As the Tile states, I have 3 GPUs How do I set GPUGRID to use only One of them and allow the other Two GPUs for other usages. Thanks

One computer gf of two cards, and other PC Another one of the most
This is the easiest solution to the problem.

As the Tile states, I have 3 GPUs How do I set GPUGRID to use only One of them and allow the other Two GPUs for other usages. Thanks

This is fairly easy to configure but do you mean any one of the three or do you want to dedicate a specific card? If no one posts the details I will when I get home in a couple of hours.

---

Thanks - Steve

As the Tile states, I have 3 GPUs How do I set GPUGRID to use only One of them and allow the other Two GPUs for other usages. Thanks

I think you can put a file like this in your c:\programdata\Boinc\projects\gpugrid folder:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
<exclude_gpu>
<url>http://www.gpugrid.net/</url>
<device_num>1</device_num>
</exclude_gpu>
<exclude_gpu>
<url>http://www.gpugrid.net/</url>
<device_num>2</device_num>
</exclude_gpu>
</options>
</cc_config>

This file will only let your first gpu, gpu zero, crunch for gpugrid. To make a different gpu crunch just change the number in the different exclude sections.
You WILL need to put a similar file in each of your other project folders so gpu zero can crunch here while a different gpu can crunch someplace else. Since you use Windows ONLY use Notepad to create the file, saving it as a txt type file with the name cc_config.xml

The http part comes from each projects home page where it tells you what to type in to crunch for the project when you first attach to it.

Right. You will want to configure the BOINC client with a customized cc_config.xml file, per the documentation, found here:
http://boinc.berkeley.edu/wiki/Client_configuration

I too have 3 GPUs, and I only allow certain projects on each of them. Below is my cc_config.xml file, that you may look at as a reference. As you'll see, exclusion blocks can even be commented out, which means they are not in-place, but you can easily switch them back on later.

Regards,
Jacob Klein

<cc_config>
	<log_flags>
		<!-- The 3 flags that are on by default are: file_xfer, sched_ops, task -->

		<file_xfer>1</file_xfer>
		<file_xfer_debug>0</file_xfer_debug>

		<sched_ops>1</sched_ops>
		<sched_op_debug>0</sched_op_debug>

		<task>1</task>
		<task_debug>0</task_debug>

		<unparsed_xml>1</unparsed_xml>

		<work_fetch_debug>0</work_fetch_debug>
		<rr_simulation>0</rr_simulation>
		<rrsim_detail>0</rrsim_detail>

		<cpu_sched>0</cpu_sched>
		<cpu_sched_debug>0</cpu_sched_debug>
		<cpu_sched_status>0</cpu_sched_status>
		<coproc_debug>0</coproc_debug>

		<mem_usage_debug>0</mem_usage_debug>
		<checkpoint_debug>1</checkpoint_debug>

		<http_debug>0</http_debug>
		<http_xfer_debug>0</http_xfer_debug>
		<network_status_debug>0</network_status_debug>

		<scrsave_debug>1</scrsave_debug>
		<notice_debug>0</notice_debug>

		<app_msg_receive>0</app_msg_receive>
		<app_msg_send>0</app_msg_send>
		<async_file_debug>0</async_file_debug>
		<benchmark_debug>0</benchmark_debug>
		<dcf_debug>0</dcf_debug>
		<disk_usage_debug>0</disk_usage_debug>
		<priority_debug>0</priority_debug>
		<gui_rpc_debug>0</gui_rpc_debug>
		<heartbeat_debug>0</heartbeat_debug>
		<poll_debug>0</poll_debug>
		<proxy_debug>0</proxy_debug>
		<slot_debug>0</slot_debug>
		<state_debug>0</state_debug>
		<statefile_debug>0</statefile_debug>
		<suspend_debug>0</suspend_debug>
		<time_debug>0</time_debug>
		<trickle_debug>0</trickle_debug>
	</log_flags>
	<options>
		<!-- =================================================== TESTING OPTIONS =================================================== -->
<!--
		<start_delay>20</start_delay>
		<ncpus>12</ncpus>
		<exclusive_app>NotepadTest01.exe</exclusive_app>
		<exclusive_gpu_app>NotepadTest02.exe</exclusive_gpu_app>
-->

		<!-- =================================================== REGULAR OPTIONS =================================================== -->
		<report_results_immediately>0</report_results_immediately>
		<fetch_on_update>0</fetch_on_update>
		<max_event_log_lines>4000</max_event_log_lines>

		<max_file_xfers>10</max_file_xfers>
		<max_file_xfers_per_project>4</max_file_xfers_per_project>

		<exclusive_app>iRacingSim.exe</exclusive_app>
		<exclusive_app>iRacingSim64.exe</exclusive_app>
		<exclusive_app>Aces.exe</exclusive_app>
		<exclusive_app>TmForever.exe</exclusive_app>
		<exclusive_app>TmForeverLauncher.exe</exclusive_app>

		<!-- ===================================================== SETUP GPUS ====================================================== -->
		<use_all_gpus>1</use_all_gpus>

		<!-- =========================================== SETUP GPU 0: GeForce GTX 660 Ti =========================================== -->
<!--
		<ignore_nvidia_dev>0</ignore_nvidia_dev>
-->

		<!-- Exclude World Community Grid's "Help Conquer Cancer" GPU app (hcc1) on main display - makes graphics slow, even on 660 Ti -->
		<!-- Commenting out, for now, since this round of hcc1 is completed, and next round may not exhibit the issue. -->
<!--
		<exclude_gpu>
			<url>http://www.worldcommunitygrid.org</url>
			<device_num>0</device_num>
			<app>hcc1</app>
		</exclude_gpu>
-->

		<!-- Exclude several projects, since work from other GPU projects should give enough work to keep this GPU busy. -->
		<!-- Commenting out, because POEM is often out of work, and GPUGrid sometimes does run out. -->
<!--
		<exclude_gpu>
			<url>http://einstein.phys.uwm.edu/</url>
			<device_num>0</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://albert.phys.uwm.edu/</url>
			<device_num>0</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://setiathome.berkeley.edu/</url>
			<device_num>0</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://setiweb.ssl.berkeley.edu/beta/</url>
			<device_num>0</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://milkyway.cs.rpi.edu/milkyway/</url>
			<device_num>0</device_num>
		</exclude_gpu>
-->

		<!-- =========================================== SETUP GPU 1: GeForce GTX 460 =========================================== -->
<!--
		<ignore_nvidia_dev>1</ignore_nvidia_dev>
-->

		<!-- Exclude POEM's "POEM++ OpenCL version" GPU app (poemcl) from a second heterogeneous GPU, since it does not work properly -->
		<!-- Note: Although 320.18 drivers successfully run smalltest_3, the drivers still do not work right with POEM. -->
		<!-- Note: Also, it appears that running POEM only on the GTX 460, does not work. So, it must run on the GTX 660 Ti! -->
		<!-- Note: Tested their new OpenCL application on 3/22/2014 -- still does not start when running only on the GTX 460. So, it must run on the GTX 660 Ti! -->
		<exclude_gpu>
			<url>http://boinc.fzk.de/poem/</url>
			<device_num>1</device_num>
			<app>poemcl</app>
		</exclude_gpu>

		<!-- Reminder: For GPUGrid.net, if going to run 2-tasks-on-1-GPU, exclude this GPU (it only has 1 GB memory) -->
		<!-- Commenting out, decided to include this GPU and run 1 task per GPU. -->
<!--
		<exclude_gpu>
			<url>http://www.gpugrid.net</url>
			<device_num>1</device_num>
		</exclude_gpu>
-->

		<!-- Exclude several projects, since work from other GPU projects should give enough work to keep this GPU busy. -->
		<!-- Commenting out, because POEM is often out of work, and GPUGrid sometimes does run out. -->
<!--
		<exclude_gpu>
			<url>http://einstein.phys.uwm.edu/</url>
			<device_num>1</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://albert.phys.uwm.edu/</url>
			<device_num>1</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://setiathome.berkeley.edu/</url>
			<device_num>1</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://setiweb.ssl.berkeley.edu/beta/</url>
			<device_num>1</device_num>
		</exclude_gpu>
		<exclude_gpu>
			<url>http://milkyway.cs.rpi.edu/milkyway/</url>
			<device_num>1</device_num>
		</exclude_gpu>
-->

		<!-- =========================================== SETUP GPU 2: GeForce GTS 240 =========================================== -->
<!--
		<ignore_nvidia_dev>2</ignore_nvidia_dev>
-->

		<!-- Exclude World Community Grid's Help Conquer Cancer GPU app -->
		<!-- GPU not supported per https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=GPU#610 -->
		<exclude_gpu>
			<url>http://www.worldcommunitygrid.org</url>
			<device_num>2</device_num>
			<app>hcc1</app>
		</exclude_gpu>

		<!-- Exclude POEM's "POEM++ OpenCL version" GPU app (poemcl) from a second heterogeneous GPU, since it does not work properly -->
		<!-- Also, GPU is not supported, as all tasks immediately error out -->
		<exclude_gpu>
			<url>http://boinc.fzk.de/poem/</url>
			<device_num>2</device_num>
			<app>poemcl</app>
		</exclude_gpu>

		<!-- Exclude GPUGrid.net -->
		<!-- GPU not supported per http://www.gpugrid.net/forum_thread.php?id=2507 -->
		<exclude_gpu>
			<url>http://www.gpugrid.net/</url>
			<device_num>2</device_num>
		</exclude_gpu>

		<!-- Exclude Milkyway@Home -->
		<!-- GPU not supported, as all tasks immediately error out -->
		<exclude_gpu>
			<url>http://milkyway.cs.rpi.edu/milkyway/</url>
			<device_num>2</device_num>
		</exclude_gpu>

	</options>
</cc_config>

I think you can put a file like this in your c:\programdata\Boinc\projects\gpugrid folder:

I believe the cc_config.xml file should go in the c:\programdata\Boinc folder.

If you were using an app_info.xml file specific for GPUGrid that would go in the c:\programdata\Boinc\projects\gpugrid folder.

I think you can put a file like this in your c:\programdata\Boinc\projects\gpugrid folder:

I believe the cc_config.xml file should go in the c:\programdata\Boinc folder.

If you were using an app_info.xml file specific for GPUGrid that would go in the c:\programdata\Boinc\projects\gpugrid folder.

If you put the cc_config file in the folder you said you need to add more lines to the file, excluding all the other gpu's for all the projects that each one crunches for. Meaning something like:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
<exclude_gpu>
<url>http://www.gpugrid.net/</url>
<device_num>1</device_num>
</exclude_gpu>
<exclude_gpu>
<url>http://www.gpugrid.net/</url>
<device_num>2</device_num>
</exclude_gpu>
<url>http://http://moowrap.net//</url>
<device_num>o</device_num>
</exclude_gpu>
<url>http://http://moowrap.net//</url>
<device_num>2</device_num>
</exclude_gpu>
<url>http://http://asteroidsathome.net/boinc///</url>
<device_num>0</device_num>
</exclude_gpu>
<url>http://http://asteroidsathome.net/boinc///</url>
<device_num>1</device_num>
</exclude_gpu>
</options>
</cc_config>

The above means gpu zero would crunch for GpuGrid only, while gpu 1 would crunch for Moo only and gpu 2 would crunch for Asteroids only.

Mikey:

Your xml is missing several open tags, and is invalid.
Also, if you happen to have spaces or tabs in it, and you want to preserve the formatting in a post (such as I have done above), you can simply encapsulate it in a

[pre]

tag.

I still think that the xml I pasted earlier, is a great reference.

Mikey:

Your xml is missing several open tags, and is invalid.
Also, if you happen to have spaces or tabs in it, and you want to preserve the formatting in a post (such as I have done above), you can simply encapsulate it in a

[pre]

tag.

I still think that the xml I pasted earlier, is a great reference.

I just learned something today, I bow to your experience and expertise!!

:) Experience, sure, but not expertise. I typically find out about features from a lot of trial and error, but I usually brute-force that trialing until I know exactly how something works. PS: Someone introduced me to the

[pre]

"pre-formatted text" command only recently (within the last year I think), so... we're all learning.

PS: <exclude_gpu> exists because of me. I was having a hell of a time trying to get work to go where it should, with my system that has 3 differing generations of GPUs in it, GTX 660 Ti, GTX 460, GTS 240. So, the BOINC dev team hooked up <exclude_gpu> options, and I tested it as best I could. It works quite well, although using it does slightly interfere with maintaining buffer levels in the work fetch logic.

This Project & Forum is the Greatest, Thanks for all your inputs.
I now have a pair of R9 290X to start testing on GPUGRID.

Not sure if serious or.............

GPUGrid has NVIDIA Cuda tasks, and CPU tasks (that work on PC and Android).
Your GPUs are AMD Radeons, right? If so, they won't work here, but if you attach to some other project that has AMD tasks, they will work there.

... Unless you meant "testing" in the sense of "getting work for a third GPU that is NVIDIA, while not getting work from the AMD GPUs", in which case, you really shouldn't have to do anything special to make that happen. BOINC knows your resource types, and BOINC knows which projects can do which resource types. Hope that makes sense.

Off Topic (OT):
Regarding Bulletin Board Code (BBCode), different projects have different versions of Boinc server software installed and AFAIK projects can select what to enable and disable. It's definitely the case that some projects use a limited range of codes whereas other projects facilitate greater use. When this site was overhauled a couple of years ago a lot of things were fixed including the BBCode functions. However, some commands such as strike-through are not enabled,

strikethrough text

---

FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Ok I need some assist. A slight background on me. I came over from Folding@Home, so I am new to crunching, and so far I don't understand how folks can say that BOINC is easier to use than FAH manager. It's much more intensive.

As I posted over here: http://forums.evga.com/FindPost/2173190 I am trying to make one of my GPU's not show up to BOINC. I don't want it to crunch at all. I have tried setting up the config as you can see that post, but then after I looked up this thread over here, I edited the cc_config.xml file to look like this.

<cc_config>
<options>
<ignore_nvidia_dev>0</ignore_nvidia_dev>
</options>
</cc_config>

As I also stated over on that other post, I am not a coder. I am great with fixing computers and hardware, but I suck when it comes to command line, scripts, and or anything else that requires a basic knowledge of writing code. I just don't get it. I have tried, and after years of reading shit I don't understand, I still try for some stupid reason, but rarely it works out for me. So I am not being lazy here.

So with that said. I would like to ask for more precise help here. I want GPU 0 in my system to NOT run at all with BOINC. Could someone please fix my script so that I can get what I want?

Offhand, that XML looks good. Are you sure you have it saved to a file exactly named "cc_config.xml" in the ProgramData\Boinc folder? Please restart BOINC, and copy/paste the first ~25 starting lines from the BOINC Event Log. Paste them here so we can see.

Edit: You have to set use_all_gpus to 1 to get BOINC to see any GPU that isn't the primary GPU (device 0) I recommend setting it to 1, even if you're going to disable 1 of the GPUs.

From what I gathered in the thread you linked to, you're trying to set this up for a system with two GTX570s and you're trying to exclude GPU0 from BOINC. However, you are still seeing tasks being processed by GPU0 even with the cc_config in place.

I concur with Jacob that the XML you posted should work.

This may seem a bit basic, but as someone new to BOINC you may not have caught this yet: Have you checked to make sure that it is actually GPU0 that is doing the crunching? I've noticed in my systems where I have two identical cards that you can't trust the BOINC manager to tell you which card a task is actually running on. Just because it says "Running (0.855 CPUs + 1 Nividia GPU)(device 0)" next to a task does not mean it is actually running on GPU0. My mixed card systems always correctly reported the actual GPU in use in the manager, but with the twin card systems it seems BOINC doesn't report accurately.

Just thought I'd throw that out there as it is an odd thing about BOINC that I didn't catch myself for a long time.

Hope we've been of some help. Please keep posting with any other questions.

Ok here's the event log from a fresh start of BOINC.

Why does that first line say my config file is not found???

6/4/2014 00:08:44 | | cc_config.xml not found - using defaults
6/4/2014 00:08:44 | | Starting BOINC client version 7.2.42 for windows_x86_64
6/4/2014 00:08:44 | | log flags: file_xfer, sched_ops, task
6/4/2014 00:08:44 | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
6/4/2014 00:08:44 | | Data directory: K:\BOINC Work
6/4/2014 00:08:44 | | Running under account Phil Gaming
6/4/2014 00:08:44 | | CUDA: NVIDIA GPU 0 (not used): GeForce GTX 570 (driver version 337.88, CUDA version 6.0, compute capability 2.0, 1280MB, 522MB available, 1425 GFLOPS peak)
6/4/2014 00:08:44 | | CUDA: NVIDIA GPU 1: GeForce GTX 570 (driver version 337.88, CUDA version 6.0, compute capability 2.0, 1280MB, 1182MB available, 1425 GFLOPS peak)
6/4/2014 00:08:44 | | OpenCL: NVIDIA GPU 0 (ignored by config): GeForce GTX 570 (driver version 337.88, device version OpenCL 1.1 CUDA, 1280MB, 522MB available, 1425 GFLOPS peak)
6/4/2014 00:08:44 | | OpenCL: NVIDIA GPU 1: GeForce GTX 570 (driver version 337.88, device version OpenCL 1.1 CUDA, 1280MB, 1182MB available, 1425 GFLOPS peak)
6/4/2014 00:08:44 | | Host name: PhilGaming-PC
6/4/2014 00:08:44 | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7 CPU 960 @ 3.20GHz [Family 6 Model 26 Stepping 5]
6/4/2014 00:08:44 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt syscall nx lm vmx tm2 pbe
6/4/2014 00:08:44 | | OS: Microsoft Windows 7: Home Premium x64 Edition, Service Pack 1, (06.01.7601.00)
6/4/2014 00:08:44 | | Memory: 11.99 GB physical, 23.98 GB virtual
6/4/2014 00:08:44 | | Disk: 148.95 GB total, 134.18 GB free
6/4/2014 00:08:44 | | Local time is UTC -5 hours
6/4/2014 00:08:44 | Asteroids@home | URL http://asteroidsathome.net/boinc/; Computer ID 92548; resource share 100
6/4/2014 00:08:44 | Milkyway@Home | URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 577447; resource share 100
6/4/2014 00:08:44 | pogs | URL http://pogs.theskynet.org/pogs/; Computer ID 39322; resource share 100
6/4/2014 00:08:44 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7298565; resource share 100
6/4/2014 00:08:44 | GPUGRID | URL http://www.gpugrid.net/; Computer ID 175670; resource share 100
6/4/2014 00:08:44 | World Community Grid | URL http://www.worldcommunitygrid.org/; Computer ID 3061693; resource share 100
6/4/2014 00:08:44 | | General prefs: from http://bam.boincstats.com/ (last modified 28-May-2014 15:57:05)
6/4/2014 00:08:44 | | Host location: none
6/4/2014 00:08:44 | | General prefs: using your defaults
6/4/2014 00:08:44 | | Reading preferences override file


After I took this report, I went and added the use all gpus to the config and now it doesnt name a GPU to a task, it just says 0.394 CPUs + 1 NVIDIA GPU. So I think problem solved.. However it still says cc_config.xml not found. Did I put it in the wrong place? I have it in both the start folder and the work folder. It must be seeing it somewhere though, cause it's showing them ignored. Weird

Ok here's the event log from a fresh start of BOINC.

Why does that first line say my config file is not found???

6/4/2014 00:08:44 | | cc_config.xml not found - using defaults

After I took this report, I went and added the use all gpus to the config and now it doesnt name a GPU to a task, it just says 0.394 CPUs + 1 NVIDIA GPU. So I think problem solved.. However it still says cc_config.xml not found. Did I put it in the wrong place? I have it in both the start folder and the work folder. It must be seeing it somewhere though, cause it's showing them ignored. Weird

The cc_config.xml file goes in the folder c:\program data\boinc if you let boinc use the default installation locations. The 'program data' folder is a hidden one and you must unhide it to put files in it. To do that just press the alt key while you are viewing the C: folder, when the new menu pops up choose tools, folder options, view and check the box to show hidden files and folders, click okay and then you can drop the file in the right folder.

It's showing 1 GPU ignored, I think, because the default BOINC settings are to only use the main GPU.

But let's solve your problem.

According to your own logs there, the data directory is "K:\BOINC Work", so the "cc_config.xml" file must be placed in that folder. Turn on the Windows folder option to show file extensions, to make absolutely sure it isn't named "cc_config.xml.txt". If you need to rename it, then do so, then restart BOINC and let us know if it finds it.