Posts by Beyond

1) Message boards : Wish list : Points nVidia / ATI (Message 5593)
Posted 27 Oct 2013 by Profile Beyond
Post:
It is, the NVidias are just slower...
2) Message boards : Number crunching : asus radeon 7790 extremly unstable (Message 5577)
Posted 25 Oct 2013 by Profile Beyond
Post:
just for curiosity i change drivers back to 13.4,no change still got the gpu and memory clocks bouncing like pinball machines,activity bounces as well,not steady.tried playing one of my graphic intense games,shows the same thing activity wise as well as the clocks,very unstable.so i think ive ruled out it being boinc.i tried varying the voltage at the bottom of ccc,+or- 2,didnt make any differance.ive googled this problem,wording it several differant ways,all im findig is alot of people having trouble with 7790 cards,and not a lot of answers.

"Clocks bouncing around", that seems to be normal in an app like Afterburner with the 7790. If it's producing results don't worry about it. It does the same thing in Collatz and DistrRTgen. It draws only about 4 watts more than a 7770 and is a lot faster. As far as 13.9 drivers, they don't seem to work correctly with my 7790 cards no matter what I do: driversweeper, you name it. Swap in any other card (like a 7850 or 7950) and everything's fine. The 13.4 drivers work fine here though with the 7790, and the 7790 runs at the same speed as the 7850 in Moo! Go figure. In Einstein the 7850 is considerably faster.

3) Message boards : Number crunching : 7970s Dont Run (Message 5534)
Posted 15 Oct 2013 by Profile Beyond
Post:
While two 7770 cards are working, I can't seem to get a 7850 to work. It just sits in low power (300MHz) mode and goes very slowly. Tried setting up 2d and 3d profiles in Afterburner to force it to run at speed but no go. Works fine on all other projects though.

Now Iv'e gotten 2 out of 3 7790 GPUs running. The 3rd one refuses to run Moo but runs everything else perfectly. Completion time for the 7790 is about 70% of the 7770. Still no luck on the 7850 cards.

Have you tried BOINC Client 7.2.16? Iam running with Catalyst 13.9 too but only one card. Dont know if this helps.

Thanks for the post. Just installed BOINC 7.2.23 and the Catalyst 13.9 drivers on that box and Moo still goes to 4% and stalls. Swapped a 7950 in and same thing, but the 7790 swapped into the other box now works on Moo. So it must be something in the software installs, but they should be the same. Also same MB and CPU, pretty much same hardware all around. All BOINC only boxes running the same projects. Moo is the only project that stalls. Any other ideas anyone?
4) Message boards : Number crunching : 7970s Dont Run (Message 5446)
Posted 19 Sep 2013 by Profile Beyond
Post:
While two 7770 cards are working, I can't seem to get a 7850 to work. It just sits in low power (300MHz) mode and goes very slowly. Tried setting up 2d and 3d profiles in Afterburner to force it to run at speed but no go. Works fine on all other projects though.

Now Iv'e gotten 2 out of 3 7790 GPUs running. The 3rd one refuses to run Moo but runs everything else perfectly. Completion time for the 7790 is about 70% of the 7770. Still no luck on the 7850 cards.
5) Message boards : Number crunching : 7970s Dont Run (Message 5383)
Posted 28 Aug 2013 by Profile Beyond
Post:
While two 7770 cards are working, I can't seem to get a 7850 to work. It just sits in low power (300MHz) mode and goes very slowly. Tried setting up 2d and 3d profiles in Afterburner to force it to run at speed but no go. Works fine on all other projects though.
6) Message boards : Number crunching : 7970s Dont Run (Message 5378)
Posted 27 Aug 2013 by Profile Beyond
Post:
I have now successful run it on HD7950 with 13.4 driver but against 5xxx/6xxx serie poor speed. It must have 30%-40% more speed increase. The client still needs some optimizations with 7xxx serie.

Get now working stream client.

Download

Download and copy files into project folder.

Thx for the infos below. Good luck!

Thanks for the link and info, got it running on 2 7770 cards. Wonder why Teemu isn't sending this app out automatically...
7) Message boards : Number crunching : Server problem (Message 3502)
Posted 9 Aug 2012 by Profile Beyond
Post:
Moo! Wrapper 08-09-12 08:01 Computation for task dnetc_r72_1344031569_12_768_0 finished
Moo! Wrapper 08-09-12 08:01 Starting task dnetc_r72_1344031567_13_823_0 using dnetc version 103 (ati14)
Moo! Wrapper 08-09-12 08:01 Started upload of dnetc_r72_1344031569_12_768_0_0
Moo! Wrapper 08-09-12 08:01 Finished upload of dnetc_r72_1344031569_12_768_0_0
Moo! Wrapper 08-09-12 08:15 update requested by user
Moo! Wrapper 08-09-12 08:15 Sending scheduler request: Requested by user.
Moo! Wrapper 08-09-12 08:15 Reporting 4 completed tasks, not requesting new tasks
Moo! Wrapper 08-09-12 08:15 Scheduler request completed
Moo! Wrapper 08-09-12 08:15 Server can't open database
8) Message boards : Number crunching : Moo does not like 7.X.X BOINC (Message 3400)
Posted 19 Jul 2012 by Profile Beyond
Post:
Indeed, but I could still add an advanced scheduler option that makes the scheduler to only ask for one resource even though the tasks will use all GPUs in the end.

Again, a very nice addition. Another useful addition for the advanced preferences would be to allow us to change the priority without having to edit the ini files.
9) Message boards : News : Network problems (Message 3380)
Posted 17 Jul 2012 by Profile Beyond
Post:
This is what I'm getting now:

Moo! Wrapper 07-17-12 08:09 Requesting new tasks for ATI
Moo! Wrapper 07-17-12 08:09 Scheduler request completed: got 0 new tasks
Moo! Wrapper 07-17-12 08:09 No tasks sent
Moo! Wrapper 07-17-12 08:09 Tasks for NVIDIA GPU are available, but your preferences are set to not accept them
Moo! Wrapper 07-17-12 08:09 Tasks for CPU are available, but your preferences are set to not accept them

Thought all the GPU tasks were the same?
10) Message boards : Number crunching : Moo does not like 7.X.X BOINC (Message 3377)
Posted 17 Jul 2012 by Profile Beyond
Post:
Indeed, but I could still add an advanced scheduler option that makes the scheduler to only ask for one resource even though the tasks will use all GPUs in the end. You just can't blame me then if you systems overheat or crawl to a stop due to overcommitting GPU resources. ;)

This would probably be easier than to make the full support for running one task per GPU but this would be first step anyway. (Such scheduler modification is needed anyway eventually.)

That would be a very useful addition. As far as overheating, they can't run any faster than 99% anyway, and TThrottle eliminates overheating worries altogether. Crawling to a stop? Haven't had that problem. The machines running the 1 ATI WUs on multiple GPUs along with POEM as outlined above are doing fine and using all the GPU finally too. This app_info trick (and hopefully soon a scheduler option) allows me to run Moo to a far greater extent than I could previously.
11) Message boards : Number crunching : Moo does not like 7.X.X BOINC (Message 3359)
Posted 13 Jul 2012 by Profile Beyond
Post:
Hello,

Hmm, if other projects are working and we are not then it must be us setting the "use all GPUs on the host for crunching one WU" switch. New BOINC Client must have some critical bugs regarding that setting where they think we never have enough resources to crunch. :(

I'll try to find some time to upgrade my host to latest stable BOINC Client and see what happens.

-w

Teemu, if you could give us a switch that will allow BOINC to think that it is going to use only 1 GPU it would really help. The dnet app still uses all GPUs but that way we can run it along with something like POEM that has low GPU usage and Moo will will use the rest of the GPU that's not utilized. I have one box with 2 5850 cards that does exactly that and it works well. Moo uses 1.5 GPUs and POEM uses .5. For some reason that box downloads Moo WUs with the settings of: 0.02C + 1 ATI GPU. This allows the above behavior to work. My other boxes always DL Moo WUs with 0.02C + 2 ATI GPU, don't know why there's a differences. But if we could force the 0.02C + 1 ATI GPU setting it would be great.

Edit: Never mind, finally figured out how to do it with an app_info. One box had the app_info and the others didn't, duh :)
12) Message boards : Number crunching : GPU usage bouncing from 99% to 0% (Message 2426)
Posted 28 Jan 2012 by Profile Beyond
Post:
Thanks!
I just set the fan control to manual while I crunch.
For some reason, even though the dips in GPU usage
are not long enough for the temps to drop much on the card
the fan dips right down when the gpu usage drops.

That's exactly why you would "Enable user defined software automatic fan control" and select the step method. You can set the steps to whatever you like and the fan will only change speed at the steps. Otherwise the fan speed changes continuously. Setting it to manual will not allow the speed to adjust if something happens to seriously escalate the GPU temps.
13) Message boards : Number crunching : GPU usage bouncing from 99% to 0% (Message 2421)
Posted 28 Jan 2012 by Profile Beyond
Post:
Double-click on the user defined fan control in Afterburner: settings/fan. This will make the graph a series of steps instead of a continuous control. Adjust the steps so that the fan doesn't constantly change speeds. Works very well here.
14) Message boards : Number crunching : App_info..xml (Message 2411)
Posted 26 Jan 2012 by Profile Beyond
Post:
The application number is shown inside the Stderr for completed WUs as:
Distributed.net Client v1.03 (ati14)
Try changing:
<version_num>130</version_num>
To:
<version_num>103</version_num>
(dnetc_wrapper_1.3_windows_intelx86__ati14.exe is correct, dont change that)
Regards Zy

Just to make things clearer, the first app_info.xml listed in this thread is all messed up, the second one much better. I modified things a bit and this is what I'm using (no api_version or gpu_ram statements, ncpu statements changed):

<app_info>
<app>
<name>dnetc</name>
<user_friendly_name>Distributed.net Client</user_friendly_name>
</app>
<file_info>
<name>dnetc_wrapper_1.3_windows_intelx86__ati14.exe</name>
<executable/>
</file_info>
<file_info>
<name>dnetc518-win32-x86-stream.exe</name>
<executable/>
</file_info>
<file_info>
<name>dnetc-gpu-1.3.ini</name>
</file_info>
<file_info>
<name>job-ati14-1.00.xml</name>
</file_info>
<app_version>
<app_name>dnetc</app_name>
<version_num>103</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>0.15</avg_ncpus>
<max_ncpus>1.0</max_ncpus>
<plan_class>ati14</plan_class>
<flops>1157115231469.729200</flops>
<file_ref>
<file_name>dnetc_wrapper_1.3_windows_intelx86__ati14.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>dnetc518-win32-x86-stream.exe</file_name>
<copy_file/>
</file_ref>
<file_ref>
<file_name>dnetc-gpu-1.3.ini</file_name>
<open_name>dnetc.ini</open_name>
<copy_file/>
</file_ref>
<file_ref>
<file_name>job-ati14-1.00.xml</file_name>
<open_name>job.xml</open_name>
<copy_file/>
</file_ref>
<coproc>
<type>ATI</type>
<count>0.5</count>
</coproc>
</app_version>
</app_info>

Like you say, about a 10% overall increase in production, GPU usage running at 97%+ and increased temps (3C-5C).
15) Message boards : Number crunching : 7970s Dont Run (Message 2292)
Posted 17 Jan 2012 by Profile Beyond
Post:
He very well may do .... Collatz has always been the premier sea of calm inside the BOINC maelstrom. Nice Project to go to if you want a respite to recharge the 'ol BOINC batteries before launching back into the tempest of BOINCLand :)

He runs a good Project, so guaranteed if he goes for one, it will be a solid good performer. He's a stand up guy who calls it how it is .... that defenitely works for me ... I detest the ego driven hype syndromes that fly around. RegardsZy

A note posted at Collatz:

I released the OS X app first because there was no AMD support for OS X. That being said, I actually created the app on Windows first because I prefer VS 2010 over XCode as an IDE. So yes, there is a Windows OpenCL version that runs on both nVidia and AMD. Performance is lousy compared to the Brook+/CAL app though. So, I'll have to come up with a way so that if the GPU supports CAL, the server will send it the ATI app and if not, will send the OpenCL app.
16) Message boards : Number crunching : 7970s Dont Run (Message 2288)
Posted 17 Jan 2012 by Profile Beyond
Post:
GPUGRID will I suspect come up with an AMD winner as they have been planning and aiming at the GNU application for many months.

years :)

I suspect Jon at Collatz will come up with an app very soon. He's always on top of things and just did an OpenCL app for the Mac.
17) Message boards : Number crunching : Low CPU utilization with differing GPU models (Message 2084)
Posted 3 Jan 2012 by Profile Beyond
Post:
Well I'm confused now. On the Moo site your single GPU box shows as 1.4.1607, which I think is 11.11:

Catalyst 11.11 WHQL is an OpenGL 4.1 and OpenCL 1.1 driver:
- OpenCL: CL_PLATFORM_VERSION: OpenCL 1.1 AMD-APP-SDK-v2.5 (793.1)
- OpenCL: CL_DRIVER_VERSION: CAL 1.4.1607 (VM)
- OpenGL version: 4.1.11251

http://www.geeks3d.com/20111116/amd-catalyst-11-11-whql/

All my 11.9 machines are reporting 1.4.1546.
18) Message boards : Number crunching : Low CPU utilization with differing GPU models (Message 2082)
Posted 3 Jan 2012 by Profile Beyond
Post:
*hic* Ta :) Zy

After recovering did you gain any insights on drivers?
19) Message boards : Number crunching : Low CPU utilization with differing GPU models (Message 2066)
Posted 2 Jan 2012 by Profile Beyond
Post:
I'll change to 11.9 see what happens - send out a search party if I am not back in 30 mins rofl :) Regards Zy

Rescue dogs sent, single malt scotch is in the flask around the neck (Guinness is in the backpack)...
20) Message boards : Number crunching : Low CPU utilization with differing GPU models (Message 2058)
Posted 2 Jan 2012 by Profile Beyond
Post:
That 100% loading stopped as soon as the fragmented units arrived - been jumping around on utilisation ever since. The closer to 12 (in the second last number in file name) it gets the more the utilisation. As crude examples ...

.... _400_768 will get around 9min + utilisation everywhere in band 65-99%

.... _12_768 will pretty well be on the money 8mins utilisation around 99%

Thanks Zydor for the explanation. Looking at the WUs, that's exactly what's happening here. The one's with the second to the last number closer to 12 are using 97-99% while the others are jumping all over the place. Completion times aren't that bad though even on the fragmented WUs. Are things running worse with your 4 GPU machine and is that the 12.1 preview driver? Better for you than 11.9?


Next 20


 
Copyright © 2011-2018 Moo! Wrapper Project