Help with GPU option

[zumbertinho]

Hi, I just started folding this week. I leave the folding on 'full' with just my cpu as a slot all the time and it's almost impossible to notice. I tried to use my GPU, and had to leave it 'on idle' because it interfered too much when I used the computer, but even when I left it on full and didn't use my computer the GPU slot would barely make it to the expiration date (not the timeout, the expiration), and because I actually USE the computer I knew it wouldn't make it to the deadline, so I had to turn it off. I saw some people talking about an option called max-packet-size and I tried to set my GPU as a slot with max-packet-size = small, but it just didn't download any WU, it just keeps retrying on and on. Does someone know if I'm doing something wrong, or maybe if this option broken (or maybe if it doesn't work the way I thought it would)? I really wanted to be able to use my GPU to do the folding too. It's a very modest video card, but I just feel like it's a waste of processing power that I could be using for this in its idle time.

Can someone help me?

Thanks

[Joe_H]

Welcome to the folding forum.

Important pieces of information are which OS you are using and what GPU you have installed. You can provide that and information about your folding configuration by posting the first hundred lines or so of your log file as described under this topic - viewtopic.php?f=61&t=26036.

[zumbertinho]

I'm assuming these lines are sufficient:

Code: Select all

*********************** Log Started 2015-03-29T11:08:47Z ***********************
11:08:47:************************* Folding@home Client *************************
11:08:47:      Website: http://folding.stanford.edu/
11:08:47:    Copyright: (c) 2009-2014 Stanford University
11:08:47:       Author: Joseph Coffland <joseph@cauldrondevelopment.com>
11:08:47:         Args: --open-web-control
11:08:47:       Config: C:/Users/Humberto/AppData/Roaming/FAHClient/config.xml
11:08:47:******************************** Build ********************************
11:08:47:      Version: 7.4.4
11:08:47:         Date: Mar 4 2014
11:08:47:         Time: 20:26:54
11:08:47:      SVN Rev: 4130
11:08:47:       Branch: fah/trunk/client
11:08:47:     Compiler: Intel(R) C++ MSVC 1500 mode 1200
11:08:47:      Options: /TP /nologo /EHa /Qdiag-disable:4297,4103,1786,279 /Ox -arch:SSE
11:08:47:               /QaxSSE2,SSE3,SSSE3,SSE4.1,SSE4.2 /Qopenmp /Qrestrict /MT /Qmkl
11:08:47:     Platform: win32 XP
11:08:47:         Bits: 32
11:08:47:         Mode: Release
11:08:47:******************************* System ********************************
11:08:47:          CPU: Intel(R) Core(TM) i5 CPU 650 @ 3.20GHz
11:08:47:       CPU ID: GenuineIntel Family 6 Model 37 Stepping 5
11:08:47:         CPUs: 4
11:08:47:       Memory: 7.93GiB
11:08:47:  Free Memory: 5.69GiB
11:08:47:      Threads: WINDOWS_THREADS
11:08:47:   OS Version: 6.1
11:08:47:  Has Battery: false
11:08:47:   On Battery: false
11:08:47:   UTC Offset: -3
11:08:47:          PID: 4276
11:08:47:          CWD: C:/Users/Humberto/AppData/Roaming/FAHClient
11:08:47:           OS: Windows 7 Ultimate
11:08:47:      OS Arch: AMD64
11:08:47:         GPUs: 1
11:08:47:        GPU 0: ATI:4 Caicos [AMD RADEON HD 6450]
11:08:47:         CUDA: Not detected
11:08:47:Win32 Service: false
11:08:47:***********************************************************************
11:08:47:<config>
11:08:47:  <!-- Network -->
11:08:47:  <proxy v=':8080'/>
11:08:47:
11:08:47:  <!-- Slot Control -->
11:08:47:  <power v='FULL'/>
11:08:47:
11:08:47:  <!-- User Information -->
11:08:47:  <passkey v='********************************'/>
11:08:47:  <team v='86565'/>
11:08:47:  <user v='zumbertinho'/>
11:08:47:
11:08:47:  <!-- Folding Slots -->
11:08:47:  <slot id='0' type='CPU'/>
11:08:47:</config>

As I said, my video card is incredible modest, it doesn't even have a direct connection to the power supply. Even so, I would really like to put it to some use, if possible.

[Joe_H]

I checked, and the server is currently listed in standby mode for the two older projects for AMD-ATI GPU's that would get assigned to card by using the max-packet-size=small option. That may mean those projects are done as they were announced as going into end of life about a year and a half ago.

Using the same option with lower end nVidia cards does result in an assignment of an older project, long deadline WU as there are 8 projects still active.