Posts by Borgholio

21) Message boards : Number crunching : Seti android app (Message 1379326)
Posted 10 Jun 2013 by Profile Borgholio
Post:
It says I need to install the bug catch version. Will this prevent me from updating automatically from the play store later?

Edit - installed the bugcheck version. The buttons for sending or saving the logs are greyed out.
22) Message boards : Number crunching : Seti android app (Message 1379163)
Posted 9 Jun 2013 by Profile Borgholio
Post:
Seti v7 keeps crashing out on my Galaxy S3. Out of a dozen workunits sent, only two successfully completed. I have tried detaching and re-attaching the project, as well as fully re-installing BOINC. No luck. Other projects seem to work fine.
23) Message boards : News : SETI@home Version 7 has been released (Message 1378650)
Posted 8 Jun 2013 by Profile Borgholio
Post:
So version includes auto correlation which, I assume based on reading the wiki, makes it more sensitive to weak signals? That's certainly a good thing. I'm still curious to know exactly how much more sensitive it makes S@H, and if there are any other improvements in v.7...
24) Message boards : Number crunching : Lunatics Windows Installer v0.41 Release Notes (Message 1376232)
Posted 3 Jun 2013 by Profile Borgholio
Post:
Installed and working great. Once again, hats off to Lunatics!

One question regarding the CUDA versions. I'm running a 9800 series GeForce so I selected 3.2. What would actually happen if I were to select a higher version of CUDA? Would it just not work, or would it not work as well as 3.2? If I stick with 3.2, will that prevent me from getting certain kinds of work from Seti?

Thanks!

They would work, it just wouldn't be as fast as the Cuda32 app, My testing on my 9800GTX+ showed a slowdown on Cuda32 and Cuda42 apps on drivers later than 301.xx, the Cuda5 app was even slower,
there was no slowdown on the Cuda23 app, so since you're running 311.06, you may want to try the Cuda23 app instead.

Claggy


So 2.3 is actually faster than 3.2 for a 9800 series card?
25) Message boards : Number crunching : Lunatics Windows Installer v0.41 Release Notes (Message 1376207)
Posted 3 Jun 2013 by Profile Borgholio
Post:
Installed and working great. Once again, hats off to Lunatics!

One question regarding the CUDA versions. I'm running a 9800 series GeForce so I selected 3.2. What would actually happen if I were to select a higher version of CUDA? Would it just not work, or would it not work as well as 3.2? If I stick with 3.2, will that prevent me from getting certain kinds of work from Seti?

Thanks!
26) Message boards : News : SETI@home Version 7 has been released (Message 1374963)
Posted 1 Jun 2013 by Profile Borgholio
Post:
Still no explanation on what v7 does. Yesterday has come and gone...


As mentioned in this very thread, there have been other more urgent issues that have required people's attention.



Sorry, didn't see it through all the spam in this thread.
27) Message boards : News : SETI@home Version 7 has been released (Message 1374906)
Posted 1 Jun 2013 by Profile Borgholio
Post:
Still no explanation on what v7 does. Yesterday has come and gone...
28) Message boards : News : SETI@home Version 7 has been released (Message 1374538)
Posted 31 May 2013 by Profile Borgholio
Post:
So any word yet on what v7 actually does differently?
29) Message boards : Number crunching : New scheduler extremely conservative? (Message 1263132)
Posted 21 Jul 2012 by Profile Borgholio
Post:
why would it still be at 1+10?

Local preferences? They take precedence over the online ones.

Gruß,
Gundolf



*bangs head against desk repeatedly*

For the love of god why do I keep forgetting about local preferences? Yep, all were still set at 1 and 10...cleared preferences and downloaded 100+ tasks from a variety of projects including many from SETI.

Ugh...
30) Message boards : Number crunching : New scheduler extremely conservative? (Message 1263038)
Posted 21 Jul 2012 by Profile Borgholio
Post:
odd...i had updated the cache to be 10 + 10 before creating the log file...why would it atill be at 1+10?
31) Message boards : Number crunching : New scheduler extremely conservative? (Message 1262756)
Posted 21 Jul 2012 by Profile Borgholio
Post:
Here is the output with my cache filled to the brim and work enabled on all but one project. If this doesn't show anything of value, I'll post another after the cache has had a day or two to deplete (when it should have stayed topped off).

20-Jul-2012 17:05:33 [---] [work_fetch] work fetch start
20-Jul-2012 17:05:33 [---] [work_fetch] ------- start work fetch state -------
20-Jul-2012 17:05:33 [---] [work_fetch] target work buffer: 86400.00 + 864000.00 sec
20-Jul-2012 17:05:33 [ABC@home] [work_fetch] REC 0.195 priority -0.046248
20-Jul-2012 17:05:33 [rosetta@home] [work_fetch] REC 15.101 priority -3.584345
20-Jul-2012 17:05:33 [Poem@Home] [work_fetch] REC 1030.412 priority -244.569158
20-Jul-2012 17:05:33 [The Lattice Project] [work_fetch] REC 9.074 priority -2.153791
20-Jul-2012 17:05:33 [boincsimap] [work_fetch] REC 1.647 priority -0.391007
20-Jul-2012 17:05:33 [superlinkattechnion] [work_fetch] REC 0.137 priority -0.000000 (project backoff 14975.04) (master fetch pending)
20-Jul-2012 17:05:33 [climateprediction.net] [work_fetch] REC 106.457 priority -25.272334
20-Jul-2012 17:05:33 [Docking] [work_fetch] REC 18.649 priority -4.426331
20-Jul-2012 17:05:33 [Einstein@Home] [work_fetch] REC 35.810 priority -8.499597
20-Jul-2012 17:05:33 [EDGeS@Home] [work_fetch] REC 0.497 priority -0.117930
20-Jul-2012 17:05:33 [Test4Theory@Home] [work_fetch] REC 0.006 priority -0.001314
20-Jul-2012 17:05:33 [LHC@home 1.0] [work_fetch] REC 0.000 priority -0.000000
20-Jul-2012 17:05:33 [Milkyway@Home] [work_fetch] REC 3.648 priority -0.865794
20-Jul-2012 17:05:33 [MindModeling@Beta] [work_fetch] REC 31.051 priority -7.369937
20-Jul-2012 17:05:33 [orbit@home] [work_fetch] REC 0.000 priority -0.000000
20-Jul-2012 17:05:33 [QMC@HOME] [work_fetch] REC 17.240 priority -4.091956
20-Jul-2012 17:05:33 [ibercivis] [work_fetch] REC 5.324 priority -1.263774
20-Jul-2012 17:05:33 [SETI@home] [work_fetch] REC 61286.359 priority -1.475791
20-Jul-2012 17:05:33 [http://spin.fh-bielefeld.de/] [work_fetch] REC 0.000 priority -0.000000
20-Jul-2012 17:05:33 [Cosmology@Home] [work_fetch] REC 0.753 priority -0.178805
20-Jul-2012 17:05:33 [Magnetism at home] [work_fetch] REC 0.000 priority -0.000000
20-Jul-2012 17:05:33 [GPUGRID] [work_fetch] REC 505.585 priority -0.000000 (no new tasks)
20-Jul-2012 17:05:33 [malariacontrol.net] [work_fetch] REC 104.985 priority -24.918381
20-Jul-2012 17:05:33 [PrimeGrid] [work_fetch] REC 421.551 priority -100.055470
20-Jul-2012 17:05:33 [yoyo@home] [work_fetch] REC 21.834 priority -5.182253
20-Jul-2012 17:05:33 [RNA World] [work_fetch] REC 0.309 priority -0.073291
20-Jul-2012 17:05:33 [uFluids] [work_fetch] REC 0.000 priority -0.000000
20-Jul-2012 17:05:33 [World Community Grid] [work_fetch] REC 1017.821 priority -0.096632
20-Jul-2012 17:05:33 [---] [work_fetch] CPU: shortfall 1675689.80 nidle 0.00 saturated 352884.09 busy 0.00
20-Jul-2012 17:05:33 [ABC@home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [rosetta@home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Poem@Home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [The Lattice Project] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [boincsimap] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [superlinkattechnion] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [climateprediction.net] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Docking] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Einstein@Home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [EDGeS@Home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Test4Theory@Home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [LHC@home 1.0] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Milkyway@Home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [MindModeling@Beta] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [orbit@home] [work_fetch] CPU: fetch share 0.164 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [QMC@HOME] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [ibercivis] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [SETI@home] [work_fetch] CPU: fetch share 0.657 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [http://spin.fh-bielefeld.de/] [work_fetch] CPU: fetch share 0.007 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Cosmology@Home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Magnetism at home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [GPUGRID] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (no apps)
20-Jul-2012 17:05:33 [malariacontrol.net] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [PrimeGrid] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [yoyo@home] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [RNA World] [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [uFluids] [work_fetch] CPU: fetch share 0.007 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [World Community Grid] [work_fetch] CPU: fetch share 0.164 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [---] [work_fetch] NVIDIA: shortfall 0.00 nidle 0.00 saturated 1035295.74 busy 0.00
20-Jul-2012 17:05:33 [ABC@home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [rosetta@home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Poem@Home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (blocked by prefs)
20-Jul-2012 17:05:33 [The Lattice Project] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (no apps)
20-Jul-2012 17:05:33 [boincsimap] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (no apps)
20-Jul-2012 17:05:33 [superlinkattechnion] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [climateprediction.net] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Docking] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Einstein@Home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (blocked by prefs)
20-Jul-2012 17:05:33 [EDGeS@Home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Test4Theory@Home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (no apps)
20-Jul-2012 17:05:33 [LHC@home 1.0] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (no apps)
20-Jul-2012 17:05:33 [Milkyway@Home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [MindModeling@Beta] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (no apps)
20-Jul-2012 17:05:33 [orbit@home] [work_fetch] NVIDIA: fetch share 0.164 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [QMC@HOME] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [ibercivis] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [SETI@home] [work_fetch] NVIDIA: fetch share 0.657 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [http://spin.fh-bielefeld.de/] [work_fetch] NVIDIA: fetch share 0.007 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Cosmology@Home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [Magnetism at home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [GPUGRID] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [malariacontrol.net] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [PrimeGrid] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) (blocked by prefs)
20-Jul-2012 17:05:33 [yoyo@home] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [RNA World] [work_fetch] NVIDIA: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [uFluids] [work_fetch] NVIDIA: fetch share 0.007 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [World Community Grid] [work_fetch] NVIDIA: fetch share 0.164 rsc backoff (dt 0.00, inc 0.00)
20-Jul-2012 17:05:33 [---] [work_fetch] ------- end work fetch state -------
20-Jul-2012 17:05:33 [---] [work_fetch] No project chosen for work fetch
32) Message boards : Number crunching : New scheduler extremely conservative? (Message 1262509)
Posted 20 Jul 2012 by Profile Borgholio
Post:
Yep, played around with the cache quite thoroughly. My resource share is set 10k on Seti, 2500 on WCG, and 1 on everything else. What flags would I need to add?
33) Message boards : Number crunching : New scheduler extremely conservative? (Message 1262347)
Posted 19 Jul 2012 by Profile Borgholio
Post:
Yep, looks like BOINC lumps the GPU scheduler into the CPU. After chugging through all my WCG units I'm back up to ~300 in my cache...and that's with the CPDN tasks still in there. Seems if my CPU cache fills up with non-SETI workunits, then my GPU cache of SETI workunits shrinks to only a day or two's worth. Time to report a bug...
34) Message boards : Number crunching : BOINC is bad at math, reason why some people can't keep a full cache? (Message 1261851)
Posted 18 Jul 2012 by Profile Borgholio
Post:
I have noticed something like this too. I am running a couple dozen projects including SETI. I have noticed that my GPU cache shrinks when I get a lot of CPU tasks from other projects, especially Climateprediction.net. Based on my testing in this thread:

http://setiathome.berkeley.edu/forum_thread.php?id=68453

I came to the conclusion that the number of tasks in queue for the GPU depend on how many tasks are in queue for the CPU. That is, the schedulers for CPU and GPU are in fact NOT separate from each other, as it was back when individual scheduling for CPU and GPU was first introduced a couple years ago. I smell a bug.
35) Message boards : Number crunching : New scheduler extremely conservative? (Message 1261364)
Posted 17 Jul 2012 by Profile Borgholio
Post:
I think I found the problem...and it's a big one. It would seem that when BOINC calculates how much work to download, it can't tell the difference between CPU and GPU tasks. Here's how I tested.

1. Set "no new work" on all projects except SETI.
2. Waited for all non-SETI tasks to complete.
3. Watched SETI queue grow to ~400 workunits.

So then I set things back to normal.

1. Resumed work fetch on all projects.
2. Downloaded a couple dozen World Community Grid workunits (~8 hours long)
3. Watched SETI queue drop to ~200 workunits.
4. Downloaded four Climate Prediction workunits (week-long tasks, for those who don't know)
5. Watched SETI queue drop to ~100 workunits.

I am setting "no new work" on all projects except SETI once again to see if my queue grows as before. If it does, then it means that getting a bunch of large CPU tasks would shrink the GPU queue down to only a day or two's worth.
36) Message boards : Number crunching : No work (Message 1261361)
Posted 17 Jul 2012 by Profile Borgholio
Post:
My boxes connect fine I just get "not downloading new work". I'm down to two day's worth for my CUDA cards.
37) Message boards : Number crunching : New scheduler extremely conservative? (Message 1249028)
Posted 21 Jun 2012 by Profile Borgholio
Post:
Yeah my computer can blow through 50 in a day...so something is obviously borked.
38) Message boards : Number crunching : New scheduler extremely conservative? (Message 1248707)
Posted 20 Jun 2012 by Profile Borgholio
Post:
Nope, running for a few days and now I have even LESS work in my queue. :-/
39) Message boards : Number crunching : New scheduler extremely conservative? (Message 1246225)
Posted 15 Jun 2012 by Profile Borgholio
Post:
Installed 7.0.28, will see how it goes over the next day or so. Will lowering the numbers actually get me more WUs in the cache?
40) Message boards : Number crunching : New scheduler extremely conservative? (Message 1246206)
Posted 15 Jun 2012 by Profile Borgholio
Post:
I know there have been some pretty heavy scheduler changes recently, and I expected the number of WUs I have cached to change, but not by as much as it has. It used to be that my old dual-core rig with one CUDA GPU would cache 300 - 400 SETI units at a time...and it would successfully crunch them all in time. Now that I've upgraded my machine to a quad core with two CUDA GPUs, I find I'm only caching 50 or so at a time. I've checked my settings and set minimum and extra work to 10 days but nothing. What's up?


Previous 20 · Next 20


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.