 -suffix 01 -version 704 -lifeline 1044 -checkpoint 9 -gpu 0 -gpu-vendor nvidia
08:03:59:WU01:FS00:Started FahCore on PID 3440
08:03:59:WU01:FS00:Core PID:4916
08:03:59:WU01:FS00:FahCore 0x21 started
08:04:00:WU01:FS00:0x21:*********************** Log Started 2015-08-22T08:04:00Z ***********************
08:04:00:WU01:FS00:0x21:Project: 9704 (Run 30, Clone 10, Gen 3)
08:04:00:WU01:FS00:0x21:Unit: 0x00000005ab404162553ec05b8c152f76
08:04:00:WU01:FS00:0x21:CPU: 0x00000000000000000000000000000000
08:04:00:WU01:FS00:0x21:Machine: 0
08:04:00:WU01:FS00:0x21:Reading tar file core.xml
08:04:00:WU01:FS00:0x21:Reading tar file system.xml
08:04:02:WU01:FS00:0x21:Reading tar file integrator.xml
08:04:02:WU01:FS00:0x21:Reading tar file state.xml
08:04:04:WU01:FS00:0x21:Digital signatures verified
08:04:04:WU01:FS00:0x21:Folding@home GPU Core21 Folding@home Core
08:04:04:WU01:FS00:0x21:Version 0.0.11
.........
08:06:17:WU00:FS00:Upload 94.55%
08:06:26:WU00:FS00:Upload complete
08:06:26:WU00:FS00:Server responded WORK_ACK (400)
08:06:26:WU00:FS00:Final credit estimate, 20848.00 points
08:06:26:WU00:FS00:Cleaning up
08:07:55:WU01:FS00:0x21:Completed 0 out of 640000 steps (0%)
08:07:55:WU01:FS00:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
08:17:50:WU01:FS00:0x21:Completed 6400 out of 640000 steps (1%)
08:27:06:WU01:FS00:0x21:Completed 12800 out of 640000 steps (2%)
08:36:24:WU01:FS00:0x21:Completed 19200 out of 640000 steps (3%)
...............................
21:25:23:WU01:FS00:0x21:Completed 544000 out of 640000 steps (85%)
21:34:45:WU01:FS00:0x21:Completed 550400 out of 640000 steps (86%)
21:44:06:WU01:FS00:0x21:Completed 556800 out of 640000 steps (87%)
21:54:09:WU01:FS00:0x21:Completed 563200 out of 640000 steps (88%)
22:00:10:WARNING:WU01:FS00:FahCore returned an unknown error code which probably indicates that it crashed
22:00:10:WARNING:WU01:FS00:FahCore returned: UNKNOWN_ENUM (127 = 0x7f)
22:00:11:WU01:FS00:Starting
22:00:11:WU01:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Arto/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 01 -suffix 01 -version 704 -lifeline 1044 -checkpoint 9 -gpu 0 -gpu-vendor nvidia
22:00:11:WU01:FS00:Started FahCore on PID 204
22:00:11:WU01:FS00:Core PID:3412
22:00:11:WU01:FS00:FahCore 0x21 started
22:00:12:WU01:FS00:0x21:*********************** Log Started 2015-08-22T22:00:12Z ***********************
22:00:12:WU01:FS00:0x21:Project: 9704 (Run 30, Clone 10, Gen 3)
22:00:12:WU01:FS00:0x21:Unit: 0x00000005ab404162553ec05b8c152f76
22:00:12:WU01:FS00:0x21:CPU: 0x00000000000000000000000000000000
22:00:12:WU01:FS00:0x21:Machine: 0
22:00:12:WU01:FS00:0x21:Digital signatures verified
22:00:12:WU01:FS00:0x21:Folding@home GPU Core21 Folding@home Core
22:00:12:WU01:FS00:0x21:Version 0.0.11
22:00:12:WU01:FS00:0x21:  Found a checkpoint file
22:02:50:WU01:FS00:0x21:ERROR:exception: Error uploading array oldPositions: clEnqueueWriteBuffer (-5)
22:02:51:WU01:FS00:0x21:Saving result file logfile_01.txt
22:02:52:WU01:FS00:0x21:Saving result file log.txt
22:03:02:WU01:FS00:0x21:Folding@home Core Shutdown: BAD_WORK_UNIT
22:03:10:WARNING:WU01:FS00:FahCore returned: BAD_WORK_UNIT (114 = 0x72)
22:03:11:WU01:FS00:Sending unit results: id:01 state:SEND error:FAULTY project:9704 run:30 clone:10 gen:3 core:0x21 unit:0x00000005ab404162553ec05b8c152f76
22:03:11:WU01:FS00:Uploading 3.20KiB to 171.64.65.98
22:03:11:WU01:FS00:Connecting to 171.64.65.98:8080
22:03:12:WU00:FS00:Connecting to 171.67.108.200:80
22:03:13:WU01:FS00:Upload complete
22:03:14:WU01:FS00:Server responded WORK_ACK (400)
22:03:15:WU01:FS00:Cleaning up
22:03:18:WU00:FS00:Assigned to work server 171.64.65.93
22:03:18:WU00:FS00:Requesting new work unit for slot 00: READY gpu:0:GK106 [GeForce GTX 650 Ti] from 171.64.65.93

[7im]

IIRC, the Kepler slowdown (needing the 327.xx drivers) was fixed in the previous version of the fahcores.