Advanced search

Message boards : Wish list : ATI Cards?

Author Message
ishmoo21
Send message
Joined: 12 Aug 08
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 1443 - Posted: 12 Aug 2008 | 5:09:29 UTC
Last modified: 12 Aug 2008 | 5:09:46 UTC

the 3xxx and 4xxx series are cheaper and run faster than the PS3, will they ever be supported?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 1497 - Posted: 17 Aug 2008 | 14:25:39 UTC - in response to Message 1443.
Last modified: 27 Nov 2008 | 19:16:29 UTC

the 3xxx and 4xxx series are cheaper and run faster than the PS3, will they ever be supported?


Unlikely, for sure not before the end of next year.
gdf

Barraud Denis
Avatar
Send message
Joined: 2 Sep 08
Posts: 15
Credit: 36,207,656
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 2023 - Posted: 2 Sep 2008 | 14:47:38 UTC - in response to Message 1497.

the 3xxx and 4xxx series are cheaper and run faster than the PS3, will they ever be supported?


Unlikely, for sure not in the next year.
gdf



So See folding@home ON http://folding.stanford.edu
for the moment they use a 32bits process to crunch with some ATI card,
i use it on my pc: q9550 / 2Gb DDR2-8500 / HD4850 512Mb and at the same time
boinc 6.2.18 (in a service), but folding use 1 of my 4 cores and one core is used for 2 WU of boinc in same time ..

Ageless
Avatar
Send message
Joined: 22 Aug 08
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 2111 - Posted: 5 Sep 2008 | 23:45:28 UTC

Worse yet. I just went and downloaded the new Catalyst drivers 8.8 for my HD3850. In the release notes (https://a248.e.akamai.net/f/674/9206/0/www2.ati.com/gamesite/Catalyst_88_release_notes.html) I found this:

ATI's Folding@Home

Folding@Home is a distributed computing project designed by the Stanford University. The application performs intensive simulations of protein folding. This simulation will help researchers uncover how certain diseases develop. Folding@Home uses distributed computing to simulate protein folding, the workload is broken up into small work units and distributed across hundreds of thousands of computers over the internet. You can help find the cure to many different diseases! To join Folding@Home, follow these steps:

1. Download ATI's Catalyst™ software suite
2. Download the Folding@Home GPU client application
Note: The latest GPU client supports the ATI Radeon™ HD 3xx0 series of products only.
3. Enter the ATI team number 51394 and start folding!

For more information on Folding@Home visit: http://ati.amd.com/technology/streamcomputing/folding.html


It's a shame. (Or doesn't that worry you?)
____________
Jord.

BOINC FAQ Service

Barraud Denis
Avatar
Send message
Joined: 2 Sep 08
Posts: 15
Credit: 36,207,656
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 2219 - Posted: 9 Sep 2008 | 15:07:03 UTC

on my P5E-Q9550-Win64-4 Gb DDR2-8500 / boinc 6.2.18-HD4850 512Mb catalyst 8.8, folding :

- gpu no success, but i will try soon under boinc 6.3.10

- cpu ok, but it take for one of my four core, and in boincmgr 2 wu on 4 core are used half on the same core.

i notice too the ccc=catalyst control center) and viewer failed sometime.

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 2669 - Posted: 30 Sep 2008 | 14:16:50 UTC

Half of the world is using ATI graphic cards. Isn't it worth to use that computing power also?? Guys from FAH already made it. I would realy like to participate in Your project but with my ATI card i am unable to do it. So i stick to the BOINC and wait for You guys to develop ATI client

Pepo
Avatar
Send message
Joined: 25 Dec 07
Posts: 19
Credit: 53,126
RAC: 0
Level

Scientific publications
watwat
Message 2713 - Posted: 3 Oct 2008 | 8:03:33 UTC

Maybe this email from ATI might interst you (forwarded yesterday to [email protected] by Dr. Anderson):

-------- Original Message --------
Subject: Working with SETI@home development team to create a ATI GPU (graphics processor unit) accelerated version of your client (plus any other BOINC projects :-)
Date: Wed, 1 Oct 2008 16:48:34 -0400
From: Dodd, Andrew <[email protected]>

...

ATI (now AMD) is now in a great position to work will a number of distributed computing projects that could benefit greatly from GPU acceleration.

We now have an SDK that will significantly help anyone trying to use the GPU for stream calculation.

I've attached a link to the AMD Stream website (which includes a link to the SDK), as well the AMD Stream user guide to give a quick over-view of things.

http://ati.amd.com/technology/streamcomputing/

We'd love to work with SETI@home or any other project that uses BOINC (if you could forward this to [email protected], it would be greatly appreciated as well).

Thanks very much for your time,

Andrew


Peter

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 3071 - Posted: 16 Oct 2008 | 10:03:23 UTC

Any status update on this case guys? There are milions of ATI cards out there just waiting to be used.

Profile Stefan Ledwina
Avatar
Send message
Joined: 16 Jul 07
Posts: 464
Credit: 215,281,659
RAC: 4,094,182
Level
Leu
Scientific publications
watwatwatwatwatwatwatwat
Message 3072 - Posted: 16 Oct 2008 | 11:42:24 UTC

GDF the project admin already said that it is unlikely, and for sure not in the next year. (See the 2nd post in this thread)

The CUDA app is still in beta and we try a few things. Also the CUDA capable BOINC clients still have some bugs that needs more work from the devs.

It doesn't help if someone asks every few weeks... ;)


____________

pixelicious.at - my little photoblog

jordi1024
Send message
Joined: 2 Nov 08
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 3608 - Posted: 2 Nov 2008 | 18:47:25 UTC - in response to Message 2669.

I'm also willing to use my ATI to join your project. You can count on me if the port is ever done. I'll keep an eye just in case. Meanwhile, i'll keep on crunching WU on IBERCIVIS and folding@home...

jordi

Quadnator
Send message
Joined: 13 Nov 08
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 3838 - Posted: 15 Nov 2008 | 19:17:15 UTC

hello

ATI Stream comes on 10 December 2008 with Catalyst 8.12
http://www.computerbase.de/news/hardware/grafikkarten/ati/2008/november/ati_stream_jedermann_dezember/
In English:
http://translate.google.de/translate?u=http%3A%2F%2Fwww.computerbase.de%2Fnews%2Fhardware%2Fgrafikkarten%2Fati%2F2008%2Fnovember%2Fati_stream_jedermann_dezember%2F&hl=de&ie=UTF-8&sl=de&tl=en

can you Support it? I will Crunch with my HD4870!

Thanks

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 4059 - Posted: 27 Nov 2008 | 19:17:59 UTC - in response to Message 3838.

We will not support ATI for the time being.

gdf

Leoman
Send message
Joined: 12 Mar 09
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 7388 - Posted: 12 Mar 2009 | 17:36:55 UTC

Good news to the ATI fans!!

The next version of BOINC (6.8) they will add support to the ATI GPUs. They put it on the Roadmap with this Note: "The turn around for this release is expected to be pretty quick". So.. we need to wait a little more!

http://boinc.berkeley.edu/trac/roadmap

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 7391 - Posted: 12 Mar 2009 | 18:17:05 UTC

The only project with a known ATI application right now is Milky Way. And then only for the 38x0 and 48x0 cards. MW may be doing an OpenCL or CUDA version of their application also. The ATI application is not that well integrated and there are issues with running it right now.

Einstein@Home has indicated that they have a GPU application in the works, but the information is more confused as to which version will be first on the block. The way I read the post they have several of the APIs in development with maybe the CUDA application in the lead.

If and when OpenCL matures and we get a couple applications done with it, well, then we will know if it is better to build native or use the more portable API ...

Boosted
Send message
Joined: 25 Dec 07
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 8040 - Posted: 1 Apr 2009 | 18:28:29 UTC

I just do not see why there is only one project that runs the ATI structure. It really seems a waste of time to exclude a significant amount of people from using very powerful cards to do work. This is not just aimed at GPU grid but the projects in general.
I do run MW@H now, but quite frankly am getting sick of it because there is hardly any work. I consider myself lucky if I get a total or 20-30 minutes running time per day.
I am fully aware that F@H also runs the ATI apps, but I have issues with how F@H is run and how they portray themselves to the other projects and even their members. They have been considered the 'bastard exiled stepson' of boinc for many years and I refuse to support them.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 8055 - Posted: 2 Apr 2009 | 9:46:06 UTC - in response to Message 8040.

The reason that no other project is using ATI card at the moment is that they are not ready. You can port only very simple applications or you have to let them do the port for you. Their hardware is moving towards better computation, but it still not there. It's not just peak performance, you need to be able to use those flops in a sensible way. For ATI, we need to wait for OpenCL and see how they perform then.

gdf

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 9430 - Posted: 7 May 2009 | 11:19:55 UTC - in response to Message 8055.

Guys from F@H did it. Guys from MilkyWay managed to do it. They didn't wait for OpenCL they just took advantage of the fact that half of the people is using ATI gpu based cards, and theese cards are just running idle, just waiting for some work. Just look at the statistics fo MilkyWay computing power. And they are not able to provide the WUs for all those ATI cards. So there is plenty of room for other projects. Folding proteins on ATI is possible and the F@H proved it. So there is no excuse like:

"The reason that no other project is using ATI card at the moment is that they are not ready. You can port only very simple applications or you have to let them do the port for you. Their hardware is moving towards better computation, but it still not there. It's not just peak performance, you need to be able to use those flops in a sensible way. For ATI, we need to wait for OpenCL and see how they perform then."

The folding simulation is quite complexed one and as seen in F@H, ATI cards are capable of running such app without OpenCL. If F@H could do that, why can't You?? Maybe You should cooperate a little bit closer with guys from ATI?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 9466 - Posted: 8 May 2009 | 10:05:00 UTC - in response to Message 9430.
Last modified: 8 May 2009 | 10:06:45 UTC

We contacted ATI. We are waiting for a decision, long due now.
For MW is a very simple application, for F@H ATI wrote the code, as far as I know.
gdf

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 10187 - Posted: 26 May 2009 | 13:27:41 UTC

Any news Guys? Pepole with ATI cards are crunching MW Wu's like crazy and there is a lot of computing power still left. For CUDA there are several other projects so there will be more competition for CPU cycles (recently AQUA joined with GPU app)

uBronan
Avatar
Send message
Joined: 1 Feb 09
Posts: 139
Credit: 575,023
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 10249 - Posted: 28 May 2009 | 10:27:48 UTC
Last modified: 28 May 2009 | 10:29:06 UTC

Guys stop asking for support for ati cards.
Just accept the choice from the project admins not all projects benefit from the ati cards.
We just need to be patient a bit more since more and more people want to contribute with their hardware, i just hope every project admin takes the time to see if its possible to do so on another brand of gpu.

If they don't then accept it and move on, lets show some respect to their choice.

If the choice is made we make the choice which we are going to help out, remember we are the ones paying the electric bill not them ;)
But i accept the choice made from every project admin
The only project which need to be pushed towards gpu work is CPDN ;), those guys simply to lazy to rewrite their crappy old fortran code ;)

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 10462 - Posted: 9 Jun 2009 | 13:43:11 UTC - in response to Message 10249.

Rewriting the code should apply to all projects (since inefficient code couses us to waste the money on electricity bill instead of upgrading our machines). I just see that more projects are about to use CUDA (AQUA@home recently) and theire giving better deadlines for WU (my card is unable to meet the deadline for GPUGrid). So there will be a lot of competition between projects for the computing power. And the GPUGrid acts like it doesn't care. Right now the only project capable of using ATI cards is MilkyWay (F@H is not a boinc project) and You can see how much they gained. I just don't understand the choices of GPUGrid (and as i do not understand, it is hard for me to accept). It seems that GPUGrid is for the people with fast/new nvidia cards. So the owners of the slower nvidia cards and owners of ATI cards are not the target for GPUGrid...

On the below graph you can see the RAC of the GPUGrid



You cna see the point when SETI and AQUA launched their CUDA apps :D

I really would like to perform some work for GPUGrid as i think this project is more important than searching for the ETI or simulating the quantum computer. But unfortunately the project is beyond my limits (i am now finishing the first WU for AQUA CUDA and it is mainly because my card is good enough to meet the deadline, which is a month or so, not a week like it does in GPUGrid).

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 10464 - Posted: 9 Jun 2009 | 17:57:57 UTC - in response to Message 10462.

Hi,
don't get us wrong, we do care, but unfortunately our complex application is not portable on ATI with their current software support.

gdf

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 10499 - Posted: 12 Jun 2009 | 19:43:28 UTC - in response to Message 10464.

So maybe You should make a petition to ATI guys :D i would gladly sign it :D

Profile Sandro
Send message
Joined: 19 Aug 08
Posts: 22
Credit: 3,660,304
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 10575 - Posted: 14 Jun 2009 | 22:10:38 UTC - in response to Message 10464.

Hi,
don't get us wrong, we do care, but unfortunately our complex application is not portable on ATI with their current software support.

gdf

Give Gipsel aka Cluster Physik, the creater of the MW_ATI-apl a try.
Send him the Code, he can than have a look if it is possible to port the code with Brooke/CAL to ATI cards.
Im shure he will handle the code secretly, so no 3rd person will see the code.
Just give him a try.

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 11127 - Posted: 12 Jul 2009 | 20:59:01 UTC - in response to Message 10575.

Gipsel helped them :D http://boinc.thesonntags.com/collatz/ and since you guys don't need double precission (like MW does) maybe You can benefit from ATI cards. Since there are alreday 5 projects using CUDA and only two using ATI. It will be more compettion for you on the CUDA power :D You really should consider creating ATI app...

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 11500 - Posted: 30 Jul 2009 | 11:42:49 UTC - in response to Message 11127.

any sign of the petition to the ATI guys??

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 12312 - Posted: 3 Sep 2009 | 14:41:55 UTC - in response to Message 11500.

Is anything moving in that topic? Do You guys even try to create ATI app??

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 12313 - Posted: 3 Sep 2009 | 14:49:42 UTC - in response to Message 12312.

There is no sign of an OpenCL for GPUs yet on ATI.

gdf

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12546 - Posted: 18 Sep 2009 | 14:21:27 UTC - in response to Message 12313.

http://boinc.berkeley.edu/trac/milestone/6.10

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 755,370,933
RAC: 212,472
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12885 - Posted: 29 Sep 2009 | 5:08:06 UTC - in response to Message 12546.

Do any of you asking for an ATI version want to contribute enough of a donation to allow hiring a computer programmer already specializing in whatever computer languages are already available for writing ATI card programs? Or would you prefer to wait for ATI/AMD to offer suitable compilers to allow using more familiar computer languages to write programs for ATI cards?

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 755,370,933
RAC: 212,472
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12886 - Posted: 29 Sep 2009 | 5:19:14 UTC - in response to Message 12546.

http://boinc.berkeley.edu/trac/milestone/6.10


I see no sign that includes any compilers for producing programs to run on ATI cards, only some support for running such programs once they are ready.

MarkJ
Volunteer moderator
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 738
Credit: 200,909,904
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12905 - Posted: 29 Sep 2009 | 11:40:38 UTC - in response to Message 12546.

http://boinc.berkeley.edu/trac/milestone/6.10


6.10.x adds CAL/Brook+ support so that BOINC can use ATI cards. It doesn't add OpenCL support which is the longer term aim.

OpenCL is supposedly supported by ATI and Nvidia so the code is supposed to be compatible. My understanding is they will need to compile a version for each brand of card, but the source code doesn't need to change. The means that GPUgrid could develop a single version instead of having a CUDA version and a CAL/Brook+ version.

Unfortunately OpenCL isn't mature and the development is well behind what CUDA offers. CAL/Brook+ is also well behind CUDA in development. There is much postulating between the vendors about who's is better and who is supporting the OpenCL standard.
____________
BOINC blog

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 12907 - Posted: 29 Sep 2009 | 12:45:54 UTC - in response to Message 12905.
Last modified: 29 Sep 2009 | 13:10:29 UTC

We will support OpenCL. The day that ATI provides a working OpenCL implementation, we will be there.
gdf

Tom Philippart
Send message
Joined: 12 Feb 09
Posts: 57
Credit: 23,376,686
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 12939 - Posted: 30 Sep 2009 | 19:09:23 UTC - in response to Message 12907.

Great news, I have 3 opencl capable ati cards waiting :D

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 755,370,933
RAC: 212,472
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 13100 - Posted: 10 Oct 2009 | 3:47:17 UTC - in response to Message 10249.
Last modified: 10 Oct 2009 | 3:49:53 UTC

The only project which need to be pushed towards gpu work is CPDN ;), those guys simply to lazy to rewrite their crappy old fortran code ;)


You might tell CPDN that Nvidia has announced that during the next few weeks, they will start providing a Fortran compiler for at least some of their chips, which should make the task of converting it to a GPU version much easier.

However, Nvidia does not seem to have gone to much effort to make it clear whether this compiler will work for the chips they're already selling, only the new 300 series of their chips they also plan to start making available in the next few weeks.

They also plan to start making some other compilers available at the same time, which may be of more interest to projects already using a few other computer languages, including C and C++. Same unclearness about whether those will work with the chips they've already sold as well.

One problem to expect with any project that currently has high memory requirements, though: Don't expect them to be able to get a full speedup by using all the GPU cores; only as many as the graphics card has memory for, at least until they do a significant rewrite to let many of the remaining GPU cores do something useful with much less memory.

For similar reasons, don't expect very many BOINC projects to start making both Nvidia GPU versions and ATI GPU versions available until some time after both Nvidia and ATI start making enough compilers available that the same computer language can be used to write both versions.

Also, expect some BOINC projects with algorithms with enough things that need to be done one after the other to start offering GPU versions that essentially run multiples of the workunits now offered, each on a small cluster of the available GPU cores, with similar restrictions on whether there's enough GPU memory to run very many of these CPU clusters at once.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 13108 - Posted: 10 Oct 2009 | 8:02:43 UTC - in response to Message 13100.

I think that this is a misconception probably given by the marketing people.
My opinion of what they mean by giving a C++ or fortran compiler is that they will support CUDA development in C++ (transparently) and fortran. Now this is only possible by using C within fortran or C++.

So good, but nothing really exceptionally different from before.

gfd

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 755,370,933
RAC: 212,472
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 13155 - Posted: 13 Oct 2009 | 0:57:05 UTC - in response to Message 10249.
Last modified: 13 Oct 2009 | 1:09:22 UTC

The only project which need to be pushed towards gpu work is CPDN ;), those guys simply to lazy to rewrite their crappy old fortran code ;)


I took a look at the CPDN project and found the following:

1. Their workunits are so large that I doubt that many Nvidia graphics cards would have enough memory to run them on more than one GPU core at a time. So don't expect a GPU version to be much faster than a CPU version, if they ever decide to have both.

2. Their applications have gone over a year without any changes. Do you expect that to continue after a major update, such as adding GPU use or rewriting the program in a different computer language?

3. Their description of their algorithm sounds like enough things have to happen AFTER other things that trying to split the calculations onto different GPU cores running at the same time wouldn't speed up the workunit. It might, however, allow running more than one workunit at once on the same GPU, but only if there is enough graphics memory.

4. Many of their workunits take months to run. How long would you expect it to take to test a new version that won't run significantly faster?

However if you participate there, you might give them this link and ask them to check whether Nvidia's new FORTRAN compiler for their chips at least makes the conversion to GPU use easy enough that they're willing to try it anyway, although without trying to rewrite their applications in any other computer language:

http://www.gpugrid.net/forum_thread.php?id=1406

Note, however, that Nvidia does NOT make it clear whether their new compilers will work for any of their chips other than the GT300 that they haven't quite started selling yet.

When ATI starts offering a similar selection of compilers, and they work well enough, I'd expect to see more BOINC projects start offering both Nvidia and ATI versions of their applications, even those that now offer only CPU versions; but rather few before then.

lorenzo palloni
Send message
Joined: 17 Dec 08
Posts: 3
Credit: 333,482
RAC: 0
Level

Scientific publications
watwatwatwatwat
Message 13185 - Posted: 16 Oct 2009 | 10:27:04 UTC - in response to Message 12907.

http://developer.amd.com/GPU/ATISTREAMSDKBETAPROGRAM/Pages/default.aspx

is it the news we were waiting for?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 13190 - Posted: 16 Oct 2009 | 16:58:45 UTC - in response to Message 13185.

Almost.
It still lacks atomic operations which are required by ACEMD.
GDF

Tom Philippart
Send message
Joined: 12 Feb 09
Posts: 57
Credit: 23,376,686
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 13192 - Posted: 16 Oct 2009 | 19:10:00 UTC - in response to Message 13190.

maybe you could send in a request as a research organization. Maybe that'll speed up the process :)

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 13194 - Posted: 16 Oct 2009 | 22:46:27 UTC - in response to Message 13192.

We are talking with them already.

gdf

Harold R. Steighner
Send message
Joined: 8 Nov 09
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 13407 - Posted: 8 Nov 2009 | 13:17:38 UTC - in response to Message 1443.

Got a new ATI 4650, alas not supported...would like to get support

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 755,370,933
RAC: 212,472
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 13417 - Posted: 9 Nov 2009 | 4:06:53 UTC - in response to Message 13407.

From who? ATI isn't providing enough software support for preparing programs to run on their graphics cards yet, so expect very few BOINC projects to put in the extra effort required to get past their lack of support before this changes.

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 14176 - Posted: 11 Jan 2010 | 17:26:25 UTC - in response to Message 13417.

The new ATI 5xxx series are openCL capable and there is increasing number of BOINC projects with CUDA support which means more competition for the GPU cycles. Just look at the Collatz or MilkyWay pojects. They are able to produce an ATI app and You can see how much work they're doing right now. Still the Folding@Home is also using the ATI cards....

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 14177 - Posted: 11 Jan 2010 | 18:10:20 UTC

work is in progress ...

http://www.gpugrid.net/forum_thread.php?id=1458
____________
Thanks - Steve

Profile Aleksander Parkitny
Send message
Joined: 30 Sep 08
Posts: 12
Credit: 116,798
RAC: 0
Level

Scientific publications
watwat
Message 14824 - Posted: 30 Jan 2010 | 10:14:37 UTC - in response to Message 14177.

Maybe talking to the guys from F@H would help? They produced app for ATI cards. So maybe they can guide You?

Profile Dan T. Morris
Send message
Joined: 19 Jun 08
Posts: 13
Credit: 806,868
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 16247 - Posted: 12 Apr 2010 | 5:46:24 UTC - in response to Message 14824.

Yeah I can see that Nvidia will no longer let a person use the crunching power that you paid for with the 480 series card. Go figure.

Maybe its now time to get serious about ATI power.. :)


DD,

____________

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 16308 - Posted: 15 Apr 2010 | 12:44:00 UTC - in response to Message 16247.

Yeah I can see that Nvidia will no longer let a person use the crunching power that you paid for with the 480 series card. Go figure.

Maybe its now time to get serious about ATI power.. :)


DD,


That's not quite exact. For GPUGRID you will be able to take full advantages of the Fermi cards.

We are working with ATI to get a faster version on their cards as well.

GDF

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 755,370,933
RAC: 212,472
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 21180 - Posted: 12 May 2011 | 1:51:55 UTC

I believe you wrote that one of the problems in starting an AMD/ATI version is the lack of a certain FFT function.

An AMD/ATI OpenCL math library that MIGHT help:

http://developer.amd.com/gpu/appmathlibs/Pages/default.aspx

Contains some FFT functions, so could you look it over and let us know if it contains something suitable for the missing function?

Post to thread

Message boards : Wish list : ATI Cards?

//