Posts by Link

61) Message boards : Number crunching : 2 GPUs for one task? (Message 8390) Posted 24 Oct 2022 by Link Post: I assume you want Moo tasks, not more MW? Coproc_debug on, Moo running. Yes, now I want Moo.
62) Message boards : Number crunching : 2 GPUs for one task? (Message 8388) Posted 24 Oct 2022 by Link Post: OK, Milkyway is definitely using both AMD GPUs as some results have Using device 1 on platform 1 and some Using device 0 on platform 1. Now crunch them all and enable Moo (with <coproc_debug> enabled). The 3 valid tasks that are left here are all crunched on GPU 1, but that could be coincidence.
63) Message boards : Number crunching : 2 GPUs for one task? (Message 8387) Posted 24 Oct 2022 by Link Post: Is that the computer we are talking about on Milkyway? 929850 Yes. How did you know? Your only computer there with 2 GPUs and a i5-8600K. Can you let Milkyway run few tasks? If we see it using sometimes dev0 and sometimes dev1 (Milkyway has that info in std_err output) than everything is OK with your setup and unfortunately I am in this case out of ideas regarding why Moo is running only 1 task. It just doesn't make any sense why it wouldn't do that. Done. And enable the <coproc_debug>, than we have that too.
64) Message boards : Number crunching : 2 GPUs for one task? (Message 8385) Posted 24 Oct 2022 by Link Post: Primegrid is running just now: http://www.primegrid.com/results.php?hostid=1164341 Great, can't see the results without being member of Primegrid. :-( Anything useful in the std_err output of the completed tasks regarding device 0 and 1? Is it running 1 WU on both GPUs or 1 WU on each GPU, i.e. 2 WUs total? Is that the computer we are talking about on Milkyway? 929850 Can you let Milkyway run few tasks? If we see it using sometimes dev0 and sometimes dev1 (Milkyway has that info in std_err output) than everything is OK with your setup and unfortunately I am in this case out of ideas regarding why Moo is running only 1 task. It just doesn't make any sense why it wouldn't do that. EDIT: or enable <coproc_debug> in cc_config and post those messages. I think that should be enough as first step.
65) Message boards : Number crunching : 2 GPUs for one task? (Message 8383) Posted 24 Oct 2022 by Link Post: 2. You have <use_all_gpus>0</use_all_gpus> in cc_config.xml. I did. I changed it to 1 when you suggested it. This didn't help Moo. Did you restart the BOINC client after the change? Reading config files isn't enough for this setting. 3. All BOINC projects except Moo run two GPU Wus at the same time. Correct. It behaves precisely like two single chip cards. Which other projects do you run on this computer? Can you post links to results on those other projects?
66) Message boards : Number crunching : 2 GPUs for one task? (Message 8381) Posted 24 Oct 2022 by Link Post: Hmm... so just to make sure I understand everything correctly. You say that 1. Crossfire is turned off. 2. You have <use_all_gpus>0</use_all_gpus> in cc_config.xml. 3. All BOINC projects except Moo run two GPU Wus at the same time. Right?
67) Message boards : Number crunching : 2 GPUs for one task? (Message 8379) Posted 24 Oct 2022 by Link Post: I vaguely remember that eventually you might need to disable crossfire, so the card becomes really two separate cards. I remember from Milkyway some discussions about cards with 2 GPUs on them, maybe you should try to search there in the forums.
68) Message boards : Number crunching : 2 GPUs for one task? (Message 8377) Posted 24 Oct 2022 by Link Post: <use_all_gpus>0</use_all_gpus> must be <use_all_gpus>1</use_all_gpus> Restart BOINC after the change and it should use both GPUs. Your <exclude_gpu> tags are wrong as well, you have only ATI device 0 and 1 and Intel device 0, I'm not even sure what you try to achive there. Ever had more than 10 GPUs in that system?
69) Message boards : Number crunching : 2 GPUs for one task? (Message 8375) Posted 24 Oct 2022 by Link Post: I can't see your 2nd link at the moment, the server isn't responding, although your first link is ok. Yeah, it was difficult to open it for me too, you need to wait like half minute before it actually starts to load, this computer has long task list. Or check dnetc_r72_1666176892_12_768_0 and dnetc_r72_1666176803_12_768_0, that are currently the two first tasks from the list. Are you saying you can't get both GPUs to use Moo, or you get one task on each? You can get one task on each GPU. I can't run Moo on two Tahiti AMD chips, which are on the same card. Boinc treats them as two seperate GPUs, and so does MSI Afterburner, but Moo doesn't notice the second one. Post your start up messages from BOINC, in particular the part where BOINC detects the GPUs, your entire cc_config.xml and your Moo app_config.xml if you have one.
70) Message boards : Number crunching : 2 GPUs for one task? (Message 8373) Posted 24 Oct 2022 by Link Post: That makes sense, but why does his boinc manager show "2NV"? Like I said, probably that's what he has put in his app_config.xml, probably when trying to get Moo to run on both cards. But that should be done via cc_config.xml, with <use_all_gpus>1</use_all_gpus>. Just guessing, but I don't see anything else, that could cause this. Can you find a computer with two identical Nvidias? Not hard, just browse the top computers list and you get for example this one. Same here, some tasks run on GPU 0, some on GPU 1.
71) Message boards : Number crunching : 2 GPUs for one task? (Message 8371) Posted 23 Oct 2022 by Link Post: you guys saw this all on my post in the other project. I can import all that here if needed. Just what you're managing to do on Moo - that one task uses 2 CPU cores, 2 GPUs, they're maxed out, and what models they are. And anything else relevant. Regardless of which of his Moo results I click on, I see this in std_err: 12:35:54 (17420): device: OpenCL: NVIDIA GPU 0: NVIDIA GeForce GTX 1080 (driver version 522.30, device version OpenCL 3.0 CUDA, 8192MB, 8192MB available, 9070 GFLOPS peak) 12:35:54 (17420): device: OpenCL: NVIDIA GPU 1 (not used): NVIDIA GeForce GTX 1050 Ti (driver version 522.30, device version OpenCL 3.0 CUDA, 4096MB, 4096MB available, 2274 GFLOPS peak) I couldn't find any result which used the 1050 Ti. So like I posted in the Rosetta thread, it would be interesting to see how the GPU load looks like with just Moo running and with just FAH running. If you check another computer with more than one Nvidia GPU (and Windows, Linux std_err does not have that info), for example this one, you will easily find both 18:29:10 (14364): device: OpenCL: NVIDIA GPU 0 (not used): NVIDIA GeForce GTX 1660 Ti (driver version 512.15, device version OpenCL 3.0 CUDA, 6144MB, 6144MB available, 5622 GFLOPS peak) 18:29:10 (14364): device: OpenCL: NVIDIA GPU 1: NVIDIA GeForce GTX 1080 (driver version 512.15, device version OpenCL 3.0 CUDA, 8192MB, 8192MB available, 9523 GFLOPS peak) and 18:10:45 (4696): device: OpenCL: NVIDIA GPU 0: NVIDIA GeForce GTX 1660 Ti (driver version 512.15, device version OpenCL 3.0 CUDA, 6144MB, 6144MB available, 5622 GFLOPS peak) 18:10:45 (4696): device: OpenCL: NVIDIA GPU 1 (not used): NVIDIA GeForce GTX 1080 (driver version 512.15, device version OpenCL 3.0 CUDA, 8192MB, 8192MB available, 9523 GFLOPS peak) My guess is Moo is running on GPU 0 and FAH on both (since the 1080 Moo runtimes are about double compared to the example computer, which has the same card as GPU 1). The 2xCPU and 2xGPU in BOINC are from app_config.xml, you can insert there any number of CPUs and GPUs, which BOINC should reserve for a task, even more than you actually have. The app will still use what it was designed for (i.e. 1 GPU). EDIT: you see that actually in the last task manager screenshot he posted: Moo is running on GPU 0 and uses 6.3% CPU (1 core), the two instances of FAH are using each one GPU and each one CPU core.
72) Message boards : Number crunching : No blocks to work on? (Message 8355) Posted 29 Aug 2022 by Link Post: Do you still have the application in your project folder? I had the same issue a while ago, since I didn't get any answer I ended up with creating my own app_info and running the app as anonymous, but if you still have it, than that's not the issue. Can you try to request CPU tasks? (with very low cache setting if you don't want too many of them) EDIT: I see you got one single task today, so in general your computer is able to receive them... no idea what's the issue than.
73) Message boards : Number crunching : Missing blocks on stats.distributed.net since 12th August (Message 8353) Posted 24 Aug 2022 by Link Post: And the stats are working again.???? Seems our blocks were delivered yesterday, you got nearly 2.6 millions, I 9839. Same huge peaks for all other crunchers I posted as examples. So everything OK again. :)
74) Message boards : Number crunching : Missing blocks on stats.distributed.net since 12th August (Message 8350) Posted 23 Aug 2022 by Link Post: Actually it seems that our results are not being send back to DNET. Non-Moo! users don't seem to be affected, so it's very likely a Moo! issue. Can be easily checked in the top list, users with @moowrap.net at the end of their name (and who are still active), have significant drop in their statistics around the 12th August (it's a bit different from user to user and sometimes not clearly visible, but if you check many, you'll see it). Here, here, here very clear, me, you? (can't see your computers, but considering your RAC I assume you crunched more than just two "large" WUs in the past 6 days). Highly unlikely that all of us nearly quit crunching same day. I'm missing about 10000 blocks by now (1 block gives us here 8 credits). EDIT: According to BOINCstats you crunched over 2 million blocks in the past 3 days, not just those 1,536 that made it back to DNET the 21st August.
75) Message boards : Number crunching : Missing blocks on stats.distributed.net since 12th August (Message 8348) Posted 18 Aug 2022 by Link Post: Same story as before. I guess something needs a kick again.
76) Message boards : Number crunching : [error] exceeded limit of 800 slot directories ??? (Message 8340) Posted 9 Mar 2022 by Link Post: Suspend all new work, complete all running tasks (if possible), exit BOINC, than delete all subfolders of your slots directory (if you didn't change it, it's C:\ProgramData\BOINC\slots).
77) Questions and Answers : Wrapper and Client : Gaining time (Message 8334) Posted 20 Feb 2022 by Link Post: BOINC needs to find out how fast your computers are, the estimates at the beginning might be very wrong. On projects not using a local DCF (duration correction factor) you need to return 10 valid results for each app to get quite good estimates, on projects using it it will just slowly get to the right value.
78) Questions and Answers : Getting started : Does this project actually support Android? (Message 8333) Posted 20 Feb 2022 by Link Post: Yes, using it as backup project on my Xperia Ray. But no idea what's causing the issue in your case, AFAICT Android devices are in general a bit difficult to get to crunch (and keep crunching) not only just on this project, I had some issues with Einstein myself. :-(
79) Questions and Answers : Windows : Can't add project on 32bit x86 (Message 8332) Posted 20 Feb 2022 by Link Post: It should work according to applications page. Which BOINC version do you use on that machine? Please post the log of what happens when you try to attach.
80) Message boards : Number crunching : please allow us to choose cuda or opencl in settings (Message 8328) Posted 31 Jan 2022 by Link Post: Is it possible to select only the OpenCL app via APP_INFO or similar configuration file? Yes, you can use the one I posted while talking here to myself as reference. ;-) The files you should already have, so skip steps 1 and 2, and the infos about them you'll find in your client_state.xml, so just need to copy to the app_info.xml from there.

Previous 20 · Next 20