More trouble with Nvidia GTX 460

[fredex]

Just installed a new-to-me GTX 460, hoping I could use it to do some folding (it replaces a 9800-GT which apparently is no longer supported by FAH).

I added a GPU slot and it starts up and keeps getting the errors shown below (see code box).

Having just done some reading here on the forums, I see that as of a few months ago people were saying that to work with the 460, we should be using a now very old Nvidia driver, e.g. 331 or maybe older. Is that still the case? Might my problems stem from my having used the latest?

Code: Select all

*********************** Log Started 2015-01-09T01:55:57Z ***********************
01:57:25:WU00:FS01:Connecting to assign-GPU.stanford.edu:80
01:57:25:WU00:FS01:News: 
01:57:25:WU00:FS01:Assigned to work server 140.163.4.231
01:57:25:WU00:FS01:Requesting new work unit for slot 01: READY gpu:0:GF104 [GeForce GTX 460] from 140.163.4.231
01:57:25:WU00:FS01:Connecting to 140.163.4.231:8080
01:57:26:WU00:FS01:Downloading 4.83MiB
01:57:28:WU00:FS01:Download complete
01:57:28:WU00:FS01:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:13001 run:177 clone:4 gen:106 core:0x17 unit:0x000000aa538b3db7532881ecc874248a
01:57:28:WU00:FS01:Downloading core from http://web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah
01:57:28:WU00:FS01:Connecting to web.stanford.edu:80
01:57:28:WU00:FS01:FahCore 17: Downloading 3.01MiB
01:57:33:WU00:FS01:FahCore 17: Download complete
01:57:33:WU00:FS01:Valid core signature
01:57:33:WARNING:WU00:FS01:FahCore has not changed since last download, aborting core update
01:57:33:WU00:FS01:Starting
01:57:33:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
01:57:33:WU00:FS01:Started FahCore on PID 4050
01:57:33:WU00:FS01:Core PID:4054
01:57:33:WU00:FS01:FahCore 0x17 started
01:57:34:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
01:57:34:WU00:FS01:Starting
01:57:34:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
01:57:34:WU00:FS01:Started FahCore on PID 4056
01:57:34:WU00:FS01:Core PID:4060
01:57:34:WU00:FS01:FahCore 0x17 started
01:57:34:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
01:58:34:WU00:FS01:Starting
01:58:34:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
01:58:34:WU00:FS01:Started FahCore on PID 4070
01:58:34:WU00:FS01:Core PID:4074
01:58:34:WU00:FS01:FahCore 0x17 started
01:58:34:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
01:59:34:WU00:FS01:Starting
01:59:34:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
01:59:34:WU00:FS01:Started FahCore on PID 4085
01:59:34:WU00:FS01:Core PID:4089
01:59:34:WU00:FS01:FahCore 0x17 started
01:59:34:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:00:34:WU00:FS01:Starting
02:00:34:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:00:34:WU00:FS01:Started FahCore on PID 4184
02:00:34:WU00:FS01:Core PID:4188
02:00:34:WU00:FS01:FahCore 0x17 started
02:00:34:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:00:34:WARNING:WU00:FS01:Too many errors, failing
02:00:35:WU00:FS01:Sending unit results: id:00 state:SEND error:FAILED project:13001 run:177 clone:4 gen:106 core:0x17 unit:0x000000aa538b3db7532881ecc874248a
02:00:35:WU00:FS01:Connecting to 140.163.4.231:8080
02:00:35:WU02:FS01:Connecting to assign-GPU.stanford.edu:80
02:00:35:WU00:FS01:Server responded WORK_ACK (400)
02:00:35:WU00:FS01:Cleaning up
02:00:36:WU02:FS01:News: 
02:00:36:WU02:FS01:Assigned to work server 140.163.4.231
02:00:36:WU02:FS01:Requesting new work unit for slot 01: READY gpu:0:GF104 [GeForce GTX 460] from 140.163.4.231
02:00:36:WU02:FS01:Connecting to 140.163.4.231:8080
02:00:36:WU02:FS01:Downloading 4.84MiB
02:00:38:WU02:FS01:Download complete
02:00:38:WU02:FS01:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:13001 run:467 clone:6 gen:83 core:0x17 unit:0x0000008d538b3db7532c80a033c52bcc
02:00:38:WU02:FS01:Starting
02:00:38:WU02:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 02 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:00:38:WU02:FS01:Started FahCore on PID 4189
02:00:38:WU02:FS01:Core PID:4193
02:00:38:WU02:FS01:FahCore 0x17 started
02:00:38:WARNING:WU02:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:00:39:WU02:FS01:Starting
02:00:39:WU02:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 02 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:00:39:WU02:FS01:Started FahCore on PID 4194
02:00:39:WU02:FS01:Core PID:4198
02:00:39:WU02:FS01:FahCore 0x17 started
02:00:39:WARNING:WU02:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:01:39:WU02:FS01:Starting
02:01:39:WU02:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 02 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:01:39:WU02:FS01:Started FahCore on PID 4224
02:01:39:WU02:FS01:Core PID:4228
02:01:39:WU02:FS01:FahCore 0x17 started
02:01:39:WARNING:WU02:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:02:39:WU02:FS01:Starting
02:02:39:WU02:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 02 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:02:39:WU02:FS01:Started FahCore on PID 4239
02:02:39:WU02:FS01:Core PID:4243
02:02:39:WU02:FS01:FahCore 0x17 started
02:02:39:WARNING:WU02:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:03:39:WU02:FS01:Starting
02:03:39:WU02:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 02 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:03:39:WU02:FS01:Started FahCore on PID 4255
02:03:39:WU02:FS01:Core PID:4259
02:03:39:WU02:FS01:FahCore 0x17 started
02:03:39:WARNING:WU02:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:03:39:WARNING:WU02:FS01:Too many errors, failing
02:03:39:WU02:FS01:Sending unit results: id:02 state:SEND error:FAILED project:13001 run:467 clone:6 gen:83 core:0x17 unit:0x0000008d538b3db7532c80a033c52bcc
02:03:39:WU02:FS01:Connecting to 140.163.4.231:8080
02:03:40:WU00:FS01:Connecting to assign-GPU.stanford.edu:80
02:03:40:WU02:FS01:Server responded WORK_ACK (400)
02:03:40:WU02:FS01:Cleaning up
02:03:40:WU00:FS01:News: 
02:03:40:WU00:FS01:Assigned to work server 140.163.4.231
02:03:40:WU00:FS01:Requesting new work unit for slot 01: READY gpu:0:GF104 [GeForce GTX 460] from 140.163.4.231
02:03:40:WU00:FS01:Connecting to 140.163.4.231:8080
02:03:41:WU00:FS01:Downloading 4.84MiB
02:03:43:WU00:FS01:Download complete
02:03:43:WU00:FS01:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:13000 run:641 clone:0 gen:74 core:0x17 unit:0x00000094538b3db753105093bbfe666a
02:03:43:WU00:FS01:Starting
02:03:43:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:03:43:WU00:FS01:Started FahCore on PID 4260
02:03:43:WU00:FS01:Core PID:4264
02:03:43:WU00:FS01:FahCore 0x17 started
02:03:43:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:03:44:WU00:FS01:Starting
02:03:44:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:03:44:WU00:FS01:Started FahCore on PID 4265
02:03:44:WU00:FS01:Core PID:4269
02:03:44:WU00:FS01:FahCore 0x17 started
02:03:44:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:04:44:WU00:FS01:Starting
02:04:44:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:04:44:WU00:FS01:Started FahCore on PID 4283
02:04:44:WU00:FS01:Core PID:4287
02:04:44:WU00:FS01:FahCore 0x17 started
02:04:44:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)
02:05:44:WU00:FS01:Starting
02:05:44:WU00:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/web.stanford.edu/~pande/Linux/AMD64/NVIDIA/Fermi/Core_17.fah/FahCore_17 -dir 00 -suffix 01 -version 703 -lifeline 3154 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
02:05:44:WU00:FS01:Started FahCore on PID 4298
02:05:44:WU00:FS01:Core PID:4302
02:05:44:WU00:FS01:FahCore 0x17 started
02:05:44:WARNING:WU00:FS01:FahCore returned: FAILED_2 (1 = 0x1)

[artoar_11]

I'm not sure if it is the driver. I use NV driver - 327.23 (or 314.22). You can try the old client Version 6.41r2 (Core_15). Works very well.

p13001 - 18,500 PPD (@800MHz, V7)
p7620-7627 - 22,000 PPD (@800MHz, V6.41r2)

[Joe_H]

Did you add the GPU slot to an existing install of the folding client that used to have a slot set up for the 9800GT? If so there may be some setting information left over from the older card in the client. Probably the best way to handle that is finish any CPU WU that you are processing and uninstall the client including the data, and reinstall from scratch.

As for the driver version, that shouldn't be causing the kind of errors shown in your log. But, many have reported that drivers newer than 327 result in a reduction in folding speed of as much as an half on older nVidia cards such as your 460.

[fredex]

Joe_H: I had, in the past, set up a slot for the 9800GT, but it never worked so I'd removed it. you're suggesting leftover cruft from that may be interfering?

artoar_11: I'm currently using the drivers from the elrepo linux repository. they offer a 325 and a 331 and a 340, but no 327. The 325 doesn't let X start, complaining about an ABI change in X-11. the 331 version is alleged to not give good performance so I haven't tried that.

I think I'll try Joe_H's suggestion and if that doesn't help will then look into finding a 327 driver.

thanks to you both!

[bruce]

I've had a lot of problems installing drivers in ways that doesn't let X start. Jesse_v has written a how-to which sounds like it might solve your problem. (I haven't installed drivers since it has been published, so I can't vouch for it, personally.)

Tips for installing Nvidia .run files in Linux