| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: OpenPandemics GPU Beta Test - Feb 18 2021 [Issues Thread] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 272
|
|
| Author |
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
FYI,
I am loading up the remaining work for this round of beta. These will distribute over the 30 minutes I would imagine. Thanks again for all your help in BETA testing. We are getting lots of valuable information to use going forward. Since this is the last group of work being loaded, after these are distributed it will probably be a few days before we have more work available. -Uplinger |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
Near the top of the device profile, you should see a Processor section with this question: "Allow research to run on my CPU?" But that seems to apply to all the WCG projects, unless I am missing something. Perhaps it really means "Do not run projects on the CPU that can be run on the GPU", as suggested above. We need something to allow us to run OPN only on the GPU, while still allowing us to run the other WCG projects on the CPU. Howdy Jim, At the moment, we do not have plans to make such a feature available. I will bring it up with our team on Monday to see if it is feasible and easy to implement something along those lines. But as you can guess our primary goal at the moment is to get GPU running properly and successfully. Thanks, -Uplinger |
||
|
|
Richard Haselgrove
Senior Cruncher United Kingdom Joined: Feb 19, 2021 Post Count: 360 Status: Offline Project Badges: 
|
Seem to be having a problem with intel_gpu version, under Windows 7 First machine - daily driver. I saw an intel HD driver failure, on both the original WU and a replacement. Second machine - no user interaction. Had stopped even before the first checkpoint, so BOINC was only showing pseudo-progress (trying to reassure me it was running, and succeeded). Aborting both sets, and suspending new task fetch for intel_gpu. NVidia tasks are running fine on "GeForce GTX 1650 SUPER (driver version 452.06, device version OpenCL 1.2 CUDA, 4096MB, 3034MB available, 4531 GFLOPS peak)". [Edit 1 times, last edit by Richard Haselgrove at Feb 20, 2021 6:11:40 PM] |
||
|
|
torma99
Cruncher Hungary Joined: Mar 30, 2020 Post Count: 12 Status: Offline Project Badges: 
|
Hi Richard! Thanks for adding the screen shot. Also, I'm about to release about 5000 work units (2 copies of each) in few minutes. These are going to be the larger workunits which on average ran about 15 minutes each. This is part of a larger amount of work units going to be released over the next couple of hours. That is going to be about 22k work units in total. The batch numbers for these are: 20010 - 20099. Thanks, -Uplinger I've just arrived home and grabbed some work units from this batch. They take only 3 minutes for my PC not 15 as you mentioned. It has a 2070 RTX. Is this a problem? batches are 20053,20055 |
||
|
|
Mumak
Senior Cruncher Joined: Dec 7, 2012 Post Count: 477 Status: Offline Project Badges: 
|
I'm attempting to catch some tasks on the Intel DG1, but nothing is coming.. No issue on several other GPUs..
----------------------------------------Wondering whether there is perhaps some GPU filter on server side? | OpenCL: Intel GPU 0: Intel(R) Iris(R) Xe MAX Graphics (driver version 27.20.100.9168, device version OpenCL 3.0 NEO, 6456MB, 6456MB available, 1190 GFLOPS peak) Nothing but: 2/20/2021 7:09:54 PM | World Community Grid | Requesting new tasks for Intel GPU 2/20/2021 7:09:56 PM | World Community Grid | Scheduler request completed: got 0 new tasks  [Edit 1 times, last edit by Mumak at Feb 20, 2021 6:17:06 PM] |
||
|
|
Dayle Diamond
Senior Cruncher Joined: Jan 31, 2013 Post Count: 447 Status: Offline Project Badges: 
|
We need something to allow us to run OPN only on the GPU, while still allowing us to run the other WCG projects on the CPU. If the GPU performance is as substantial as it appears, no additional CPU work should be created. The GPU crunchers are going to be sucking up enough work units to keep turnaround times at a minimum. All of yesterday's entire crunching could be replaced with 1 year, 7 months, 22 hours of GPU folding. Why would we ask contributors to run OPN at 1/200th efficiency when we can ask them to run Mapping Cancer Markers at 100% efficiency? |
||
|
|
zombie67 [MM]
Senior Cruncher USA Joined: May 26, 2006 Post Count: 228 Status: Offline Project Badges: 
|
Is there an app for the OSX/intel-GPU combination? I haven't been able to get any work for my two machines with that.
---------------------------------------- |
||
|
|
Bartlomiej Batkowski
Advanced Cruncher Poland Joined: Jun 21, 2007 Post Count: 50 Status: Offline Project Badges:
|
4 machines, 43 workunits = 43 errors..
|
||
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 145 Status: Offline Project Badges: 
|
Just so it is shown by the same thing, here is a snip from MSI afterburner. The first part is running folding at home. The middle part where there are small spikes upwards is when boinc found another beta gpu unit before I noticed and paused folding at home. The last part is with just the beta gpu unit and shows what I was referring to.
---------------------------------------- This may not be as noticeable on less powerful graphics cards. The GTX 750Ti that pututu showed has the loaded parts to be wider. Much more so with the intel uhd 620, given how much less powerful it is.  |
||
|
|
|