| 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 - March 26 2021 [ Issues Thread ] |
| No member browsing this thread |
|
Thread Status: Locked Total posts in this thread: 511
|
|
| Author |
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2068 Status: Recently Active Project Badges: 
|
Hehe, 8000 work units lasted less than 40 minutes. Edit:I wonder if some people here are running those "hacked" BOINC versions, with which you can trick the server to think that you have as many GPU's as you wish, and thereby download tons of GPU WU's? How many people worldwide are doing this. 8k in 40 minutes is not hard to believe. 1k crunchers, 8 tasks each, gone. JMHO Never heard of a GPU hack for BOINC. Any links? That was/is a "semi-secret" hack, used on SETI@home, mainly by members of the "GPU Users Group" team. It was never spread much outside of that team afaik. And no, there's no specifik link to it, but there are discussions about it on the SETI@home message board, in the "Number crunching" section. What specific threads though out of thousands, is impossible for me to remember. [Edit 2 times, last edit by Grumpy Swede at Mar 27, 2021 1:32:58 AM] |
||
|
|
maeax
Advanced Cruncher Joined: May 2, 2007 Post Count: 142 Status: Offline Project Badges: 
|
Nanoprobe, What graphics card do you have? Ellesmere is a strange name for one, but this shows that you got the opencl_ati version of the application. So it is not Intel GPU. Thanks, -Uplinger Ellesmere is a AMD RADEON WX7100 for example.
AMD Ryzen Threadripper PRO 3995WX 64-Cores/ AMD Radeon (TM) Pro W6600. OS Win11pro
|
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
Hehe, 8000 work units lasted less than 40 minutes. Edit:I wonder if some people here are running those "hacked" BOINC versions, with which you can trick the server to think that you have as many GPU's as you wish, and thereby download tons of GPU WU's? How many people worldwide are doing this. 8k in 40 minutes is not hard to believe. 1k crunchers, 8 tasks each, gone. JMHO Never heard of a GPU hack for BOINC. Any links? That was/is a "semi-secret" hack, used on SETI@home, mainly by members of the "GPU Users Group" team. It was never spread much outside of that team afaik. And no, there's no specifik link to it, but there are discussions about it on the SETI@home message board, in the "Number crunching" section. What specific threads though out of thousands, is impossible for me to remember. Seems like a lot of coding work for minimal gain. Doubt that would get past Keith here. 
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
SOS_Ready
Cruncher Joined: Sep 23, 2008 Post Count: 19 Status: Offline Project Badges:
|
Got 12 WUs on Win 10 PRO, Intel i7 860, ATI 7950
and 41 WUs on Win 10 PRO, 2 x Opteron 6386 SE, 2 x Nvidia TITAN. All completed successfully. |
||
|
|
ChrisR1964
Cruncher Usa Joined: Dec 5, 2005 Post Count: 17 Status: Offline Project Badges: 
|
@Uplinger, I'm seeing the same issue I saw on the previous beta tests for OpenPandemics. The work unit completes 100% but continues to "run" for hours. I have work unit BETA_OPNG_0000075_00245_0 and it has been running for 3+ hours. I have AMD Radeon HD 6530D. Thank you That was one of the symptoms i had from anti malware blocking BOINC processes Agreed, can you check if you have some 3rd party software that is preventing it from continuing? Thanks, -Uplinger @uplinger, I don't have any third party antivirus or anti malware. I do have Windows 10 default Antivirus (Windows Defender Antivirus) I turned it off and we will see. BETA_OPNG_0000075_00245_0 has been running for 5+hours with the windows antivirus on. I paused the antivirus and then paused this work unit. BETA_OPNG_0000084_00168_1 hit 100% in 32 minutes. I think it took about 10 minutes to get to 99%. At an hour it is still running at 100%. The progress percentage increases much slower the closer it gets to 100%. The last one percent takes longer than the first 99 percent. Thanks |
||
|
|
maeax
Advanced Cruncher Joined: May 2, 2007 Post Count: 142 Status: Offline Project Badges:
|
AMD HD6670(Turks) Win10pro run into a Error:
----------------------------------------https://www.worldcommunitygrid.org/ms/device/...og.do?resultId=1588667200
AMD Ryzen Threadripper PRO 3995WX 64-Cores/ AMD Radeon (TM) Pro W6600. OS Win11pro
|
||
|
|
robertmiles
Senior Cruncher US Joined: Apr 16, 2008 Post Count: 443 Status: Offline Project Badges: 
|
[snip]
@Ulrich Metzner, I do not know why, but it seems like the GT cards are generally on my naughty list of graphics cards that can't work. I will need to do some research to see what the differences are between GT and GTX cards. [snip] Thanks, -Uplinger The best I can tell, the only difference between GT cards and GTX cards is that the GT cards tend to have GPUs from the lower end of the range and GTX cards have GPUs from the higher end of the range. At least, you no longer have to worry about G cards and GTS cards - those haven't been offered for years. If you seen an M added to the end of the card type, that means a model for laptops, which often doesn't have double precision. |
||
|
|
FritzB
Cruncher Joined: May 8, 2012 Post Count: 17 Status: Offline Project Badges: 
|
Thanks uplinger for the continous work and updates!
----------------------------------------Works fine here on RX 5700, HD 7970, 280X, 1660 Super, Tonight one crashed after some good wus on Ryzen 4800H/Vega 7. Graphics driver 21.3.1. On previous beta, it happened everytime! Now I mentioned a message of AMD bug reporting tool, that something went wrong. But it didn's say what exactly. I have seen this behavior with my RX 5700 at Einstein@Home. But there, the next wus just ran normally. So it did here: https://www.worldcommunitygrid.org/ms/device/...og.do?resultId=1589244151 I got two types of errors. The first one only one time and the second one three times. The message "Die Umgebung stimmt nicht." means something like "the environment doesn't fit/suit/is incorrect" in german. https://www.worldcommunitygrid.org/ms/device/...og.do?resultId=1589244152 Result Name: BETA_ OPNG_ 0000108_ 00129_ 2-- <core_client_version>7.16.11</core_client_version> <![CDATA[ <message> (unknown error) - exit code 4294967295 (0xffffffff)</message> <stderr_txt> projects/www.worldcommunitygrid.org/wcgrid_beta29_autodockgpu_7.28_windows_x86_64__opencl_ati_102 -jobs OPNG_0000108_00129.job -input OPNG_0000108_00129.zip -seed 59670422 -wcgruns 1300 -wcgdpf 26 INFO: Using gpu device from app init data 0 INFO:[03:42:13] Start AutoGrid... autogrid4: Successful Completion. INFO:[03:42:27] End AutoGrid... INFO:[03:42:27] Start AutoDock for ZINC000884515945-ACR2.7_RX1--fr2266benz_001--CYS114.dpf(Job #0)... OpenCL device: gfx902 INFO:[03:43:26] End AutoDock... INFO:[03:43:26] Start AutoDock for ZINC000419890294-ACR2.18_RX1--fr2266benz_001--CYS114.dpf(Job #1)... OpenCL device: gfx902 INFO:[03:43:51] End AutoDock... INFO:[03:43:52] Start AutoDock for ZINC001353303786-ACR2.5_RX1--fr2266benz_001--CYS114.dpf(Job #2)... OpenCL device: gfx902 INFO:[03:44:28] End AutoDock... INFO:[03:44:30] Start AutoDock for ZINC000442624186-ACR2.19_RX1--fr2266benz_001--CYS114.dpf(Job #3)... OpenCL device: gfx902 INFO:[03:45:02] End AutoDock... INFO:[03:45:03] Start AutoDock for ZINC000236791699-ACR2.7_RX1--fr2266benz_001--CYS114.dpf(Job #4)... OpenCL device: gfx902 INFO:[03:45:33] End AutoDock... INFO:[03:45:34] Start AutoDock for ZINC001156986311-ACR2.18_RX1--fr2266benz_001--CYS114.dpf(Job #5)... OpenCL device: gfx902 INFO:[03:46:14] End AutoDock... INFO:[03:46:14] Start AutoDock for ZINC000905840968-ACR2.15_RX1--fr2266benz_001--CYS114.dpf(Job #6)... OpenCL device: gfx902 INFO:[03:46:48] End AutoDock... INFO:[03:46:49] Start AutoDock for ZINC001327487852-ACR2.5_RX1--fr2266benz_001--CYS114.dpf(Job #7)... OpenCL device: gfx902 INFO:[03:47:26] End AutoDock... INFO:[03:47:27] Start AutoDock for ZINC000519165799-ACR2.5_RX1--fr2266benz_001--CYS114.dpf(Job #8)... OpenCL device: gfx902 INFO:[03:47:51] End AutoDock... INFO:[03:47:51] Start AutoDock for ZINC001327514600-ACR2.19_RX1--fr2266benz_001--CYS114.dpf(Job #9)... OpenCL device: gfx902 INFO:[03:48:28] End AutoDock... INFO:[03:48:29] Start AutoDock for ZINC000357075516-ACR2.14_RX1--fr2266benz_001--CYS114.dpf(Job #10)... OpenCL device: gfx902 INFO:[03:49:09] End AutoDock... INFO:[03:49:10] Start AutoDock for ZINC000883442957_2-ACR2.15_RX1--fr2266benz_001--CYS114.dpf(Job #11)... OpenCL device: gfx902 </stderr_txt> ]]> Update: Just mentioned that these errored on 1650 Ti mobile. The other three: https://www.worldcommunitygrid.org/ms/device/...og.do?resultId=1589095562 https://www.worldcommunitygrid.org/ms/device/...og.do?resultId=1589063440 https://www.worldcommunitygrid.org/ms/device/...og.do?resultId=1589079992 e.g. Result Name: BETA_ OPNG_ 0000107_ 00310_ 1-- <core_client_version>7.16.11</core_client_version> <![CDATA[ <message> Die Umgebung stimmt nicht. (0xa) - exit code 10 (0xa)</message> <stderr_txt> projects/www.worldcommunitygrid.org/wcgrid_beta29_autodockgpu_7.28_windows_x86_64__opencl_nvidia_102 -jobs OPNG_0000107_00310.job -input OPNG_0000107_00310.zip -seed 1017051530 -wcgruns 1250 -wcgdpf 25 INFO: Using gpu device from app init data 0 INFO:[01:33:13] Start AutoGrid... autogrid4: Successful Completion. INFO:[01:33:28] End AutoGrid... INFO:[01:33:28] Start AutoDock for ZINC000655925119-ACR2.1_RX1--fr2266benz_001--CYS114.dpf(Job #0)... GPU not found: type=NVIDIA, opencl_device_index=0, device_num=0 Error: boinc_get_opencl_ids() failed with error -1 01:33:42 (6752): called boinc_finish(10) </stderr_txt> ]]> [Edit 1 times, last edit by FritzB at Mar 27, 2021 9:00:10 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
For all who are interested, my (admittedly older) machines have so far done all wu's without any problems:
Greeting from Germany Mario |
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1313 Status: Offline Project Badges:
|
As some have noticed, my theory on why some are getting more than others is incorrect. I'm not worried about some getting more than their threads at the moment, but it does introduce something I should look into for the mechanism which schedules beta work units. Thanks, -Uplinger Hello Keith, I noticed that the number of sent tasks only are exceeding the number of cpu-threads when the GPU is of the integrated type APU or iGPU. Good luck! CP  |
||
|
|
|