GPU crunching question

[Vyper]

 

Impressive Devaster, good news really.. Page me if u want me to test on Vista x64 and 8800GTX gfx.. It could be fun to see if eveything works and validates..

Kind Regards Vyper

[Devaster]

today i make a validation check and if would be correct i will upload a "technology preview" ....

[Devaster]

sorry i was yesterday too tired (12 hours @work) and only i have do was convert to console app and add some info messages to console so no app today ... as i have wrote i havent too much time for coding 

[Vyper]

No worries dude, it's done when it's done so to speak..

Take your time, the good things is that u let us know the progress of your work instead of sitting in silence and by all means that is very valuable..

Don't exhaust yourself ...

//Vyper

[Devaster]

aka Duke Nukem 4ever ? ? ?
i hope ,not

[Vyper]

Hehe god heavens no!  I wasn't refering to that at all..

Health and well being before charity and sacrifices..

//Vyper

[paydirt76]

[edit: I didn't notice this thread was 8 pages folks, sorry, I'll leave my reply mostly as written though]

Citroja, I understand your excitement.  I actually run a GPU crunching rig for F@H.  If you want to stick with Seti@Home, then just be patient.  F@H has issues trying to run on nVidia cards, but they don't do it through either GPGPU or CUDA.  My advice would be to wait for this app to be released to the community for beta testing and then to follow the beta testing on the forums... before you make a purchase.  There are probably plenty of folks who can test out the cards.  I'd hate to see you spend $500 on a card to crunch with only to turn out that people are unable to get it to crunch up to its potential.

Devaster, I really hope that you can Git R Done!   It would be nice to get a legion of G80 cards crunching for science.

[Devaster]

last week i havent wrote any line of code 
i was under heavy workload (@work,@home)

and in tv is WC of Ice Hockey so ....

i have show wrong progress so i must repair it...

[Devaster]

rapidmind has released a final version of its enviroment so i have downloaded it and install it ...

MOOre better documentation, forums  and so on ....

Aleluja ....

[Devaster]

is FFTLen same for whole computaton

[Josef W. Segur]

is FFTLen same for whole computaton ???

No.

8, 16, 32, 64, 128, 256, 512, 1024, 2048, 4096, 8192, 16384, 32768, 65536, and 131072 lengths are all used.

Technically, some simple changes in the parameters sent in the workunit header could require additional lengths but it's highly unlikely the project will do that.
                                                                                      Joe

[Devaster]

ok.

for fft sizes <128k is a gpu version (GPUFFTW and rapidmind) maximaly uneffective - by time when  driver upload datas , cpu calculates 2- 10 small ffts ...

also seti uses fft lower <128k only so i think that any development for seti gpu client is lost of time for now, until whole analyse (fft,powerspectrum,pulses,triplets and so on) is on gpu. but there is a problem on older cards with shaders size.

so for now i have stopped a development of this client    ...

[Vyper]

Isn't gpufftw and rapidmind parallell utilized?? If so a WU can be arranged in chunks and u can upload a datastream as a whole block to the gfxram and then let multiple "threads" calculate that particular area..

That would speed up the process alot.. If that isn't doable then ofcourse it't no use using either API..

Nv Cuda (which u unfortunately can't test) have 12-16 "processors" that can execute multiple threads each.. If it were possible to move one block of data once and then chop it up into small different pieces and try to align ram reads/writes then the client would be highly effecient.. The problem is that the Cuda API atm can only accept FFT sizes up to 16K and that would require either alot of datashuffling or a complete custom FFT library for that particular API and different gfx setups = "Verrrry time consuming" , but if someone is really keen of low level programming and implementation can surely write a massive parallell FFT routine ..

Well i hope that there is another approach in implementing a S@H client to the GPU world, it would surely be a shame if only Folding@home is going away with having that accomplished, but then again i don't know what type of calculations that client want . The only thing i know is that good o'l ATI X19XX series card has huge increased throughput and even PS3 client has strugles keeping up with that gfxcards FPU power..

So all in all im hoping for a parallell execution routine to accomplish that goal now atm...

Devaster, u have my deepest respect for even trying the 1st step to GPU S@H with ur reported progress, hope that the API gets better as times goes by..

Kind Regards Vyper

[Simon]

Same from me - sad to see you stop development for the time being. Maybe Vyper's comments gave you some new ideas, hopefully anyway

Take care, and stay around!

Regards,
Simon.

[Devaster]

so here is a technology preview...
DO NOT USE IT FOR OFFICIAL CRUNCHING
stop boinc unpack run exe with parameter -standalone and see how is it running ...
if dont running then send stderr.txt to this forum, but in 90 % of cases is missing some OpenGL extension ...

to development : i am tired and i need some pause and thinking about it so dont be worry ...

Ill be back ! 

[attachment deleted by admin]