Posts by AllenIN


log in
1) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1763642)
Posted 2 days ago by Profile AllenIN
-use_sleep -unroll 10 -oclFFT_plan 256 16 512 -ffa_block 6144 -ffa_block_fetch 1536


Jim try these. Got this from the old 750Ti thread

http://setiathome.berkeley.edu/forum_thread.php?id=75919#1588675

this as the last recommendation on a 750Ti

-use_sleep -unroll 6 -oclfft_plan 256 16 256 -ffa_block 2830 -ffa_block_fetch 2830 -tune 1 64 4 1 -tune 2 64 4 1

See if it helps


I was just wondering, would the same commands work on a GTX650ti or are they specific to only the 750?


Try -use_sleep -unroll 8 -oclfft_plan 256 16 256 -ffa_block 2830 -ffa_block_fetch 2830 -tune 1 64 4 1 -tune 2 64 4 1.


Thanks Mike, I'll give it a try.
2) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1763351)
Posted 4 days ago by Profile AllenIN
-use_sleep -unroll 10 -oclFFT_plan 256 16 512 -ffa_block 6144 -ffa_block_fetch 1536


Jim try these. Got this from the old 750Ti thread

http://setiathome.berkeley.edu/forum_thread.php?id=75919#1588675

this as the last recommendation on a 750Ti

-use_sleep -unroll 6 -oclfft_plan 256 16 256 -ffa_block 2830 -ffa_block_fetch 2830 -tune 1 64 4 1 -tune 2 64 4 1

See if it helps


I was just wondering, would the same commands work on a GTX650ti or are they specific to only the 750?
3) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1763019)
Posted 5 days ago by Profile AllenIN
No way to do that by web Settings (unless you edit them all the time) or app_config.xml

It is done by app_info.xml
Since you already use Lunatics - it is a matter of selecting only apps you want to run.

If you now don't have AstroPulse tasks for CPU you can do that right now.
Re-run Lunatics installer and select e.g.: for CPU only SETI@home v8 + for GPU only AstroPulse v7
(or any other combination you want)

!!! Anything not selected but you have tasks now for it - tasks will be Deleted without warning by BOINC !


I had a feeling that I needed to change something in the app_info file, but I had not thought about using the Lunatics installer to make the changes. That will make it very easy.

Thanks BilBg!
4) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1762323)
Posted 7 days ago by Profile AllenIN
As part of those settings, check or unchecked the "Allow work for CPU" box. (Keeping in mind that if you're using the Work/Home/School option as a way to preset the combination there's a limited amount you can do without having to edit before every work fetch. Otherwise it would be easier to use app_config.xml.)


Exactly! I just don't know what to put in the app_config.xml to make it use only a gpu or a cpu for a given task. Can you help with that?

Thanks
5) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1762319)
Posted 7 days ago by Profile AllenIN
Yes. Set up Home to not run AP, and Work to allow AP. Then set up each computer to use Home or Work for preferences as desired.



Ahh yes, I see. Nice solution to a otherwise tough problem. Now I see how that would allow you to run or not run specific types, but is there a way to just run a type of wu to a specific cpu or gpu only, on the same machine? Say I want to run AP's only on GPU and not on CPU.

Thanks a bunch.
6) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1762299)
Posted 7 days ago by Profile AllenIN
What has changed is that the rate of consumption of "tapes" by MB v8 is lower than that of v7. Thus the rate of "tapes" being loaded for splitting is lower, and so the rate of AP production is lower.

I think there are also more people that set their computers up to accept AP work when the opti v8 MB app was not ready, and they are still doing some AP now.
So when AP does split, there are more hungry mouths to feed.


Is it possible to not get AP's on certain computers, when you have a few computers and don't want to run them on some? Is it possible to get them only on GPU and not on CPU?

If so, how does one go about this?
7) Message boards : Number crunching : Anything relating to AstroPulse tasks (Message 1762054)
Posted 8 days ago by Profile AllenIN
Looking for a little help here....I have an astropulse_v7 7.3 (sse) currently running on one of the two cores of my cpu. It's just a 2.6Ghz machine, but it still seems to be taking an awfully long time to complete this chore. It is 34.628% done and it has taken 16:19:10 so far. At that rate I expect it to be done in a total of 48 hours. Is that nuts or am I mistaken?

Thanks!
8) Message boards : Number crunching : Notes on Nvidia driver bloat [361.60 Hotfix] (Message 1761174)
Posted 11 days ago by Profile AllenIN
Jason,

That's what screwed me up on my 650ti rig, that damned "hotfix". It locked up my system too and I finally shut it down after a night of hammering the HDD. When I fired it back up it had no driver at all so I reinstalled the version that I had on it before the attempted hotfix and all is well now.

I have wondered many a time about taking the chance doing MS updates and wondered if I really should. I always do, but that may end soon too.

You know, at any time MS could sabotage all of our machines by putting a time bomb in one of their updates. I know it's not likely, but who knows for sure?

Thanks for sharing your experience!

Allen
9) Message boards : Number crunching : Forced to abandon GPU processing of SETI work (Message 1761171)
Posted 11 days ago by Profile AllenIN
OK - so not 100% fantastic. I've had two SETI@Home v8 GPU units stall in the last four days. I'm not sure what they have in common, but I suspect they're the ones that are scheduled to take the longest times to run.

However, I can live with that.

Would increasing the number of iterations in AllenIN's tweak (above) be likely to help - and if so, why?


Graham,

I haven't had any errors at all since setting my APU's, both A8 3870K's, to these settings.........

-sbs 256 -cpu_lock
-period_iterations_num 450

If you have problems at 450, then I would try something higher.

Good luck!
10) Message boards : Number crunching : Lunatics Windows Installer v0.44 - new release for v8 (required upgrade) (Message 1760162)
Posted 14 days ago by Profile AllenIN
Thanks for checking it out for me, but I must have misspoken or didn't make myself understood (not rare for me) but what I had done was to upgrade the driver in the first place to the latest version and everything when South.

When you have a driver that works why would you "upgrade"? I've learned the hard way many times if it's not broken don't fix it.


Good question. I guess I just thought that perhaps they had made some improvements in the code and it might perform better. I guess I've learned my lesson.
11) Message boards : Number crunching : Lunatics Windows Installer v0.44 - new release for v8 (required upgrade) (Message 1760141)
Posted 14 days ago by Profile AllenIN
I was just investigating my last failed unit on the same system...650ti...and it said that it had 8 spikes and 7 triplets, but bad computation. One of the completed okay units done by someone else said 8 spikes and 7 triplets but it finished fine. How does this work? If we both found the same thing, why is one bad and one okay? Just trying to learn.

Allen
12) Message boards : Number crunching : Lunatics Windows Installer v0.44 - new release for v8 (required upgrade) (Message 1760129)
Posted 14 days ago by Profile AllenIN
could it be because you were running cuda 5.0 on a 650ti that has a compute capability of 3.0

Could be.

"CUDA driver version is insufficient for CUDA runtime version."

From task 4681187393:

setiathome_CUDA: CUDA Device 1 specified, checking...
Device cannot be used
Cuda device initialisation retry 1 of 6, waiting 5 secs...
Cuda error 'Couldn't get cuda device count
' in file 'C:/[Projects]/__Sources/sah_v7_opt/Xbranch/client/cuda/cudaAcceleration.cu' in line 138 : CUDA driver version is insufficient for CUDA runtime version.
setiathome_CUDA: cudaGetDeviceCount() call failed.
setiathome_CUDA: No CUDA devices found
setiathome_CUDA: Found 0 CUDA device(s):
nVidia Driver Version {blank}

This is what it looks like now you've gone back up to a more recent driver:

v8 task detected
setiathome_CUDA: Found 1 CUDA device(s):
nVidia Driver Version 358.91
Device 1: GeForce GTX 650 Ti, 1024 MiB, regsPerBlock 65536
computeCap 3.0, multiProcs 4
pciBusID = 2, pciSlotID = 0
In cudaAcc_initializeDevice(): Boinc passed DevPref 1
setiathome_CUDA: CUDA Device 1 specified, checking...
Device 1: GeForce GTX 650 Ti is okay
SETI@home using CUDA accelerated device GeForce GTX 650 Ti
pulsefind: blocks per SM 4 (Fermi or newer default)
pulsefind: periods per launch 100 (default)
Priority of process set to BELOW_NORMAL (default) successfully
Priority of worker thread set successfully

setiathome enhanced x41zi (baseline v8), Cuda 5.00

That's a Kepler core (GK106), so cuda50 should be a good choice - stick with that, and see how it goes. Don't ever drop to a driver earlier than 306.23 with that application.


Thanks for checking it out for me, but I must have misspoken or didn't make myself understood (not rare for me) but what I had done was to upgrade the driver in the first place to the latest version and everything when South. I believe this is the same driver that I had before everything went wrong....when I tried to upgrade...ugh!

Thanks, I will keep an eye on it and get back to you if work keeps getting dumped into the abyss.

Thanks a bunch Richard. Btw, this card was running v7 version with Cuda 50 forever.
13) Message boards : Number crunching : Lunatics Windows Installer v0.44 - new release for v8 (required upgrade) (Message 1760091)
Posted 14 days ago by Profile AllenIN
Hey Richard,

Suddenly I'm having trouble with one of my systems. It's 6260596. It is throwing GPU work in the scraper on a regular basis. I don't know if it's something I've done (more than likely) or just a fluke. I was hoping you might take a look and see if you can tell me what's happening.

I'm using your software and the latest Boinc. It was running fine with 2 GPU units running at one time and then I saw it was stumbling and I changed the app_config to only run one at a time.

I should tell you that I tried to upgrade the driver on the nVidia card and the whole system locked up and I had to shut it down and reboot. When it came up it said there was no driver, so I went back to an older driver and then everything appeared to be okay until I checked it the next day and it was acting up again.

I'd sure appreciate you taking a look see and letting me know how I screwed it up.

Thanks a bunch,

Allen
14) Message boards : Number crunching : Forced to abandon GPU processing of SETI work (Message 1759096)
Posted 18 days ago by Profile AllenIN
When SETI@home v8 units were first released I hoped that the problems I'd had with v7 units and my ATI GPU might go away.

With longer work units (those estimated to take around 2 hours on my GPU, as opposed to the shorter 45-minute units which were fine, the unit would complete a certain percentage of its work, often around 20%, then cause the display driver to stop working. The display driver would recover, but processing on the unit would then stop (i.e. the percentage of work done would remain the same while both the elapsed time and the remaining time would increase).

When I discussed this on the SETI message board a while back, several people made helpful suggestions, typically involving tweaking my config file. Thanks to everyone who made these suggestions, but alas nothing worked.

So, I hoped things would change with v8, but they haven't. Longer units still stop at around 20% completed. I haven't received any shorter units to compare yet. I suspect they'd be OK, and Astropulse units - on the rare occasions when I get any - are also fine.

However, in the absence of an option to have only Astropulse and/or shorter SETI@home v8 units processed on the GPU, I've decided to allow no SETI units to run on the GPU rather than having to check which ones to abort several times a day. It means I'll run a lot fewer SETI units than before, but there are plenty of other machines out there so it's OK.

I don't expect anyone to try to fix the problem, because I doubt whether enough people have the same GPU to make it worthwhile.

Here are the details of my setup, in case anyone else has the same problem:

23/01/2016 18:56:21 | | CAL: ATI GPU 0: AMD Radeon HD 6520G/6530D/6550D/6620G (SuperSumo) (CAL version 1.4.1848, 512MB, 479MB available, 960 GFLOPS peak)
23/01/2016 18:56:21 | | OpenCL: AMD/ATI GPU 0: AMD Radeon HD 6520G/6530D/6550D/6620G (SuperSumo) (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 512MB, 479MB available, 960 GFLOPS peak)
23/01/2016 18:56:21 | | OpenCL CPU: AMD A8-3850 APU with Radeon(tm) HD Graphics (OpenCL driver vendor: Advanced Micro Devices, Inc., driver version 1800.11 (sse2), device version OpenCL 1.2 AMD-APP (1800.11))

Graham,

I had the same problem on my APU and I had a little help from the people here and it fixed my problem. Maybe yours can be fixed as mine was.
In your, mb_cmdline_win_x86_SSE2_OpenCL_ATi_HD5.txt, located in your seti directory, add the lines below, save and then quit and restart Boinc. It is a text file and you can open it with Notepad.

-sbs 256 -cpu_lock
-period_iterations_num 450

You may be able to reduce the 450 number, but on mine it didn't really take until I hit 450. I started with 256 as I recall and it was better but still erred.

I wish you success!
Allen
15) Message boards : Number crunching : Seti@home v8 (Message 1758956)
Posted 18 days ago by Profile AllenIN
Jeff,

Just received 3 ap's on one of my other rigs. This one is running an apu and it automatically shut down one cpu core and is running two of the ap's on the gpu. I'm a bit surprised as I didn't believe the apu/gpu could handle that kind of load. It says I'm at 55c which is in the good range for this system. Still, I see all 4 cores of the cpu working, but only 3 cpu wu's running and 2 gpu. Very interesting as I've never seen this before.
16) Message boards : Number crunching : Seti@home v8 (Message 1758954)
Posted 18 days ago by Profile AllenIN
BTW, I see I have some ap's coming up and they are running on the 730. The config file has the gpu and the cpu usage both at .5, which is different than the non ap units at .5 on gpu and .04 on cpu. Will this be a problem when it goes to run the ap's? I noticed that they are opencl nvida 100 too.

Personally, I prefer not to run 2 APs at a time on a single GPU, although 1 AP plus 1 MB is usually fine. To allow for either 2 MBs at a time or 1 AP plus 1 MB, change your GPU usage to 0.51 for Astropulse tasks and 0.49 for the MultiBeam (setiathome_v8 and _v7).

As far as CPU usage goes, there are strong feelings around here that a full core should be reserved for each AP task. I tend not to subscribe fully to that philosophy, but on my multi-GPU systems, where an AP might be running on each of my GPUs at the same time, I use that 0.5 figure for AP CPU usage, effectively reserving a full core only when at least 2 APs are running. The CPU usage value only comes into play when the total value of all running tasks equals or exceeds an integer amount (1, 2, 3, etc.). In your current setup, you'd only reach that threshold if you did, in fact, let 2 APs run at the same time.

If you do choose to reserve a full core for AP, simply change your cpu_usage value to 1.0. What you'll notice, in that case, is that one of your running CPU tasks will drop into a "Waiting to run" state, and will remain that way either until the AP task or another CPU task finishes, at which point it will automatically resume.


Sounds like very good advice. I will reconfigure my gpu settings for the ap's. If that works, great, if not, I will go with the full cpu for the ap's. I would have never figured this out for myself.
17) Message boards : Number crunching : Seti@home v8 (Message 1758938)
Posted 19 days ago by Profile AllenIN
Allen, I see both your 730 and 650 have cuda 32/42/50 tasks, So just leave it alone for 24 hours and see what application BOINC settles in with and wants to run.

Will do. I noticed the same and was wondering why I had all of them since I set for the 50's in the install.

Thanks!

Your task list is just showing you what the tasks were originally assigned when they were downloaded as "stock", before you switched to Lunatics. If you look at the Stderr for any that have run since you switched, you'll find they were run under "setiathome enhanced x41zi (baseline v8), Cuda 5.00", in other words, Cuda50, regardless of the original assignment.


Once again I'm amazed. I guess I've just not spent enough time devouring all the technical information about the software side of SETI. I really appreciate all of you taking the time to explain the intricacies of the software.

I guess I'm not as dumb as I had thought though, since I had surmised the possibility that the d/led wu's were d/led prior to installing Lunatics software. I really am grateful to Lunatics as well...they make it so damned easy to set up.
18) Message boards : Number crunching : Seti@home v8 (Message 1758934)
Posted 19 days ago by Profile AllenIN
Allen, I see both your 730 and 650 have cuda 32/42/50 tasks, So just leave it alone for 24 hours and see what application BOINC settles in with and wants to run.

Will do. I noticed the same and was wondering why I had all of them since I set for the 50's in the install.

Thanks!
19) Message boards : Number crunching : Seti@home v8 (Message 1758930)
Posted 19 days ago by Profile AllenIN
BTW, I see I have some ap's coming up and they are running on the 730. The config file has the gpu and the cpu usage both at .5, which is different than the non ap units at .5 on gpu and .04 on cpu. Will this be a problem when it goes to run the ap's? I noticed that they are opencl nvida 100 too.
20) Message boards : Number crunching : Seti@home v8 (Message 1758927)
Posted 19 days ago by Profile AllenIN
This is really interesting....at least to me....the 730 is using about 80-90 percent of the gpu and the 650 is using about 50-70 percent. I would have thought that they both would be nearly maxxed out. They both do bounce around a lot, but on average, that's about what they are using. I do realize that the 650ti has more cuda cores than the 730.

The 730 is much older hardware, and the software is more optimised for it. The 650Ti is more recent & more powerful, but as the software hasn't been optimised specifically to take advantage of it, the load on it is much less.

When you ran the Lunatics installer, what version of CUDA did you select for each card?


From the Lunatics Windows Installer v0.44 thread,

Rule of thumb selection guide:
Cuda 2.3: PreFermi - NOT to be used on Fermi/Kepler/Maxwell
Cuda 3.2: mixed systems
Cuda 4.2: Fermi
Cuda 5.0: Kepler/Maxwell

I selected Cuda 5.0 on both cards.


Next 20

Copyright © 2016 University of California