Posts by Magiceye04

1) Message boards : Number crunching : Why èarn some WUs a very low credit? (Message 77045) Posted 20 days ago by Magiceye04 Post: [quote] one central element of the CreditNew algorithm is the use of moving averages over time. /quote] I think I know this effect from other projects. But there nearly all WUs are affected and show a bit lower credit at the beginning. In Milkyway there are only a few WUs affected and the have significantly low credit. But if this also can be explained by "new credit", I am fine with it. Last night no new low credits were seen. In the meantime I have also seen some of the "skipping" WUs, but they run only a few seconds, so the 0,x credits are OK for this short time.
2) Message boards : Number crunching : Why èarn some WUs a very low credit? (Message 77042) Posted 20 days ago by Magiceye04 Post: Tasks like this one have no usable work involved. So no or very little credit. 0.11 credits https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=943807851 If you inspect the stderr.txt output you see this kind of statement. Number of particles in bins is very small compared to total. (0 << 1). Skipping distance calculation Such tasks I have not seen in my list.
3) Message boards : Number crunching : Why èarn some WUs a very low credit? (Message 77037) Posted 21 days ago by Magiceye04 Post: low: https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=943784423 high: https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=943780641
4) Message boards : Number crunching : Option in project preferences to set max CPUs (Message 77036) Posted 21 days ago by Magiceye04 Post: Finn the Human wrote: Thanks, I've set mine to also run multiple lower CPU count WUs. Is there any performance increase you see for doing this? CPU utilization at high thread counts per task is not very good, therefore it occurs to me that host utilization is better with lower thread counts per task and higher number of concurrent tasks on a host. I am considering to eventually measure whether or not there is merit to my assumption, but haven't taken the time yet to set up a respective experiment, which would involve running a fixed workunit outside of BOINC for hopefully direct and precise comparison of different threads/tasks settings. I have also noticed that. On my 16core CPU I had to use 3x15Threads to get around 100% CPU load. Now I try with 4x10Threads which is also running fine. I have told my CPU to be a 45core via cc_config.xml. With 2x16Threads I only saw about 50% load. The first minute only one Thread is under load and the second minute 12 Threads were loaded. This is really bad efficiency.
5) Message boards : Number crunching : How can I fix the estimated duration time? (Message 77033) Posted 21 days ago by Magiceye04 Post: Boinc shows me estimated time = 2 hours. But the WU run only 2-3 minutes. In other projects I can fix this via duration correction factor. But for milkyway this shows no change and I see this option in the client-state.xml: <dont_use_dcf/> Sounds to me that I can not change it?
6) Message boards : Number crunching : Why èarn some WUs a very low credit? (Message 77031) Posted 21 days ago by Magiceye04 Post: About 70% of my results look like this: run time \| CPU time \| credits 223.89 \| 1,807.51 \| 28.25 But the other 30% look like this: run time \| CPU time \| credits 224.09 \| 1,863.43 \| 1.41 Why are there so many results with such low credits? This happens on 2 different PCs.
7) Message boards : Number crunching : bad argument #0 to 'calculateEps2' (Expected 3 or 6 arguments) (Message 77017) Posted 21 days ago by Magiceye04 Post: Thanks!
8) Message boards : Number crunching : bad argument #0 to 'calculateEps2' (Expected 3 or 6 arguments) (Message 77015) Posted 21 days ago by Magiceye04 Post: How can I prevent getting the bad WUs ? Something I can add in the app_config?
9) Message boards : News : Separation Project Coming To An End (Message 75862) Posted 22 Jun 2023 by Magiceye04 Post: We would appreciate your input on this because we expect that it will probably take some time for GPU-oriented users to swap that hardware over to different projects. How long would you like us to wait before we shut down Separation? It takes 10 seconds to switch to another project. Just do it and don't let people who do not read the forum waste their resources!
10) Message boards : News : Server Outages (Message 71501) Posted 13 Dec 2021 by Magiceye04 Post: Thank you! Idea: not send out new work until the pile is gone.
11) Message boards : News : Validator Outage (Message 71480) Posted 12 Dec 2021 by Magiceye04 Post: Wouldn't it be a good idea to give the validator some air to breathe? I will stop working until the problem is fixed.
12) Message boards : News : New Poll Regarding GPU Application of N-Body (Message 71120) Posted 18 Sep 2021 by Magiceye04 Post: I would be very happy to see longer running work units on GPUs. Especially on high performance AMD cards (with a lot of double precision performance) i have to run several WUs in parallel, which causes driver issues. I got some Fixed by AMD, but not all. If i could run a N-Body WU on my GPU and it takes several hours and loads the GPU well, it would be great. The comparison specifically preformed was between a i9-10900k and RTX 3070, using all available compute cores for both. When is is the comparison, then AMD cards with a lot of double precision performance should do well, as well as Nvidia cards. A 3070 has roughly 0.3 TFlop double precision performance. A 10900K should turn out the same, since memory bandwith is limited on the CPU. I would expect a R9 280X to be 3 times as fast as a 3070 then. I totally agree. Even with 4 WUs in parallel my VII finishes the WUs in less than 1 minute. Why are the WUs so short? In my opinion a new GPU app makes only sense if the work is done more efficient then on the CPU. My VII and the 3950X are set both to ~90W and in this config I would expect the VII to be at minimum 2-3x faster then the 32 CPU threads. Otherwise the next jump from AMD to 32 CPU cores for the consumer platform would make this advantage for the GPU obsolete. Threadripper already have more CPU cores, but also much more power consumption.
13) Message boards : Number crunching : 80 WU limit (Message 67585) Posted 9 Jun 2018 by Magiceye04 Post: I would be happy about 80 WUs. Also about 4 WUs. But i only get 1. This means, the GPU is more or less in idle state. How do i get more than 1 ?
14) Message boards : Number crunching : Again a lot of inconclusive WU (Message 67584) Posted 9 Jun 2018 by Magiceye04 Post: I run several hosts, differents CPU/GPU and drivers. All are visible. About 50% of my WU finish in "inconclusive" Reset project not change anything. Who can explain ? The proper phrase is "Validation inconclusive". It means they have not been compared with the results from other users. About half of mine are in that status too. That will eventually be compared, and usually validated. But the correct state for this would be "waiting for validation" like in other projects. Why is using Milkyway this misunderstanding state that is normally used, when 2 PCs deliver different results - and therfore really need a 3rd or 4th run?