Hi Team,
I am running SETI and Einstein @ home on the GPU with no log issues, I am hitting lag (slow desktop interface response) with the following:

Intel i7 2600K CPU Overclocked to 4.5Ghz, nVidia 580 GTX GPU, 16GB RAM
Boinc version 7.2.33
OS: Ubuntu 12.10 Linux 64bit

With project specs as:

Resource share
Determines the proportion of your computer's resources allocated to this project. Example: if you participate in two BOINC projects with resource shares of 100 and 200, the first will get 1/3 of your resources and the second will get 2/3. 50
Use CPU
Enforced by version 6.10+ no
Use ATI GPU
Enforced by version 6.10+ no
Use NVIDIA GPU
Enforced by version 6.10+ yes
Is it OK for MilkyWay@Home and your team (if any) to email you?
Emails will be sent from boinc@cs.rpi.edu; make sure your spam filter accepts this address. yes
Should MilkyWay@Home show your computers on its web site? yes
Default computer location home
Maximum CPU % for graphics
0 ... 100 25
Run only the selected applications (all applications)
If no work for selected applications is available, accept work from other applications? yes
Frequency (in Hz) that should try to complete individual work chunks. Higher numbers may run slower but will provide a more responsive system. Lower may be faster but more laggy.
default 60 (corresponds to 60 fps) 10000
Period (in milliseconds) to check for scene updates 30
Change how fast the view randomly moves around 5
Make particles bigger or smaller 250
Make particles bigger or smaller when using uglier particles 40

Lag occurs when running Separiation (Modified Fit) 1.28 (opencl_nvidia).
(It may be lagging for other milkyway apps as well, this is just the one running atm causing lag).

I have also tried it with running CPU apps with no apparent difference.

Any ideas would be appreciated.

My best guess is to use an app_info.xml and play with the command line parameters.
The bad thing is that you need to define every app that you want to use in milkyway, not only the modified fit one.

Check this line:
Frequency (in Hz) that should try to complete individual work chunks. Higher numbers may run slower but will provide a more responsive system. Lower may be faster but more laggy.
default 60 (corresponds to 60 fps)
in your MW preferences.

yea, ive tried that, even with figures all the way up to 2000 instead of 60 with no real difference. I may have to try the config file option and play with the settings. Its just annoying that both Seti and Einstein work out of the box without the need to go to the config file route unless one wants to further tinker - not to fix up an issue for something that should be a non-obtrusive background task. Anyway, its tinkering time :)

yea, ive tried that, even with figures all the way up to 2000 instead of 60 with no real difference. I may have to try the config file option and play with the settings. Its just annoying that both Seti and Einstein work out of the box without the need to go to the config file route unless one wants to further tinker - not to fix up an issue for something that should be a non-obtrusive background task. Anyway, its tinkering time :)

You must remember that each project ONLY uses the out of the box Boinc Server side of the software, then build there own applications to use it. Each projects is separate and distinct from every other one. Now a very few do have the same people running them, but they are the teeny tiny ones not those like the ones you are talking about. Having different people in charge and running each project means much different thought processes in how to make things work, and naturally some work much better then other right 'out of the box'. Some projects take great pride in making their project work very well with almost any pc configuration. Others not so much and only try to make it work with the average pc, not those with non standard setups. We crunchers, nd gamers too, tend to push the envelope with our pc's trying to get near the top end of our pc's capabilities, within our own set of limits of course.

And also let's not forget, that 'out of the box' BOINC is not using the PC when it is in use, so any screen lags should not be an issue with standard configuration.

---

Further Update

It seems to be (so far that I have noticed anyway) only on "Separation (Modified Fit) programs/work units. The standard Separation program/work units go through fine without causing any lag.

So I guess the questions should be asked - what is the Modified Fit version doing that the standard Separation isnt doing that could cause a nVidia GTX580 to lag?

BTW I am running Ubuntu linux - so any Windows related issues wont be applicable.

Cheers

Mark

Hi Link

Im not talking about out of the box BOINC (mines been set to crunch while using the PC for years now), I am talking about what is essentially meant to be a low priority background task that causes lag. Ive narrowed it down to I think just one Milkyway program, the trick now will be to find why its causing lag when its "sister program" is not (see my other post)

Cheers

Mark

mw gpu apps are designed to get as much as possible crunching power out of them with only very few calculations done on the cpu. The gpu is stressed far more than with other projects.
The frequency of 60 was choosen because it showed the best balance between max gpu use and a still responsive system. Still some combinations of hardware / software / driver can lead to a less responsive system or lagging. That's where the commandline params come into play. They can although help to increase your gpu use if needed.

Hi Len

The frequency setting does nothing with my set-up, I've had it up to 10,000 instead of 60 (updated the project etc) but it had little to no noticeable affect.

I will look into the command line options once I find where they are listed. The silly thing is i can run the non-modified fit application without lag, so the real question is whats the main difference, processing wise, between the standard and the modified app? That may help work out what may need altering, otherwise im going to playing hit and miss with the options.

Addition: Just noticed that the modified fit program is an OpenCL - nVidia application that is currently in beta, that maybe the issue there.

Cheers

Mark

Hi Len

The frequency setting does nothing with my set-up, I've had it up to 10,000 instead of 60 (updated the project etc) but it had little to no noticeable affect.

I will look into the command line options once I find where they are listed. The silly thing is i can run the non-modified fit application without lag, so the real question is whats the main difference, processing wise, between the standard and the modified app? That may help work out what may need altering, otherwise im going to playing hit and miss with the options.

Addition: Just noticed that the modified fit program is an OpenCL - nVidia application that is currently in beta, that maybe the issue there.

Cheers

Mark

One of your options would be to go onto the website and change your settings to not get the modified-fit units, these are my selections right now under Home:

Run only the selected applications
MilkyWay@Home: yes
MilkyWay@Home N-Body Simulation: no
Milkyway@Home Separation: yes
Milkyway@Home Separation (Modified Fit): yes

I have multiple pc's so use the default, home, school and work settings to make each pc do what I want it to do, not strictly the default settings.

hi Mikey

Thanks for that info which i did before making my initial post, however since the issue is with beta software i wouldnt mind testing it out so that we can help improve it so future versions dont have this issue.