Advanced search

Message boards : Graphics cards (GPUs) : GPUGRID and ATI

Author Message
cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 24435 - Posted: 14 Apr 2012 | 9:13:23 UTC - in response to Message 24409.
Last modified: 14 Apr 2012 | 9:14:04 UTC

How's the progress coming along?
I'm guessing the HD 7770 is being tested in-house.

Thanks.

http://www.gpugrid.net/forum_thread.php?id=2778

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 24409 - Posted: 12 Apr 2012 | 16:54:11 UTC

How's the progress coming along?
I'm guessing the HD 7770 is being tested in-house.

Thanks.

Spatzthecat
Send message
Joined: 26 Nov 09
Posts: 27
Credit: 150,056,844
RAC: 0
Level
Ile
Scientific publications
watwatwatwatwatwatwatwat
Message 23404 - Posted: 12 Feb 2012 | 0:05:52 UTC - in response to Message 23395.

Sounds fishy. Somehow feels like a scam. Didn't expect this from medical boinc project. I think i will quit gpugrid and move to other projects untill they too start to try make money out of crunching power.



Can you please clarify what you mean by "Sounds fishy" and "Somehow feels like a scam"

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 23399 - Posted: 11 Feb 2012 | 17:28:37 UTC

It's to sponsor their project, not personal gain.

LTDAkiles
Send message
Joined: 29 Sep 10
Posts: 4
Credit: 2,389,778
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwat
Message 23395 - Posted: 11 Feb 2012 | 9:24:04 UTC

Sounds fishy. Somehow feels like a scam. Didn't expect this from medical boinc project. I think i will quit gpugrid and move to other projects untill they too start to try make money out of crunching power.
____________

Profile rilian
Send message
Joined: 4 Feb 09
Posts: 6
Credit: 16,374
RAC: 0
Level

Scientific publications
watwat
Message 23383 - Posted: 10 Feb 2012 | 21:10:26 UTC

GDF, why don't you give away resources of people, who came here to crunch for science, to render commercial graphics? it will bring you even more money, and stable payouts
____________

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 23382 - Posted: 10 Feb 2012 | 20:56:33 UTC
Last modified: 10 Feb 2012 | 21:04:18 UTC

no comment

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 23337 - Posted: 8 Feb 2012 | 22:14:50 UTC - in response to Message 23336.

yes the right url is
http://donateathome.org

gdf

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 23336 - Posted: 8 Feb 2012 | 22:12:52 UTC
Last modified: 8 Feb 2012 | 22:13:31 UTC

Keeps telling me I'm using the wrong url even though I've tried the proper one 3 times ...

PBOYZTOY16 Donate@Home 2/8/2012 5:09:34 PM You used the wrong URL for this project. When convenient, remove this project, then add http://donateathome.org/
____________
STE\/E

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 23335 - Posted: 8 Feb 2012 | 21:59:33 UTC - in response to Message 23331.

HI Steve. We will add that this is using BOINC.
I did not understand your first point.
thanks


The Project says:

You don’t gain bitcoins in this project.


was my 1'st point. Maybe the GPUGrid Project gets the BitCoins ??? but from what I read the Participants don't. Not that that bothers me as I don't have anything to do with BitCoins anyway, at least not at this time anyway ...
____________
STE\/E

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 23333 - Posted: 8 Feb 2012 | 21:37:56 UTC
Last modified: 8 Feb 2012 | 21:39:20 UTC

If I understand correctly this is just a testing phase to get the project started.
I don't think they will actually get BitCoins until the bugs are ironed out (I could be wrong).

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 23331 - Posted: 8 Feb 2012 | 21:34:00 UTC - in response to Message 23328.

HI Steve. We will add that this is using BOINC.
I did not understand your first point.
thanks

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 23328 - Posted: 8 Feb 2012 | 21:12:45 UTC - in response to Message 23326.
Last modified: 8 Feb 2012 | 21:17:47 UTC

For me the bitcoin network is an impressive idea. Read about it:
http://bitcoin.org/bitcoin.pdf

I understand that somebody might want to do scientific calculations instead, that's why we kept it separate. So you can choose to participate. It's very different from molecular simulations of course.

gdf


But the Project it's self say's you won't be getting any Bitcoin's ... Also other than saying it gives BOINC Credits no where at or on the Site does anything indicate it's a BOINC Project, at least I couldn't find anything ...


What is donate@home?

Your compute time will support science and at the same time participate to the bitcoin experiment. You don’t gain bitcoins in this project. You do receive BOINC credits which have only a symbolic value of your participation. So, the only reason to participate is to make a difference, nothing else is gained. You might also want to volunteer your computing time to do scientific calculations, then give a look at gpugrid.net

____________
STE\/E

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 23327 - Posted: 8 Feb 2012 | 21:06:28 UTC

I'll give it a try.
There are still bugs in the other projects that refrain me from running them 24/7 so it can't hurt.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 23326 - Posted: 8 Feb 2012 | 21:02:40 UTC - in response to Message 23322.

For me the bitcoin network is an impressive idea. Read about it:
http://bitcoin.org/bitcoin.pdf

I understand that somebody might want to do scientific calculations instead, that's why we kept it separate. So you can choose to participate. It's very different from molecular simulations of course.

gdf

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 23322 - Posted: 8 Feb 2012 | 20:08:01 UTC

It's a nice idea, but I'd rather donate my GPU resources in a more "scientific" way through GPUGRID.
I'll run other projects (Milkyway, POEM) on my 7970 until GPUGRID is ready.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 23320 - Posted: 8 Feb 2012 | 19:08:36 UTC

You've been serious about this, haven't you? ;)

MrS
____________
Scanning for our furry friends since Jan 2002

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 23314 - Posted: 8 Feb 2012 | 18:46:26 UTC - in response to Message 23175.

If you have an ATI card please attach to this satellite project of gpugrid.
http://donateathome.org

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 23175 - Posted: 29 Jan 2012 | 11:27:42 UTC

Cite from another thread:

GDF wrote:
If all goes well, there are good chances that we will be able to provide a new AMD application very soon for all type of cards which are supporting OpenCL, not only GCN. We are already testing now.

I'll keep you posted by the end of the next week.

MrS
____________
Scanning for our furry friends since Jan 2002

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22945 - Posted: 11 Jan 2012 | 12:06:08 UTC - in response to Message 22942.

Or Timo from POEM since he got the application running with OpenCL.

[P3D] Crashtest
Send message
Joined: 9 Nov 09
Posts: 3
Credit: 11,700
RAC: 0
Level

Scientific publications
wat
Message 22942 - Posted: 11 Jan 2012 | 6:13:12 UTC

Big Question at the Projectteam here:

Did you test AMD APP SDK with FFToptimized API ( http://developer.amd.com/libraries/appmathlibs/Pages/default.aspx )

AMDs current GPUs should crunsh nVIDIA without any problems like they do on:
- Collatz
- Milkyway
- POEM
- Einstein@Home Beta called Albert@Home

By the way if OpenCL is the problem - use the working one : CAL

It would be easier if you would allow external developer so people like Gipsel could help ...

Profile SMTB1963
Avatar
Send message
Joined: 27 Jun 10
Posts: 35
Credit: 142,831,243
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 22825 - Posted: 25 Dec 2011 | 6:38:04 UTC - in response to Message 22818.

I do not intend to spend countless €s on new state-of-the-art GPU hardware every year just to be able to run your project!


Christopher, there are plenty of low cost options for running GPUGRID if you really want to. I recently bought a b-stock GTX275 SC from EVGA for < $150 shipped. Damn thing gets ~100K/day as long as my network's not futzing out. I've seen GTX260s on e-bay for as little as $40, and they're good for 45-50K/day (maybe more).

My point is that with a little looking, one doesn't have to spend much at all to volunteer here.

But if you're just dead set on running AMD and nothing else, I'm afraid you're going to have to be a little more patient.

Cheers!


Dagorath
Send message
Joined: 16 Mar 11
Posts: 509
Credit: 179,005,236
RAC: 2
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwat
Message 22823 - Posted: 25 Dec 2011 | 1:50:45 UTC - in response to Message 22818.

A project that thinks its volunteers can afford high-priced high-end cards every year, is severely mistaken, has lost it and obviously is living in a dream world!


You are the one living in a dream world. They don't think volunteers can afford expensive cards every year. The reason they will not support the current crop of AMD GPUs is because they cannot get their algorithm to work well on it. It does *not* matter how many other projects have their code running on AMD, the only thing that matters is the fact that the GPUgrid code does not run well on current AMDs. Maybe with more time and effort it could be done but the developers have made it clear they have invested all the time and effort they can. You have to get used to the idea that project admins are not born to serve us and cater to our demands to make their project run on our hardware and no amount of whining and bellyaching from you is going to change that. Accept it like an adult and move on.

I do not intend to spend countless €s on new state-of-the-art GPU hardware every year just to be able to run your project!


You don't have infinite amounts of money and they don't have infinite amounts of time. Yet you expect them to spend countless hours developing for your hardware! That isn't the way the world works. Successful people know when it's time to stop flogging a dead horse and move on. As far as GPUgrid computations are concerned, your AMD is a dead horse and it isn't going to run no matter how hard you/they flog it. It's time to bury the horse and move on to something productive.

And btw regarding that there has allegedly not been enough time to take care of this matter earlier: was the new website REALLY that important for the project or the science? The old one did it for me personally quite well IMHO.


The old website and server sucked. This one is much better and the time they invested in that was well spent.

Sorry, this becomes frustrating and unnerving! Every new post in here makes one think: "Oooh, now there is good news!" and the news just go from bad to worse!


Awwwwww! I'm so sad for you. Maybe in the future the developers should just say nothing about anything they try just so you won't be disappointed? Do you promise to then say nothing concerning the fact the developers say nothing?

Have a nice Christmas though and I hope Santa is putting a thousand brand new, top model AMD Radeon HD 7990 under your Christmas trees even before release date, or however the enthusiast card will be named then! ;)


Merry Christmas. I hope Santa brings you one too.

Ban me, if you must, for I am beginning to not care anymore!


Do you think we care that you don't care? The only one you seem to care about is you. You don't care about what the devs have tried to do and why it won't work. Why should anybody care about you not caring?

Profile Retvari Zoltan*
Avatar
Send message
Joined: 20 Jan 09
Posts: 925
Credit: 3,490,350,637
RAC: 4,440,969
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 22822 - Posted: 24 Dec 2011 | 21:05:47 UTC - in response to Message 22818.

I have two fairly newish AMD Radeon HD 6990/ATI Radeon HD 5970 cards here, that of course do well on other projects, but will supposedly not here...

If you want to contribute to this project so badly, should't you have bought a couple of not so high end nVidia cards (GTX 470, 570) instead?

A project that thinks its volunteers can afford high-priced high-end cards every year, is severely mistaken, has lost it and obviously is living in a dream world!

You are obviously mistaken. You have a card like that, but from the wrong vendor.

Why not simply making the prerequisite of a working quantum computer, if you want to sort out only wealthy gals and guys? I do not intend to spend countless €s on new state-of-the-art GPU hardware every year just to be able to run your project!

Your frustration speaking. I guess you would be equally frustrated from a ridiculously performing AMD client. You can use well your AMD cards elsewhere, so there is no loss from scientific point of view.

And btw regarding that there has allegedly not been enough time to take care of this matter earlier: was the new website REALLY that important for the project or the science? The old one did it for me personally quite well IMHO.

This is a selfish point of view. Maybe the old server software and website didn't do so well for the project.

Sorry, this becomes frustrating and unnerving! Every new post in here makes one think: "Oooh, now there is good news!" and the news just go from bad to worse!

See this post.

Ban me, if you must, for I am beginning to not care anymore!

AMD should be banned for making their customers believe that their GPUs can do everything that nVidia GPUs can. However they will catch up with their new GCN architecture.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22821 - Posted: 24 Dec 2011 | 20:29:49 UTC - in response to Message 22819.
Last modified: 25 Dec 2011 | 8:37:52 UTC

Hi,
let me clarify because I see that there are a lot of people that would like to use their cards and fear that they cannot. I am not saying anything new and we are not working against you, but with you.

1) Bad, well known news. The VLIW5/4 AMD cards are very powerful cards and usually cheaper than nvidia, also the top cards beat nvidia in terms of raw performance. However, only very numerical intense algorithms can benefit. For gpugrid, a top VLIW5 card consumes almost like a gtx580 but has a performance of a gt8800, a card of several years ago which is not recommended. It's not a decision, there is nothing we can do about it and we tried a lot. Other algorithms are better with this and electricity costs 3 times the card.

2) Good news, to be confirmed. In the future, new Amd cards based on GCN should level up with Nvidia fermi/kepler. We could not check it but all the indications are good. So take this with a level of uncertainty. Expect to have an equal representation of nvidia and amd at gpugrid in the future. This will be great because half (or more) of the world cannot crunch for us now. In the past few months we have spent plenty of hours to prepare for AMD.

Of course, don't go and buy a new card because of this! The HD7950 is expensive because now it's the fastest card worldwide. It will cost 100USD or less as soon as nvidia wakes up with kepler. There will be other reduced GCN cards also cheaper and less power hungry. According the news websites, AMD will put out all the new cards within 6 months, so there should be plenty.

We only have a single VLIW5 card now. That's all. Maybe soon a GCN card. We will test on those. If it works, meaning it is really stable, the application will go out. If you want to use it on VLIW5 do it, but your are not using it for its capabilities here and there is nothing, absolutely nothing we or amd can do about it.

Enjoy holidays and merry Christmas.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22819 - Posted: 24 Dec 2011 | 19:23:17 UTC - in response to Message 22818.
Last modified: 24 Dec 2011 | 19:34:31 UTC

I'm far from disappointed or frustrated; I was holding onto an HD5850 in the hope I could use it here, but now that POEM has launched an AMD GPU app, I am delighted I can use it there on a Bio-science project, somewhat similar to this project. Even if GPUGrid launched the AMD app, I don't think I would want to use it here - well not if it would only contribute 1/8th the performance of a comparable NVidia GPU, and we are talking about a 2088GFlops GPU performing like a GT240, and not if the performance at POEM is 4 or 5 times better. Performance is very much down to the research requirements, and GPUGrid does some of the most complex GPU research anywhere.

You have a GTX 480 running here, and doing good research. Be happy with that, and use your AMD GPU's at POEM.

GDF and the other researchers/developers have to make decisions like this. They have limited numbers and resources, so if they feel now is not a good time to launch an AMD app (due to poor performance, numerous bugs over several generations, and other projects) then I can live with that. The time and effort required to facilitate and support a poor performing app is probably not worth it. Kepler is also coming out soon, and the team has to prepare to facilitate those cards, and they will probably be twice as fast at crunching.

A lot of effort went into testing both a new server system and site. Other ongoing considerations include new apps, and better recognition systems. The site definitely needed an overhaul and in my opinion has resolved many issues and made many significant improvements. The server upgrade simply had to be done.

I will not be rushing out to buy a 7950, and I would not suggest anyone else does - they will be too expensive, for me, and who knows what the performance will be like. Six months or a year from now, when we know what the performances are and cost relative to comparable NVidia cards, then perhaps I will get one, or perhaps I will get one for another project, POEM, or possibly WCG.

PS. Nobody is getting banned for speaking their mind. If you say it like you see it, keep it clean, and don't attack anyone, I have no issues. Throw a wobbler and crazy posts just get deleted, not your account. Have a good Christmas.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Profile Christopher Herr
Send message
Joined: 1 Apr 10
Posts: 13
Credit: 106,905,353
RAC: 180
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 22818 - Posted: 24 Dec 2011 | 18:05:22 UTC

Well, if this should or will indeed be the final verdict here, then this will be it for me! If that is all, you were able to conclude in around two years, then this is disappointing and poor!
I have two fairly newish AMD Radeon HD 6990/ATI Radeon HD 5970 cards here, that of course do well on other projects, but will supposedly not here...

A project that thinks its volunteers can afford high-priced high-end cards every year, is severely mistaken, has lost it and obviously is living in a dream world! Why not simply making the prerequisite of a working quantum computer, if you want to sort out only wealthy gals and guys? I do not intend to spend countless €s on new state-of-the-art GPU hardware every year just to be able to run your project!

And btw regarding that there has allegedly not been enough time to take care of this matter earlier: was the new website REALLY that important for the project or the science? The old one did it for me personally quite well IMHO.

Sorry, this becomes frustrating and unnerving! Every new post in here makes one think: "Oooh, now there is good news!" and the news just go from bad to worse!

Have a nice Christmas though and I hope Santa is putting a thousand brand new, top model AMD Radeon HD 7990 under your Christmas trees even before release date, or however the enthusiast card will be named then! ;)

Ban me, if you must, for I am beginning to not care anymore!
____________

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22817 - Posted: 24 Dec 2011 | 17:21:22 UTC - in response to Message 22816.

We never tested on VLIW4. Yet, I would expect marginal improvements compared to VLIW5. It's not just a problem of VLIWX, there are many more problems in those cards for compute.

As I said as long that older cards are stable, we can support them. What we cannot support is too many generations because each one comes with their bugs. The more bugs there will be, the less we can support. A VLIW5 card would probably be 8 times slower than Kepler (consider the save in electricity).

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22816 - Posted: 24 Dec 2011 | 16:55:04 UTC - in response to Message 22815.
Last modified: 24 Dec 2011 | 17:11:33 UTC

Probably doesn't mean definitely. So no GCN for 7800 series, fair enough. That changes the picture, for me. In Jan there would only be two very expensive GCN cards. Don't know how worth while it would be supporting a project with say ten 7900 GPU's attached?

As for that wiki page, I have my usual doubts about it's accuracy.

My take on things is that the initial 7000 release will use GDDR5, and not XDR2?
If so, and if future cards will use XDR2 then is it also possible later editions of the 7800 will be released with GCN?
Will the 7800's use VLIW5, just like the present (40nm) 7600M cards, and will it be any better than VLIW4 and VLIW3?

Was the AMD app ever tested on VLIW4 and VLIW5 GPU's?
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22815 - Posted: 24 Dec 2011 | 15:45:30 UTC - in response to Message 22813.
Last modified: 24 Dec 2011 | 15:47:51 UTC

GDF made the decision that GPUGrid will only try to use GCN based AMD GPU's due to the relatively poor performance of existing AMD models (up to and including the 6000 series), and presumably logistical reasons. Basically, this will limit any GPUGrid app to the high and probably mid range 7000 series AMD GPU's; Radeon HD 7970, 7950, 7990 and probably 7870 and 7850. I doubt if the intermediate 7770 and 7750 GPU's would be usable (depends on performance/GCN). Obviously testing can't happen until the GPU's have been released and the researchers have a GPU to test. STEVE pointed out the high costs, 'over $500 for the 7970'. Good point, there might not be a massive demand for an AMD app. I suspect cards might trickle out in pairs rather than all get released at once and the 7990 is not due until March.

As pointed out you can now use your old and present AMD GPU's at POEM. 4000 through to 6000 series.
Requirements are Boinc Beta version 7.x and I think Catalyst 11.x drivers. Works for Linux and Windows and is OpenCL.


You're wrong.
http://en.wikipedia.org/wiki/Comparison_of_ATI_graphics_processing_units#Southern_Islands_.28HD_7xxx.29_series
The only cards that will support GCN will be 79x0. Just like 69x0 were the only ones from the 6xxx series that supported VLIW4.

From my point of view, if poem did it, if folding did it, if wcg (most probably) did it, why couldn't gpugrid not do it?
I would have preferred to have a low gpu utilization just like over at folding, instead of not heaving any app. The guys from SETI faced the same problem, but they bypassed by running two applications simultaneously.
I actually would have wanted a low utilization since I would have let the app running non-stop while also doing other stuff on the computer.

All I can say is ...¡Feliz Navidad! and see you in 1 year+, when the new fusion series is release which, hopefully, will have GCN on it.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22814 - Posted: 24 Dec 2011 | 15:42:20 UTC - in response to Message 22813.

HI,
just to specify we still have not tried any GCN based card. We expect to be good, but don't have any number, nor we know if it works at all. We have amd to send us one, let's see.
The code is ready, so as soon as we get one in, we should be able to send out the numbers.

Looking at the numbers on compute published by some sites, I still expect kepler to be faster, but probably quite close.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22813 - Posted: 24 Dec 2011 | 15:15:26 UTC - in response to Message 22808.
Last modified: 24 Dec 2011 | 15:23:50 UTC

GDF made the decision that GPUGrid will only try to use GCN based AMD GPU's due to the relatively poor performance of existing AMD models (up to and including the 6000 series), and presumably logistical reasons. Basically, this will limit any GPUGrid app to the high and probably mid range 7000 series AMD GPU's; Radeon HD 7970, 7950, 7990 and probably 7870 and 7850. I doubt if the intermediate 7770 and 7750 GPU's would be usable (depends on performance/GCN). Obviously testing can't happen until the GPU's have been released and the researchers have a GPU to test. STEVE pointed out the high costs, 'over $500 for the 7970'. Good point, there might not be a massive demand for an AMD app. I suspect cards might trickle out in pairs rather than all get released at once and the 7990 is not due until March.

As pointed out you can now use your old and present AMD GPU's at POEM. 4000 through to 6000 series.
Requirements are Boinc Beta version 7.x and I think Catalyst 11.x drivers. Works for Linux and Windows and is OpenCL.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 22811 - Posted: 24 Dec 2011 | 15:04:26 UTC - in response to Message 22796.

Expect to pay over $500 for the 7970, at least right at first anyway ...

Profile Szopler
Send message
Joined: 17 Oct 08
Posts: 4
Credit: 1,058,161
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 22808 - Posted: 24 Dec 2011 | 11:37:20 UTC

POEM did it, and you ?

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 303
Credit: 52,518,184
RAC: 21,933
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 22796 - Posted: 22 Dec 2011 | 14:58:15 UTC

I'm about to look up the HD 7970 and check if it fits my rather tight power limit requirements, or if I need to wait for a later board in that series that does.

Tapio
Send message
Joined: 24 Jan 09
Posts: 37
Credit: 14,726,012
RAC: 1,706
Level
Pro
Scientific publications
watwatwatwatwatwatwatwat
Message 22792 - Posted: 22 Dec 2011 | 8:51:56 UTC - in response to Message 22790.


I'm still waiting for the windows version and the gpu version of hcc over at wcg.
The system requirements on this project are too high.
I'm out.


I wait also windows app. :-)

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22790 - Posted: 21 Dec 2011 | 20:12:21 UTC - in response to Message 22786.

if you like to "fold" some aminoacids/proteins with your GPU, you now can try POEM@Home and their new open CL app.
A the moment onlx Linuy is supported, Windows will follow. All ati-cards from the 4000, 5000 and 6000er serie are supported.

I'm still waiting for the windows version and the gpu version of hcc over at wcg.
The system requirements on this project are too high.
I'm out.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 22789 - Posted: 21 Dec 2011 | 19:33:50 UTC

I'm still hoping to get some wu's for my AMD-Cards ...

Was ist neu im AMD APP SDK V2.6?

Key features supported in SDK 2.6 and the Catalyst 11.12 drivers include:

OpenCL runtime integration into Linux and Windows® Catalyst drivers.
Inclusion of the Khronos C++ wrapper API.
Multi-GPU support on Linux platforms.
PX5 support.
Preview: Support for AVX instructions on CPUs that support AVX.
Support for FMA4 instructions in OpenCL built-in function libraries on CPUs that support FMA4.
Kernel reflection, query kernel parameters, and enable use of OpenCL kernels in data-driven applications.
Support for atomic counters on APUs.
Redesign of OpenCL run-time on CPU, significantly improving performance.
Support for the cl_amd_media_ops2 extension, exposing hardware capabilities for accelerating image-related processing.
Async copies preview (set environment variable GPU_ASYNC_MEM_COPY=2 to enable).


The OpenCL 1.2 preview includes the following capabilities (requires 8.93.10 preview drivers):

Host access flags for memory objects enables more efficient buffer handling.
Pattern-based GPU buffer and image initialization eliminates need for certain buffer/image transfers.
Memory objects migration supports early transfer of buffers in preparation for when they are needed.
New generalized image creation API.
Enhanced image/buffer map operations.
OpenCL 1.2 CPU device partition, including partition of a CPU after addition to a context.
Generalized 1D and 2D images, image arrays, and image <--> buffer interop.
The 8.93.10 preview drivers also enable use of the static C++ kernel language.


gDEBugger version 6.1 is a major improvement in performance and robustness over version 6.0.
It can be downloaded for use with this SDK from http://developer.amd.com/gDEBugger.

Integrated with Microsoft® Visual Studio
Release Notes


APP KernelAnalyzer v 1.1:

Support for AMD Radeon™ HD7000 series GPUs (compilation only, no analysis).
Support for Catalyst revisions through 11.11.
Support for compiling kernels with the installed driver (select Installed Driver under the CAL version in the Options panel).
Format and Target Object Code are now separated.


APP Profiler v2.4 includes several key new features, including:

A kernel occupancy analyzer that estimates, for each kernel dispatch, the number of in-flight wavefronts on a compute unit as a percentage of the theoretical maximum number of wavefronts that the compute unit can support. In addition to reporting the occupancy percentage, the profiler can display a report that can help the developer achieve a higher occupancy percentage.
The ability to navigate from the API trace to the source code that called an OpenCL API.
Improved OpenCL API analysis that provides performance suggestions to the developer.
The ability to filter which OpenCL APIs are traced.
Several UI enhancements, including the ability to rename sessions from the Session Explorer Window, and the ability to automatically delete Profiler sessions when closing a Microsoft Visual Studio solution.
Preview: Support for profiling with AMD Radeon HD 7000 series GPUs (requires AMD APP SDK v2.6 and an AMD Catalyst version that supports this hardware).

Hope that helps!

Profile Sandro
Send message
Joined: 19 Aug 08
Posts: 22
Credit: 3,660,304
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 22786 - Posted: 21 Dec 2011 | 10:46:36 UTC - in response to Message 22766.

if you like to "fold" some aminoacids/proteins with your GPU, you now can try POEM@Home and their new open CL app.
A the moment onlx Linuy is supported, Windows will follow. All ati-cards from the 4000, 5000 and 6000er serie are supported.

Norman_RKN
Send message
Joined: 22 Dec 09
Posts: 16
Credit: 23,522,575
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwat
Message 22766 - Posted: 20 Dec 2011 | 17:48:48 UTC - in response to Message 22758.

maybe opencl can open a door and could be easier to handle within a timescale ??
____________
http://www.rechenkraft.net

Profile oldDirty
Avatar
Send message
Joined: 17 Jan 09
Posts: 22
Credit: 3,140,107
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwatwat
Message 22758 - Posted: 20 Dec 2011 | 16:38:21 UTC

GCN is now 1D Shader or like some say 16D? :confused
____________

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22742 - Posted: 19 Dec 2011 | 21:19:53 UTC - in response to Message 22740.

As well as VLIW5, I take it?

MrS
____________
Scanning for our furry friends since Jan 2002

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22740 - Posted: 19 Dec 2011 | 19:40:02 UTC - in response to Message 22737.

We simply have not resources to take on the debugging on all VLIW4 AMD cards. The present and the future is GCN. Then, if it happens a miracle and the stability of VLIW4 is good, then we can also release for that. But again, running gpugrid on VLIW4 cards is very inefficient.

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22737 - Posted: 18 Dec 2011 | 21:43:38 UTC - in response to Message 22736.

Performance of alpha versions has been pretty lacking compared to what's possible on nVidia hardware. Folding@Home is facing a similar problem. The current ATIs are hard to utilize effciently. GCN is changing this, it should be approximately equal to Fermi in terms of flexibility. If any ATI can be useful here it's GCN.

Rumored launch is in 4 days, or maybe the 9th january.

MrS
____________
Scanning for our furry friends since Jan 2002

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22736 - Posted: 18 Dec 2011 | 19:38:38 UTC - in response to Message 22735.
Last modified: 18 Dec 2011 | 19:40:00 UTC

We don't know yet, but we will start testing on GCN cards and initially support those. I am expecting enough problems already with a single compiler generation.

gdf

If this is the best answer after 2 years of waiting, to support an architecture that it's not even out yet, then ... good luck with all that

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22735 - Posted: 18 Dec 2011 | 18:31:33 UTC - in response to Message 22734.

We don't know yet, but we will start testing on GCN cards and initially support those. I am expecting enough problems already with a single compiler generation.

gdf

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22734 - Posted: 18 Dec 2011 | 17:33:05 UTC - in response to Message 22733.

HI,
GCN is the suggested core for GPUGRID, its performance should rival Nvidia cores.

http://www.tomshardware.com/news/amd-HD-7900-XDR2-Rambus-Memory,13408.html

gdf

In an older post you said that the amd app should support OpenCL 1.0, meaning 4xxx series and up.
Are you trying to say that the 2 previous architectures will be abandoned?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22733 - Posted: 18 Dec 2011 | 15:28:00 UTC - in response to Message 22612.

HI,
GCN is the suggested core for GPUGRID, its performance should rival Nvidia cores.

http://www.tomshardware.com/news/amd-HD-7900-XDR2-Rambus-Memory,13408.html

gdf

Skip Da Shu
Send message
Joined: 13 Jul 09
Posts: 38
Credit: 41,955,165
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 22612 - Posted: 28 Nov 2011 | 22:39:25 UTC - in response to Message 22610.

Oh, wait, this IS software development! Just not at work. ;-) LOL

Not your work :D

MrS
ROFLMAO... Excellent point!

____________
- da shu @ HeliOS,
"A child's exposure to technology should never be predicated on an ability to afford it."

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22610 - Posted: 28 Nov 2011 | 19:56:30 UTC - in response to Message 22601.

Oh, wait, this IS software development! Just not at work. ;-) LOL

Not your work :D

MrS
____________
Scanning for our furry friends since Jan 2002

Skip Da Shu
Send message
Joined: 13 Jul 09
Posts: 38
Credit: 41,955,165
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 22601 - Posted: 27 Nov 2011 | 20:21:52 UTC - in response to Message 22591.

For people outside research it's sometimes difficult to imagine how it works... [beamed Kirk & Scotty out of quote] ...there's always something unexpected crossing your way. The question is just how severe that is and how much effort it takes to get around it. Personally I wouldn't dare committing to a release date for a highly experimental piece of software. Well, one could release a half baked alpha in time.. but that wouldn't please people either.

And people tend to be impatient and tend to hear what they want to hear. So "we hope to get an alpha out by x" quickly turns into "Woow, there'll be an app by x!". Which is quite different..

MrS


Hmmm, This sounds a lot like software development at work. Oh, wait, this IS software development! Just not at work. ;-) LOL
____________
- da shu @ HeliOS,
"A child's exposure to technology should never be predicated on an ability to afford it."

mikey
Send message
Joined: 2 Jan 09
Posts: 135
Credit: 49,647,590
RAC: 608,635
Level
Val
Scientific publications
watwatwatwatwatwat
Message 22599 - Posted: 27 Nov 2011 | 14:43:30 UTC - in response to Message 22591.
Last modified: 27 Nov 2011 | 14:44:09 UTC

Seems like things have calmed down again, but I'll add some more cents anyway:

For people outside research it's sometimes difficult to imagine how it works. In reality there's no

Kirk: Scotty, how much time do you need?
Scotty: 2 Days!
Kirk: I'll give you 1!
Scotty: I'll do it in a half!

and there's always something unexpected crossing your way. The question is just how severe that is and how much effort it takes to get around it. Personally I wouldn't dare committing to a release date for a highly experimental piece of software. Well, one could release a half baked alpha in time.. but that wouldn't please people either.

And people tend to be impatient and tend to hear what they want to hear. So "we hope to get an alpha out by x" quickly turns into "Woow, there'll be an app by x!". Which is quite different..

MrS


My Genealogy software did exactly that, they gave a hoped for release date and then caught holy he!! for missing it. They came on and gave the reasons for the delay and it did not help!! They now say it will be relased when it is ready and not before, they still catch some flak but not nearly as much!!

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22591 - Posted: 26 Nov 2011 | 19:22:08 UTC - in response to Message 22588.

Seems like things have calmed down again, but I'll add some more cents anyway:

For people outside research it's sometimes difficult to imagine how it works. In reality there's no

Kirk: Scotty, how much time do you need?
Scotty: 2 Days!
Kirk: I'll give you 1!
Scotty: I'll do it in a half!

and there's always something unexpected crossing your way. The question is just how severe that is and how much effort it takes to get around it. Personally I wouldn't dare committing to a release date for a highly experimental piece of software. Well, one could release a half baked alpha in time.. but that wouldn't please people either.

And people tend to be impatient and tend to hear what they want to hear. So "we hope to get an alpha out by x" quickly turns into "Woow, there'll be an app by x!". Which is quite different..

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Christopher Herr
Send message
Joined: 1 Apr 10
Posts: 13
Credit: 106,905,353
RAC: 180
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 22588 - Posted: 26 Nov 2011 | 15:19:24 UTC - in response to Message 22583.

Either give your volunteers a definitive date when there will be an app, and I mean definitive, or a definitive statement 'There will never be an ATI/AMD OpenCL app for the following reasons'! But please not this "sitting between two chairs" any longer!


Please tell me you're joking.

I'm not trying to start an argument here, but I really hope that you - a VOLUNTEER - are not making demands of the GPUGRID scientists & staff. They have limited resources, and we as volunteers are here to support them, not the other way around.

Look, if you've got some AMD cards lying around, I can appreciate you wanting to put them to work on this project. But IMO we should all remember our role as volunteers. The AMD app will get here when it gets here. Or not. Getting in a twist and griping because your expectations aren't being met is counter-productive.

Maybe I've misinterpreted your post...if so, I apologize in advance.

Cheers!


No, SMTB1963, I am not joking, but I have not 'demanded' anything as well. I said 'please' as you see in the second phrase of your quoted text of mine. :)

Yes, you are right, we are the volunteers, but we are also the volunteers! :)
We support science and the scientists more or less personally with volunteering a lot of money for hardware and power. In some cases this munificence makes the science possible in the first place, and I think this is one of these cases.

I do not see myself as a 'Megacruncher', if that is a part of my statement that should have been misunderstood. But I know a lot of guys in my team and elsewhere, that could be interested in unleashing a lot of power AMD-wise (and CUDA-wise) on this project.
And it is my point of view that all volunteers deserve to be kept up to date on the progress of such undertakings on a regular basis by an official. With this kind of answer of GDF and others:
We would like to get some alpha app for ati/amd out by the end of the year. However, we don't know if it will good enough to go in production. The update of the server has required a very large amount of work so far.

gdf

I am satisfied, because it contains all the official information I waited for. First, there is an app in alpha development status, second, it could (and perhaps will) be buggy, third, the reasons that hindered the progress...

Thank you to all officials and I look forward to help (alpha- or beta-)testing the app and will continue waiting with fingers crossed.
Please continue keeping us up to date whenever you can find the time!

Cheers and sorry again, if my chose of words in English was misunderstanding, as I am German and not a native speaker!
____________

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22584 - Posted: 25 Nov 2011 | 20:27:31 UTC - in response to Message 22583.

We would like to get some alpha app for ati/amd out by the end of the year. However, we don't know if it will good enough to go in production. The update of the server has required a very large amount of work so far.

gdf

Profile SMTB1963
Avatar
Send message
Joined: 27 Jun 10
Posts: 35
Credit: 142,831,243
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 22583 - Posted: 25 Nov 2011 | 16:29:43 UTC - in response to Message 22579.

Either give your volunteers a definitive date when there will be an app, and I mean definitive, or a definitive statement 'There will never be an ATI/AMD OpenCL app for the following reasons'! But please not this "sitting between two chairs" any longer!


Please tell me you're joking.

I'm not trying to start an argument here, but I really hope that you - a VOLUNTEER - are not making demands of the GPUGRID scientists & staff. They have limited resources, and we as volunteers are here to support them, not the other way around.

Look, if you've got some AMD cards lying around, I can appreciate you wanting to put them to work on this project. But IMO we should all remember our role as volunteers. The AMD app will get here when it gets here. Or not. Getting in a twist and griping because your expectations aren't being met is counter-productive.

Maybe I've misinterpreted your post...if so, I apologize in advance.

Cheers!

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22581 - Posted: 25 Nov 2011 | 15:57:48 UTC - in response to Message 22579.

There is an app, and it will be beta tested when the scientists are in the position to do so, and have the time. The scientists built, tested (as well as previous unusable apps) and reported what they observed in this thread. Server upgrades and site improvements are presently being tested/developed. These are required for the project as a whole, and to facilitate the introduction and maintenance of an AMD app. When these are finished and implemented the scientists can then begin work on preparing to release an AMD Beta app.
I'm not in a position to give a definitive release date, and I cannot see how the researchers would be, until just before being ready. As for commitment to an AMD app, it's my understanding that the researchers have put off the release of a new NVidia app at least in part to help forward the release of an AMD app.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Profile Christopher Herr
Send message
Joined: 1 Apr 10
Posts: 13
Credit: 106,905,353
RAC: 180
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 22579 - Posted: 25 Nov 2011 | 10:11:19 UTC

Right, TomaszPawel, 'this calls for immediate discussion', this seems to be all that is happening on the ATI/AMD front...
'Nothing new in the east, west, north and south'?!
I am getting frustrated with waiting; right now I have an AMD Radeon HD 6990 and an ATI Radeon HD 5970, the former quite new, the latter sort of newish, but both ready to jump train, as soon as there might even be a beta app and work for it! So you so to say waste the power of four high-end GPUs...
Either give your volunteers a definitive date when there will be an app, and I mean definitive, or a definitive statement 'There will never be an ATI/AMD OpenCL app for the following reasons'! But please not this "sitting between two chairs" any longer!
____________

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 119
Credit: 28,568,536
RAC: 25,059
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 22578 - Posted: 25 Nov 2011 | 7:30:13 UTC

What can I say after two years of waiting to be able to count on ATI ...

This is my comment:

http://www.youtube.com/watch?v=5P2jJdrz9bY

And on HD79xx in 48 seconds it is on target ...

:D
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22539 - Posted: 17 Nov 2011 | 20:32:36 UTC - in response to Message 22536.

me wrote:
20% at best


.. that's for a full Cypress (24/20), which the HD5850 is not ;)

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22536 - Posted: 16 Nov 2011 | 22:04:34 UTC - in response to Message 22535.

You lost me at 20% faster per clock and 24/18 is 33%...
Anyway, I understand now :)


____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22535 - Posted: 16 Nov 2011 | 21:21:06 UTC - in response to Message 22531.

No, that 20% comes from the fact that Cayman has 24 shader clusters compared 20 in a full Cypress. On the HD5850 2 of them are deactivated. That's where the first factor 24/18 comes from ;)

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22531 - Posted: 16 Nov 2011 | 18:50:47 UTC - in response to Message 22525.
Last modified: 16 Nov 2011 | 19:44:12 UTC

We agree that it's worth testing, and would be a worthwhile project.
As for performances, the speculation can end when we have Beta tasks :)

If you meant 'Cayman was 20% faster clock-for-clock' then you could add that factor into the equation, for a more agreeable middle-ground ;)

    24/18 * 900/725 * 1.2 = 1.98.


- Almost forgot about Antilles, and all those mid range AMD cards with only single precision.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22525 - Posted: 15 Nov 2011 | 20:53:41 UTC - in response to Message 22521.

Definitely worth it. For many GPU-Grid would be the first worthwhile project to run on AMDs :)

The HD5850 has 18 shader clusters at 725 MHz. Cayman has at best 24 and OC'ed about 900 MHz. You could go higher with the clock (given enough voltage), but so could the HD5850.
That works out to be a factor of 24/18 * 900/725 = 1.65. So.. I must confess I underestimated the difference a bit. Let's agree to meet in the middle, shall we? ;)

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22521 - Posted: 15 Nov 2011 | 0:59:28 UTC - in response to Message 22519.

Cayman may only be 20% faster than a 5850 per clock, but there are some monsters compared to a humble 5850. For example, you beast cuts through MW tasks in around a minute, while an HD5850 takes about three times as long. I'm not saying AMD's will be great, but some people might want to use their AMD GPU here.

If an HD 5850 really is about 1/4 the speed of a GTX580, then your Cayman might be somewhere around 2/3rds the speed of a GTX580.

Even if performance is poor compared to a GTX580, there are other considerations:
People with existing AMD GPU's have expressed an interest in using them here.
Their performance per Watt or per Euro/$/£ is not known for the app.
Some ranges of cards might be more competitive than others; mid-range AMD GPU's would be competing against many CC2.1 cards, rather than CC2.0 GPU's.

So, worth testing sometime I think.

____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Profile verlyol_[belgium]
Send message
Joined: 9 Sep 08
Posts: 31
Credit: 6,491,359
RAC: 17
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 22520 - Posted: 14 Nov 2011 | 20:46:48 UTC

Yes I think that the performance gap between AMD and Nvidia should be large enough.

when is that AMD GPUs will be supported ?

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22519 - Posted: 14 Nov 2011 | 20:46:09 UTC

AMDs current VLIW architecture is much harder to utilize efficiently than nVidias scalar shaders. That's likely why they've got these problems at F@H and why they're not yet rocking GPU-Grid. Sure, software is also a very important piece of the puzzle.. and AMDs compiler has a harder job keeping these execution units fed. But it's not always possible to work around fundamently architecture limitation by software - so it's not only about software maturity and "trying hard enough".

AMDs upcoming "Graphics Core Next" (GCN) will be the first departure from the VLIW since many years. It should be a lot easier to utilize. However, I doubt the same code path will run good on both, GCN and VLIW. That will lead to a very interesting, if not at times frustrating situation.

BTW: actually there's not that much of a difference between Cypress (HD58x0) and Cayman (HD69x0) apart from switching from VLIW5 to VLIW4. At best Cayman is 20% faster per clock. That matters a lot if you're competitive, but is nothing if you've got a catch up an entire order of magnitude :p

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22518 - Posted: 14 Nov 2011 | 15:06:20 UTC - in response to Message 22517.

It's hard to compare performances between different projects, especially Boinc to non-Boinc. However I would still expect a high end AMD GPU (HD6900) to significantly outperform a high end CPU (i7-2600), which is not the case at folding.

We know the performance is not great compared to high end GTX500 series GPU's by NVidia, but again we are comparing mature CUDA apps to a pre-Beta OpenCL/GL app.

Basically a GPUGrid entry level NVidia card (say GT240 to GT430) would be on par with a mid-range AMD GPU (say HD5850 to HD5870). Obviously the high end 6000series should fair much better; perhaps approaching say a GTX470. I would expect the high end Southern Islands GPU's (7000 series), supposedly due out before the end of the year to at least match a GTX470. That said, Kepler would stretch the gap again, when released.

Would be beta testing will come down to time/timing; some work presently under way has to come first. How long that takes depends on the problems that are encountered. Not sure if a new NVidia app will happen first (presently on hold until next 2012), perhaps just for Linux, what the situation is WRT publications, seminars..., and the researchers might want to finish off some work using older code before anything happens.

Should a beta test run, be successful, and a live app be successfully used for a while, development might mean the gap here between AMD and NVidia apps would fall for GPUGrid, perhaps even with a mature OpenGL 4.2. However, such positive speculation should be made cautiously; a Beta could demonstrate unreliability, and require too much work from the researchers to get an AMD project off the ground.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Profile verlyol_[belgium]
Send message
Joined: 9 Sep 08
Posts: 31
Credit: 6,491,359
RAC: 17
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 22517 - Posted: 13 Nov 2011 | 21:22:32 UTC

I am also interested by the future support for ATI / AMD, but I hope it will be stable and the performance will be a good level ?

I am part of the beta-team on folding @ home and for this project the performance of ATI / AMD is not very good even if there are improvements.

Digital Savior
Send message
Joined: 17 Sep 09
Posts: 1
Credit: 10,327,677
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwat
Message 22479 - Posted: 7 Nov 2011 | 3:03:58 UTC

I have two AMD GPUs I would love to use on this project! I hope you can get it out soon!

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22474 - Posted: 4 Nov 2011 | 20:43:18 UTC - in response to Message 22204.

Any news on the AMD application?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22204 - Posted: 2 Oct 2011 | 20:56:56 UTC - in response to Message 22202.

We are waiting to have time... We are submitting 2 new papers and two new applications for grants.

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22202 - Posted: 2 Oct 2011 | 20:16:13 UTC - in response to Message 22197.

What you waiting for?


Maybe for the app being ready for public alpha testing?

MrS
____________
Scanning for our furry friends since Jan 2002

Tapio
Send message
Joined: 24 Jan 09
Posts: 37
Credit: 14,726,012
RAC: 1,706
Level
Pro
Scientific publications
watwatwatwatwatwatwatwat
Message 22197 - Posted: 1 Oct 2011 | 16:01:05 UTC - in response to Message 22177.

gpu load is around 85%

g


any rough estimate for alpha testing?...like next week? :)


More like next month.

gdf


Ok. Now is 1.10.2011. What you waiting for?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22177 - Posted: 27 Sep 2011 | 6:47:38 UTC - in response to Message 22170.

gpu load is around 85%

g


any rough estimate for alpha testing?...like next week? :)


More like next month.

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22171 - Posted: 25 Sep 2011 | 22:20:54 UTC - in response to Message 22167.

...it's time to burn some fusion chips


The current Fursions are HD6000 series, so they should work :)
(don't expect performance miracles, though)

MrS
____________
Scanning for our furry friends since Jan 2002

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22170 - Posted: 25 Sep 2011 | 22:00:36 UTC - in response to Message 22169.

gpu load is around 85%

g


any rough estimate for alpha testing?...like next week? :)

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22169 - Posted: 25 Sep 2011 | 17:23:00 UTC - in response to Message 22167.

gpu load is around 85%

g

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22167 - Posted: 25 Sep 2011 | 13:11:07 UTC - in response to Message 22166.
Last modified: 25 Sep 2011 | 13:12:28 UTC

opencl 1.1 and jut single precision required.
H5000, H6000 and the new H7000 cards should work.

gdf


...it's time to burn some fusion chips
From my point of view, it doesn't matter if the gpu load is going to be 20-25%, I think my laptops are going to handle it

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22166 - Posted: 25 Sep 2011 | 11:49:45 UTC - in response to Message 22165.

opencl 1.1 and jut single precision required.
H5000, H6000 and the new H7000 cards should work.

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22165 - Posted: 25 Sep 2011 | 11:13:10 UTC - in response to Message 22162.

The new amd app will require OpenCL 1.1 and double precision cards?


The nVidia code doesn't require dp, so there's no reason why the ATI code would. Can't tell you the OpenCL version, though.

MrS
____________
Scanning for our furry friends since Jan 2002

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 431
Credit: 395,421,274
RAC: 167,058
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 22163 - Posted: 25 Sep 2011 | 10:44:58 UTC
Last modified: 25 Sep 2011 | 10:45:58 UTC

Is it possible to update the server for OpenCL without the *new credit* code?
If not, I would like to suggest that you run some NVidia WUs against the new server and let us know what we might see for credit per WU.

Not that I am participating in this excellent project for the points but we all know how fussy crunchers get when it comes to credits :-)
____________
Thanks - Steve

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22162 - Posted: 25 Sep 2011 | 9:40:36 UTC - in response to Message 22161.

The new amd app will require OpenCL 1.1 and double precision cards?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22161 - Posted: 25 Sep 2011 | 7:33:21 UTC - in response to Message 22159.
Last modified: 25 Sep 2011 | 11:43:42 UTC

We are practically ready with the app alone. We have to test it on the context of boinc now.

First thing to do is to update the server to a more recent version, then wait for problems. Our server software is quite old and there could be problems which would only appear in a couple of weeks. I don't want to test AMD with the old server as no opencl was there then.

In the meanwhile we will do our tests locally on a private new server.

An other thing to say that it will be windows only for the first several months.

We have tested on one OS (XP32) and one card (H5850). So expect few issues. Hopefully, it all works and we can start submitting real workunits for amd. In such case they will be on a different app, not to affect the work on nvidia cards, but the porting of the code is complete, so they will be of the same usefulness of nvidia wus.

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22159 - Posted: 24 Sep 2011 | 17:20:43 UTC - in response to Message 22153.

It is also four times cheaper.


And *only* consumes half the power..

MrS
____________
Scanning for our furry friends since Jan 2002

Skip Da Shu
Send message
Joined: 13 Jul 09
Posts: 38
Credit: 41,955,165
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 22156 - Posted: 23 Sep 2011 | 15:01:43 UTC - in response to Message 22153.

It is also four times cheaper.


LOL! Excellent point!

____________
- da shu @ HeliOS,
"A child's exposure to technology should never be predicated on an ability to afford it."

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 119
Credit: 28,568,536
RAC: 25,059
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 22153 - Posted: 23 Sep 2011 | 11:13:20 UTC - in response to Message 22152.

It is also four times cheaper.
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22152 - Posted: 23 Sep 2011 | 7:00:37 UTC - in response to Message 22150.

H5850 is about 4 times slower than the top card a gtx580.

gdf

Profile Retvari Zoltan*
Avatar
Send message
Joined: 20 Jan 09
Posts: 925
Credit: 3,490,350,637
RAC: 4,440,969
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 22150 - Posted: 22 Sep 2011 | 22:02:48 UTC - in response to Message 22144.

... Performance could be better...

Could you please specify this?

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 22149 - Posted: 22 Sep 2011 | 21:09:52 UTC - in response to Message 22144.

People have been waiting for over 2 years for an AMD application. Some of us are willing to test even nightly builds asap.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22144 - Posted: 22 Sep 2011 | 7:49:09 UTC - in response to Message 21965.

We have a more or less advanced stage application for AMD. Performance could be better, but at least it seems to run stably in a local machine. We might have a release of an alpha-app in the next couple of months.

gdf

cenit
Send message
Joined: 11 Nov 09
Posts: 23
Credit: 668,841
RAC: 0
Level
Gly
Scientific publications
wat
Message 21965 - Posted: 5 Sep 2011 | 16:27:34 UTC - in response to Message 21881.

there's a bug in the latest sdk that makes a full use of a cpu-core whenever an opencl app is running.
They promised a fix, but still not here in 11.8
maybe in 11.9??

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 21881 - Posted: 23 Aug 2011 | 18:12:25 UTC - in response to Message 21869.

11.7 has amd app sdk 2.5 Is it any good or not?

In the meantime 11.8 is released. Makes heavy use of the CPU as well.

What makes heavy use of the CPU?

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 21869 - Posted: 21 Aug 2011 | 19:08:30 UTC - in response to Message 21791.

11.7 has amd app sdk 2.5 Is it any good or not?

In the meantime 11.8 is released. Makes heavy use of the CPU as well.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 21816 - Posted: 7 Aug 2011 | 21:43:38 UTC - in response to Message 21791.

we will be looking again at AMD only after the summer and will post comments here.

gdf

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 21791 - Posted: 2 Aug 2011 | 20:18:53 UTC

11.7 has amd app sdk 2.5 Is it any good or not?

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20995 - Posted: 18 Apr 2011 | 9:22:28 UTC - in response to Message 20961.
Last modified: 18 Apr 2011 | 9:25:37 UTC

On ATI development, I don't hold up much hope for the immediate future,

The quality of any OpenCL implementation is simply poor so far. At every new version one bug is fixed and two new appear. For us, it was just a big loss of time, but we will be here for AMD when they are ready.
There is no plan of release for now.

gdf

On language,
My advice is to post in English if you can. If your English is not great, post in your preferred language as well; a bad English translation is unlikely to get a response, but if you also use your preferred language someone will probably be able to speak it or make a better translation. You can use Google Translate, to get a reasonable translation. Using simplified language helps.
I have no objections to a few friendly words in any language, but if anyone is going to ramble on use the Personal Messaging facility.

If you see a post in a language you don't understand, it's probably small talk, otherwise they would probably have posted in English.

Profile Christopher Herr
Send message
Joined: 1 Apr 10
Posts: 13
Credit: 106,905,353
RAC: 180
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 20961 - Posted: 14 Apr 2011 | 17:35:50 UTC

Hello everyone,
I would like to ask everyone to keep their posts in English so everyone can read them, despite being not a native English speaker, thank you! I am no Englishman neither, but I give my best to do postings in English only...

Greetings from Germany and I am waiting for a working ATI OpenCL app with fingers crossed, whenever I can spare them as well!
Christopher
____________

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 119
Credit: 28,568,536
RAC: 25,059
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 20941 - Posted: 13 Apr 2011 | 20:39:20 UTC - in response to Message 20910.

Nie wartościuje punktów ani projektów. Każdy projekt jest wartościowy, każde punkty cieszą. Liczę sobie tutaj, liczę sobie tam - dobrze się bawię i pozdrawiam wszystkich :D
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20910 - Posted: 12 Apr 2011 | 22:15:31 UTC - in response to Message 20906.
Last modified: 18 Apr 2011 | 8:55:14 UTC

Yeeee, sure :P

I sold HD5970 and bought... GTX480 - more ceredits then HD5970 but in.... PrimeGrid LOL

Still have HD5870... and waiting since 2009 for gpugrid app...

Ale dlaczego?
Czy uważasz, że punkty PrimeGrid są ważniejsze niż punkty GPUGRID?
Chodź tutaj.

en.
But why?
Do you think that the PrimeGrid points are more important than GPUGRID points?
Come here.

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 119
Credit: 28,568,536
RAC: 25,059
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 20906 - Posted: 12 Apr 2011 | 17:34:05 UTC - in response to Message 20898.

Yeeee, sure :P

I sold HD5970 and bought... GTX480 - more ceredits then HD5970 but in.... PrimeGrid LOL

Still have HD5870... and waiting since 2009 for gpugrid app...
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 20898 - Posted: 11 Apr 2011 | 5:46:20 UTC - in response to Message 20851.

SDK 2.4 was released a few days ago.
I hope this changes things for the better.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 20851 - Posted: 2 Apr 2011 | 21:48:54 UTC - in response to Message 20835.

The quality of any OpenCL implementation is simply poor so far. At every new version one bug is fixed and two new appear. For us, it was just a big loss of time, but we will be here for AMD when they are ready.
There is no plan of release for now.

gdf

Wolfram1
Send message
Joined: 24 Aug 08
Posts: 45
Credit: 3,431,862
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 20845 - Posted: 2 Apr 2011 | 10:37:16 UTC - in response to Message 20844.

I have running DNETC on 5870 and ist runninmg fine. I chaned yesterday tu 11.3 and it is a little quicker.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 20844 - Posted: 2 Apr 2011 | 9:10:05 UTC - in response to Message 20835.
Last modified: 2 Apr 2011 | 9:12:14 UTC

GDF said before that it's crashing a lot. I don't know anything else, though.
BTW: I've got Milkyway@Home running on my HD6950. It was fine with the release data beta Cat 10.12, but with Cat 11.2 it keeps crashing in conjunction with Win7 and Firefox 4 or a modern Winamp skin. Going to install 11.3 now and hope things improve.

MrS
____________
Scanning for our furry friends since Jan 2002

cenit
Send message
Joined: 11 Nov 09
Posts: 23
Credit: 668,841
RAC: 0
Level
Gly
Scientific publications
wat
Message 20835 - Posted: 1 Apr 2011 | 9:24:54 UTC - in response to Message 20450.

Stanford just released their OpenCL client for ATi (OpenCL 1.1, so only Radeon HD5xxx and 6xxx cards, older cards are being abandoned by that project), based on their opensource toolkit OpenMM. They seems to be happy with SDK 2.3, and a lot more for the speed improvements in 2.4 (in beta/RC now, with the latest 11.3/11.4 drivers).

What about you? Can I ask what are your real biggest problems deploying ATi OpenCL client?
You keep saying that it's buggy and I could be fine with that, but could you just elaborate a little bit on that? It's my curiosity asking!

MAD!
Send message
Joined: 9 May 10
Posts: 6
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 20450 - Posted: 15 Feb 2011 | 2:52:43 UTC

Oh, will check back here in 6 months and see then...

Dragokatzov
Send message
Joined: 31 Oct 08
Posts: 2
Credit: 3,771,990
RAC: 145
Level
Ala
Scientific publications
watwatwat
Message 20449 - Posted: 14 Feb 2011 | 23:26:44 UTC - in response to Message 20428.

It's now the middle of February and nothing?
I really don't mean to be pushy, but there are several ATi users just waiting to contribute to this project.

Even if it's buggy, there are several of us here that would help you guys out with testing and hopefully provide useful feedback.
Like I've said before, something's better than nothing.


Myself included, I am going to upgrade back to an nvidia GPU soon, but still would like to beta test an ATI app

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20448 - Posted: 14 Feb 2011 | 22:05:57 UTC - in response to Message 20440.

That is good news. I hope it goes well.

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 20440 - Posted: 14 Feb 2011 | 18:40:53 UTC

That's sad to hear. :(
On the bright side Folding@home for ATi GPU's should have a new core and client soon (March). It's OpenCL/OpenMM based. :)

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20432 - Posted: 14 Feb 2011 | 7:35:31 UTC - in response to Message 20428.

Given the expected ETA's of the next two APP SDK's and some knowledge of the researchers timetable, I don't expect another sustained attempt until around Aug.

If the drivers are not stable for a range of cards then by the time you debug/find work arounds, the next APP SDK will be out, making your efforts worthless. ATI even changed the name from SDK to APP SDK. It's down to ATI to have a useable APP SDK from which to work from.

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 20428 - Posted: 13 Feb 2011 | 20:37:14 UTC

It's now the middle of February and nothing?
I really don't mean to be pushy, but there are several ATi users just waiting to contribute to this project.

Even if it's buggy, there are several of us here that would help you guys out with testing and hopefully provide useful feedback.
Like I've said before, something's better than nothing.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20294 - Posted: 31 Jan 2011 | 13:11:27 UTC - in response to Message 20288.
Last modified: 31 Jan 2011 | 13:54:59 UTC

I cant find a future AMD Accelerated Parallel Processing SDK (APP SDK) , but my guess is SDK2.4 will be released in mid of April:
The last release dates have been in Dec 2009, Feb 2010, May 2010, Aug 2010, and Dec 2010. It looks like the release dates are spreading out; the gap has been 2months, 3months, 3months and 4months. So being optimistic my guess is about 4months from the 13th Dec 2010 release date – mid April 2011.

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 20288 - Posted: 30 Jan 2011 | 16:42:47 UTC - in response to Message 20286.

If not then we will have to wait for SDK 2.4, and testing will basically start over again.

When should the new SDK be available?

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20286 - Posted: 30 Jan 2011 | 0:07:11 UTC - in response to Message 20285.

GDF said that the problem was numerous bugs in the present SDK package. We know speed is not great but getting a working app is the main problem. Perhaps they will do better using Windows with the present SDK, but I am not holding up much hope. If not then we will have to wait for SDK 2.4, and testing will basically start over again.

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 20285 - Posted: 29 Jan 2011 | 23:27:16 UTC - in response to Message 20283.


If speed is the issue, then I'm sure most of us won't mind (it's better than nothing).


+1

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 20284 - Posted: 29 Jan 2011 | 20:43:53 UTC

Yesterday Lunatics released their first ATI-OpenCL App for SETI (Multibeam).
:-(
~25min/task @ HD6950 6.10 (ati 13ati)
~21min/task @ GTX460 6.08 (cuda)

looks like ATI and OpenCL are not really best friends, but at least they run stable (I updated the ati and nVidia-drivers today).

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 20283 - Posted: 29 Jan 2011 | 20:21:21 UTC

So... Delayed until after SDK 2.4 is released?

If speed is the issue, then I'm sure most of us won't mind (it's better than nothing).

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 20228 - Posted: 22 Jan 2011 | 14:22:18 UTC

If one can accuse SK of being a fan then it would be a GPU-Grid fan ;)

The projects currently working on ATIs are comparably simple. That's why it was possible (for a single person in case of MW and CC) to optimize them this good and to get a stable app. The more you have to rely on calling complex functions from ATIs libraries the more you depend on them fixing their stuff. Which is not exactly easy. NVidia invested a huge amount of money into this software and still GPU-Grid does find quite some bugs in their code / drivers.

Anyway, the ATIs traditionally provide lot's of raw computation power per $ and good power efficiency. And in Cayman (HD69x0) this power became easier to use. So.. hopefully at some point it'll work. Don't put too much hope into OpenCL yet, though. It's not quite where people want to have it. And being able to run on anything doesn't mean one code path will be fast & efficient on anything..

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20227 - Posted: 22 Jan 2011 | 12:55:24 UTC - in response to Message 20225.

You have spent a lot of time crunching mathematical projects such as DNETC, PrimeGrid, and Collatz, but not so much for the more humanitarian projects. Do you put your Cypres to use at Folding@home? Perhaps you should have thought through your choice of GPU a bit better before purchasing it and limiting yourself to mathematical analyses.

While OpenCL is a cross-device open language, each device still requires working drivers.

[boinc.at] Nowi
Send message
Joined: 4 Sep 08
Posts: 44
Credit: 3,685,033
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwat
Message 20225 - Posted: 22 Jan 2011 | 7:25:36 UTC
Last modified: 22 Jan 2011 | 7:41:47 UTC

@skgiven: I apologize for my misinterpretational words. I wrote it "it looks like..." not "there is ...".

Of course I can´t help you, because my experiences in molecular modelling are more then 17 years ago and then these calculations were on an very early stage done on an 80286. And I´m not a programmer, only a lecturer and very interested in boinc and the projects. I want to spent my power and energy in "humanic" projects.

My thoughts were that from 2,111,537 boinc users (boinc-stats today) some have the knowledge to help this project. Of course akosh (I think that is the right spelling) was no gravitational expert "only" an interest user of E@H, who improved the performance of the apps. Now his optimations are included in the standard code of E@H apps.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 20221 - Posted: 21 Jan 2011 | 20:17:56 UTC - in response to Message 20219.

It’s easier to write simple apps and if ATI doesn’t adequately support the creation of more complex apps then its better to wait for the next SDK than to continue to debug and report endless errors, knowing that they will not be fixed until the next SDK, if at all.

I don’t know of another Boinc project that manipulates large molecules.
Perhaps there will be better support for Vista x86 and Win7 x86.


http://dnahome.cs.rpi.edu/dna/

This project is in an very early stage and I can not say for shure, that it is as complex as GPUGRID, I'm no scientist. And until now there is no GPU-app, but moderators say they plan to do that.

Another thing: as far as I understood, OpenCL is a hardware independent programming language. Since GDF did some tests with an openCL app and it looks like he has an ready to go app (a lot of effort has been put it to get there with large structural changes), he could release that as an nVidia-app. This gives the possibility to test the app in the field und to compare speeds, at least again the cuda-app. This could also solve some problems with the apps for fermi only and non-fermi apps.

MW did that; the openCL and the cuda-apps take exact the same time to complete (only a few seconds apart for a 17min task).

Please do not misinterprete my words, I really don't want to bother anyone. Your efford to develop the project is appreciated. It's my interest for this project that leds me to encourage you to distribute an ATI-app, even if it is unstable. I will not post angry words.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20219 - Posted: 21 Jan 2011 | 19:25:13 UTC - in response to Message 20217.

It’s easier to write simple apps and if ATI doesn’t adequately support the creation of more complex apps then its better to wait for the next SDK than to continue to debug and report endless errors, knowing that they will not be fixed until the next SDK, if at all.

I don’t know of another Boinc project that manipulates large molecules.
Perhaps there will be better support for Vista x86 and Win7 x86.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 20217 - Posted: 21 Jan 2011 | 17:33:47 UTC - in response to Message 20214.

MW, Collatz and Einstein dont run molecular dynamic simulations, their apps are more simplistic and very different.


Yes of course, you are right. But one word to MW/Collatz: both projects have Cuda and ATI apps. They are simpler for both types, nVidia AND Ati, but both apps are faster on ATI and use much less power also.

So don't let the ATI-apps die!

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 20216 - Posted: 21 Jan 2011 | 16:33:31 UTC - in response to Message 20214.

We have done our bit, it's just them that have to produce a working SDK. We have filed so many bug reports in the process.
We will keep the possibility to run ATI,a lot of effort has been put it to get there with large structural changes.
The last attempt we have planned to do is to see if Windows is more stable than Linux. All our tests were on Linux so far.
ATI gpus are quite good, especially the latest generation. So hopefully we should be able to run on them.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20214 - Posted: 21 Jan 2011 | 13:39:01 UTC - in response to Message 20213.

Looks like a real fan-boy speaking...

I expect you came to that conclusion by reading one post.

Is ATI only a mathematical GPU???

You answered yourself,
Akatosh: Einstein@Home
ClusterPhysik(Gipsel): Milkyway/Collatz


By the way: Why not using external support? there are excellent examples in the history of grid-computing

Funds and expert ATI OpenCL programmers familiar with molecular dynamics don’t grow on trees.
Might have been an excellent example if it came with a link, but I doubt it; MW, Collatz and Einstein dont run molecular dynamic simulations, their apps are more simplistic and very different.

[boinc.at] Nowi
Send message
Joined: 4 Sep 08
Posts: 44
Credit: 3,685,033
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwat
Message 20213 - Posted: 21 Jan 2011 | 13:13:46 UTC - in response to Message 20204.
Last modified: 21 Jan 2011 | 13:17:21 UTC

I’m not in the least surprised to hear that ATI support is insufficient, and I seriously doubt the merits of perusing an ATI application suitable to the complexities of this projects research. [..] Some projects have spent over a year trying to test and develop ATI apps only to find it is impossible/worthless.


Looks like a real fan-boy speaking...

Is ATI only a mathematical GPU???

By the way: Why not using external support? there are excellent examples in the history of grid-computing:

Akatosh: Einstein@Home
ClusterPhysik(Gipsel): Milkyway/Collatz

kbot
Send message
Joined: 18 Jan 11
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 20207 - Posted: 21 Jan 2011 | 1:51:01 UTC

to let the AMD-app die, is not a good choice!
the developers said, an AMD-app is possible, only the background driver is instable
so, if the app will work in future, you can (at least) double the amount of results ... there are so much people which have ATI/AMD cards only to make millions of credits per day and what is their result? they prove if a mathematical theory is right or wrong ...

imagine the millions of Watts (Energy) which are useless wasted, this energy could be used to heal people, to take away the pain of so much people

i agree with cenit, please tell us the problems, maybe someone has a solution ... a little patch can help millions of people...

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20204 - Posted: 20 Jan 2011 | 23:40:59 UTC - in response to Message 20203.
Last modified: 20 Jan 2011 | 23:46:39 UTC

I’m not in the least surprised to hear that ATI support is insufficient, and I seriously doubt the merits of perusing an ATI application suitable to the complexities of this projects research. I suggest you concentrate on CUDA for the foreseeable future; I know of several aspects to the present CUDA apps that could do with your attention (I spoke too soon, being addressed here). If an ATI app is not going to happen, put it to rest and let it be. Some projects have spent over a year trying to test and develop ATI apps only to find it is impossible/worthless.

cenit
Send message
Joined: 11 Nov 09
Posts: 23
Credit: 668,841
RAC: 0
Level
Gly
Scientific publications
wat
Message 20203 - Posted: 20 Jan 2011 | 23:11:53 UTC - in response to Message 20197.

SDK2.3 is still unstable unfortunately. I am getting a bit tired of this level of software support from ATI.

gdf

could you please share the problems you found with the community? I think that many of us would like to read something about the problems you are finding and how you're tackling them.
May I also say that maybe some problems rely on your side? I'm reading many positive comments about SDK 2.2 and 2.3... Surely a lot of work is still required for optimization, but an app should be working by now!

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 20197 - Posted: 20 Jan 2011 | 19:00:49 UTC - in response to Message 20182.

SDK2.3 is still unstable unfortunately. I am getting a bit tired of this level of software support from ATI.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20182 - Posted: 18 Jan 2011 | 10:11:10 UTC - in response to Message 20181.

There has been continuous speculation as to when ATI cards would be facilitated by this project. While another SDK has been released and the programmers might be working on this or at least planning to, it would be wise to accept that so far there have been lots of disappointments and false dawns. I don’t want to get anyone’s hope up on the ATI front because of this, but I also understand some of the hardware requirements might not be in place to support such an addition to the project. Also note that if ATI cards are supported at some stage, the initial phase of their introduction would be bumpy with lots of testing, problems and periods of inactivity. So don’t think someone can just flick a switch and suddenly GPUGrid will support ATI cards.

Over the last year and a half I replaced my ATI cards with NVidia cards, to support this project. While I can’t see many people leaving the mature NVidia apps for infant ATI apps, I agree there is a lack of Biological and Medical GPU projects for ATI crunchers and respect that many would be happy to crunch here even at 30 to 50% the performance of an NVidia card. However, the problem is not just development; it is implementation, maintenance and support and these things require funding, equipment, space, time and staff.

So while I look forward to having at least one ATI card in the future, it's likely to be at least a year before I fork out on one, and I do have a GTX260 and several GT240's that could be replaced should something come along.

kbot
Send message
Joined: 18 Jan 11
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 20181 - Posted: 18 Jan 2011 | 3:17:38 UTC

SDK 2.3 is out, some news about AMD app?

i would really love to use my 5870 + 6970 for bio crunching ...
these mathematical and astro projects are boring -.-

and its okay if the app is 2-3 times slower, i would do it for science, not for credits and i think many AMD cruncher think the same way...


greetings

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 20167 - Posted: 15 Jan 2011 | 15:16:38 UTC - in response to Message 20145.

Boudicca,

you must not install BOINC in "protected mode" in order to use GPUs. It's an option during installation which is active by default. Reinstall 6.10.58 (to the same paths, so you can keep all your configuration files and WUs), look for this option and deactivate it.
Regarding the driver: just go to amd.com and get the latest from there.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 303
Credit: 52,518,184
RAC: 21,933
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 20145 - Posted: 9 Jan 2011 | 2:24:15 UTC

Boudicca,

You may want to determine which model of HD 5570 card that is, find it on this list and click on that model name , then click on Details on the resulting page to get a page including more accurate information about what rating of power supply is needed. For example, if you scroll down in the details for the first model on the list, you'll see that it needs at least a 400 watt power supply (possibly less if the rest of the computer is on a separate power supply).

http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=HD+5570&x=11&y=25

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20137 - Posted: 7 Jan 2011 | 13:33:28 UTC - in response to Message 20122.

Boudicca, ATI cards do not presently work on this project (GPUGrid), hence the message, "No usable GPU's found".

Try Folding@home if you want to run a molecular folding project; they also do useful science.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 20134 - Posted: 7 Jan 2011 | 8:09:34 UTC - in response to Message 20122.

then again to V8.8 when I installed what they call, "Catalyst Control".

HOWEVER - - When starting up Boinc V6.10.58, it gives a message "No usable GPU's found".

Is that the only graphic card in your PC or is there also an on-board card? AMD (unlike nVidia) needs a monitor plugged in (or a dummy-plug) to be detected by BOINC.

Profile Boudicca
Send message
Joined: 5 Jan 11
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 20122 - Posted: 5 Jan 2011 | 21:01:44 UTC

excuse me if this has already been answered. Or if I should know better. But, i need a bit of help here.

Just installed a Radeon ATI HD 5570 GPU on my husband's computer (Intel Pentium dual cpu E2160). Increased the Windows experience rating from 3.8 to 5.9 immediately. Boinc says for ATI 5500 series GPU's you need to have driver V8.12. Windows vista went on the web and found ATI HD 5500 driver V8.791. So, I installed it. Next, when I installed the driver that came with the graphic card, it went to V8.77. Whatever :-| When *that* updated itself, it went to driver V8.791, then again to V8.8 when I installed what they call, "Catalyst Control".

HOWEVER - - When starting up Boinc V6.10.58, it gives a message "No usable GPU's found".

I like the 5570. It was cheap ($69.99), fits our Dell 530s SLIM profile case, however occupying the only single PCI X16 slot on the MB (the other three are PCI1, PCI2, and PCIX). It is passive (no fan - no noise)with massive heat fins on both sides of the card (and over the top - still fitting the SLIM profile case), and has a nifty utility for panning across multiple displays http://ati.amd.com

The stock Dell 530s has only a 250 watt PSU, but I have been assured (on another forum) that it will work OK. So far so good!

Two HIS Radeon ATI HD 5570's can be installed on a MB with two PCI X16 slots if you have them; but for that a 400 Watt PSU is recommended by ATI. Then, it lets you run, what they call, "Crossfire Mode" which, I suppose really zooms it up.

Did I say, - - When starting up Boinc V6.10.58, it gives a message "No usable GPU's found".

Am I barking up the wrong tree?


tuktuk1984
Avatar
Send message
Joined: 24 May 10
Posts: 3
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 20119 - Posted: 5 Jan 2011 | 18:39:07 UTC

Anny news for beta for ATi cards ?

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 20067 - Posted: 29 Dec 2010 | 7:56:13 UTC

So the beta for ATi cards is expected some time within January?
I'm really looking forward to it.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 20012 - Posted: 23 Dec 2010 | 11:21:04 UTC - in response to Message 20009.
Last modified: 23 Dec 2010 | 11:24:31 UTC

Boinc Projects are going to want to facilitate Open-CL:

The forthcoming Sandy Bridge contains 6 or 12 integrated Execution Units (an on-die GPU). These 32nm processors will replace the mainstream LGA 1155 i7’s, i5’s and i3’s. While they are unlikely to be useful to GPUGrid for GPU tasks, other projects such as Einstein might be able to utilize them to good effect, and I expect some big projects will use Open-CL in order to employ these CPU’s more fully. Intel have doubled the SSE data path and introduced Advanced Vector eXtensions (AVX), an advanced SSE.

Open-CL is knocking at the door.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 303
Credit: 52,518,184
RAC: 21,933
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 20009 - Posted: 23 Dec 2010 | 2:23:38 UTC

From Extra Terrestrial Apes:

The feature needs OpenCL, though
---

Looks like time to ask the BOINC to provide one of two possible new features, whichever one they'd prefer:

1. Add an OpenCL interface to BOINC.

2. Allow workunits to tell BOINC to reserve a GPU for them, then turn loose of that GPU in all other ways until the workunit finishes. The application program would then have to contain all the code needed for interfacing to that GPU, but at least that would be better than not being able to use it at all in the desired fashion.

Another idea to consider: Provide workunits with two independent parts taking about as much time each. On Fermi processors, use the new feature to run one a while, then while the other one is running, do the CPU interfacing for the first one. On older processors, run one after the other instead.

MAD!
Send message
Joined: 9 May 10
Posts: 6
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 19995 - Posted: 19 Dec 2010 | 1:09:58 UTC - in response to Message 19984.
Last modified: 19 Dec 2010 | 1:11:06 UTC

I especially like the new feature "asynchronous dispatch". It lets different apps be executed on the GPU at the same time. Fermi only allows different kernels from the same app. This could help with GPU utilisation, if the problem is using more Multiprocessors, or if the GPU is not used all the time. And it could make crunching GPUs more responsive, which is always welcome. Might even help with the CPU support / synchronization issues we're seeing at GPU-Grid now. By that I mean the CPU is not acting fast enough to keep the GPU busy - which would be much easier if the GPU was working on 2 WUs concurrently and with half a step phase difference. That way there'd be a rather large buffer or time window until the next CPU support would be required.



WOW, that feature alone is peaking my interest in GPU coding. I imagine this a result of AMD's Fusion direction.

More reading to do...

M.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19984 - Posted: 17 Dec 2010 | 23:15:52 UTC - in response to Message 19929.

I especially like the new feature "asynchronous dispatch". It lets different apps be executed on the GPU at the same time. Fermi only allows different kernels from the same app. This could help with GPU utilisation, if the problem is using more Multiprocessors, or if the GPU is not used all the time. And it could make crunching GPUs more responsive, which is always welcome. Might even help with the CPU support / synchronization issues we're seeing at GPU-Grid now. By that I mean the CPU is not acting fast enough to keep the GPU busy - which would be much easier if the GPU was working on 2 WUs concurrently and with half a step phase difference. That way there'd be a rather large buffer or time window until the next CPU support would be required.

The feature needs OpenCL, though.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 19929 - Posted: 15 Dec 2010 | 15:13:01 UTC - in response to Message 19911.
Last modified: 15 Dec 2010 | 15:13:19 UTC

http://www.semiaccurate.com/2010/12/14/look-amds-new-cayman6900-architecture/

The new AMD cards seem much better for compute.

gdf

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 41
Credit: 16,295,862
RAC: 163
Level
Pro
Scientific publications
watwatwatwatwat
Message 19911 - Posted: 14 Dec 2010 | 18:50:07 UTC - in response to Message 19908.

Yeah SDK 2.3 has been released.
http://www.geeks3d.com/20101213/opencl-ati-stream-sdk-v2-3-and-image-convolution-optimization-techniques/

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 19908 - Posted: 14 Dec 2010 | 17:07:17 UTC - in response to Message 19907.

No,
nothing until end of January but then we plan to release several things.
In the meanwhile, we can keep updating here on the performance.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19907 - Posted: 14 Dec 2010 | 16:54:56 UTC - in response to Message 19897.

Looks like 10.2 turned up yesterday (13th Dec).

The researchers may still want to wait until after the holiday period before beginning internal testing, and there is no guarantee they will be successful any time soon. They have to get it to work first and then optimize it to a reasonable level. I see little point in rushing an application into test release only to disappoint lots of ATI users, especially at this time of year.
If and when they are ready to go to Beta they will make the very welcome announcement.

tuktuk1984
Avatar
Send message
Joined: 24 May 10
Posts: 3
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 19897 - Posted: 14 Dec 2010 | 0:45:28 UTC

SDK2.3 is in the new Driver Pack 10.12

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 19892 - Posted: 13 Dec 2010 | 22:15:50 UTC - in response to Message 19889.

well....hmmm..it's out :)

It's out?
developer.amd.com has still 2.2 ready for download.
Anyway, maybe we get a christmas present ...

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 19888 - Posted: 13 Dec 2010 | 20:07:55 UTC - in response to Message 19877.

well....hmmm..it's out :)

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19877 - Posted: 12 Dec 2010 | 14:12:23 UTC - in response to Message 19876.

ATI's roadmap suggests they will release SDK2.3 this month.
Even if ATI are on target, I don't think the researchers will want to spend their holiday period debugging test apps. My guess is that it will be next year before any Beta's are released.

Wolfram1
Send message
Joined: 24 Aug 08
Posts: 45
Credit: 3,431,862
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 19876 - Posted: 12 Dec 2010 | 13:25:44 UTC - in response to Message 13411.

If you have a new 5870 ATI GPUs please accept beta work from us, in the following days we will try to upload a new ATI application.

gdf


Hi,

I am still waiting. Can I hope to get the bata in this year?

woever

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 19871 - Posted: 11 Dec 2010 | 17:06:33 UTC - in response to Message 19868.

We are waiting for SDK2.3.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19868 - Posted: 11 Dec 2010 | 14:19:19 UTC - in response to Message 19860.

There are no tasks for ATI cards, and as far as I am aware no immediate plans to use ATI cards.

tuktuk1984
Avatar
Send message
Joined: 24 May 10
Posts: 3
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 19860 - Posted: 11 Dec 2010 | 10:14:39 UTC

I have accept Beta work and it comes no Work

I use 2 ATI 5870 with the newest Driver and the Stream SDK 2.2

zpm
Avatar
Send message
Joined: 2 Mar 09
Posts: 159
Credit: 13,639,818
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 19574 - Posted: 20 Nov 2010 | 19:24:06 UTC - in response to Message 19533.

yea you kinda got that right.... Jack is the Main admin on the DD@H and the Developer of Hydrogen....

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19533 - Posted: 17 Nov 2010 | 20:16:47 UTC - in response to Message 19488.

Sorry, I have no idea what you're talking about. I don't know who Jack is.

I just know you said you were 100% sure it worked and then said it didn't. So I assumed you actually saw the machine with the HD2600 running such a WU. Which would actually be quite possible, since openCL can in principle run on anything as long as there's a driver mediating between the hardware and openCL. The first devices on which openCL was run were CPUs.

That's why I could imagine you saw that WU running, but openCL had determined the CPU as the only appropriate computing device and was using that one instead of the HD2600.

MrS
____________
Scanning for our furry friends since Jan 2002

zpm
Avatar
Send message
Joined: 2 Mar 09
Posts: 159
Credit: 13,639,818
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 19488 - Posted: 15 Nov 2010 | 17:45:42 UTC - in response to Message 19483.

but on jack computer it runs on the ati card...

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19483 - Posted: 15 Nov 2010 | 9:20:20 UTC - in response to Message 19480.

I guess it was running on the CPU instead?

MrS
____________
Scanning for our furry friends since Jan 2002

zpm
Avatar
Send message
Joined: 2 Mar 09
Posts: 159
Credit: 13,639,818
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 19480 - Posted: 14 Nov 2010 | 20:37:55 UTC - in response to Message 19475.

hello, me again!!!!

I have to admit... i was wrong... OpenCL will not run on a 2600 XT ati Card......

after going to pc wizard and looking PC activity and gpu temp. and Opencl line. i have come to the realization.

inserts foot into mouth and naws on foot.

zpm
Avatar
Send message
Joined: 2 Mar 09
Posts: 159
Credit: 13,639,818
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 19475 - Posted: 14 Nov 2010 | 18:47:28 UTC - in response to Message 19473.
Last modified: 14 Nov 2010 | 18:56:06 UTC

yes, i'm sure it's running on the 2600 xt... it's my computer...

If you go here, https://simtk.org/project/xml/downloads.xml?group_id=161 it will redirect to amd gpu page, which has support for opencl only gor 4xxx and up.
On the 2600 card, is it going to take around 20h for the WU described?



I know, i'm not the developer, if he made it compatible with older cards, then so be it, progress.... you should be congratulating us instead of criticizing it... ati may only support 4300 and up but we know the true reason for that.... ( They want to MAKE MONEY ON THE CARDS!!!!!)

but the truth it is opencl is compatible with the 2600 xt, my card running it proves it...


We know that our research, will take sometime... that's why we've been wanting to go to gpu apps as they are much faster than the single cpu...

and well have cuda to go along with ati.

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 19473 - Posted: 14 Nov 2010 | 18:14:17 UTC - in response to Message 19470.
Last modified: 14 Nov 2010 | 18:15:41 UTC

yes, i'm sure it's running on the 2600 xt... it's my computer...

If you go here, https://simtk.org/project/xml/downloads.xml?group_id=161 it will redirect to amd gpu page, which has support for opencl only gor 4xxx and up.
On the 2600 card, is it going to take around 20h for the WU described?

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19471 - Posted: 14 Nov 2010 | 18:06:58 UTC - in response to Message 19468.

Looks like they are just testing, and have no work at the minute.
http://boinc.drugdiscoveryathome.com/
http://hydrogenathome.org/

I expect they will attract many ATI card users when they are up and running. I think DD will be the first Biological research for ATI cards via Boinc.

zpm
Avatar
Send message
Joined: 2 Mar 09
Posts: 159
Credit: 13,639,818
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 19470 - Posted: 14 Nov 2010 | 17:49:29 UTC - in response to Message 19468.
Last modified: 14 Nov 2010 | 18:00:00 UTC

yes, i'm sure it's running on the 2600 xt... it's my computer...

OpenMM Loads Library - C:\0\cygintl-8.dll
OpenMM Loads Library - C:\0\cygreadline6.dll
OpenMM Loads Library - C:\0\cygwin1.dll
OpenMM Loads Library - C:\0\libfftw3f-3.dll
OpenMM Loads Library - C:\0\msvcp100.dll
OpenMM Loads Library - C:\0\msvcr100.dll
OpenMM Loads Library - C:\0\OpenMM.dll
OpenMM Loads Library - C:\0\OpenMMOpenCL.dll
Successfully loaded plugin cygwin1.dll
Successfully loaded plugin libfftw3f-3.dll
Successfully loaded plugin msvcp100.dll
Successfully loaded plugin msvcr100.dll
Successfully loaded plugin OpenMM.dll
Successfully loaded plugin OpenMMOpenCL.dll
OpenMM Platform: OpenCL
OpenCLDeviceIndex: 0
starting mdrun 'Protein in water'
100500 steps, 201.0 ps (continuing from step 100000, 200.0 ps).


and to answer your 2nd question... the ancient card the 2600 xt... will take forever it's been running since 9 am and it's still running and it's almost 1 pm.... but its trying to simulate 1 Nano second.

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 19468 - Posted: 14 Nov 2010 | 16:59:06 UTC - in response to Message 19466.

I'm pleased to announce that www.boinc.Drugdiscovery@home.com now has a working beta of a ati_opencl of MDRUN, it has been tested on a 5700 ati, a 2600 xt, as well the developers computer.

Are you sure opencl is running on 2600?
How long are the WUs on the 2600?

zpm
Avatar
Send message
Joined: 2 Mar 09
Posts: 159
Credit: 13,639,818
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 19466 - Posted: 14 Nov 2010 | 14:45:31 UTC - in response to Message 19406.

I'm pleased to announce that www.boinc.Drugdiscovery@home.com now has a working beta of a ati_opencl of MDRUN, it has been tested on a 5700 ati, a 2600 xt, as well the developers computer.
____________

I recommend Secunia PSI: http://secunia.com/vulnerability_scanning/personal/

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 19406 - Posted: 10 Nov 2010 | 9:19:14 UTC - in response to Message 19405.

For me, SWAN_SYNC increases GPU usage from 62% or so to 67% (Win 7 64 bit)



The problem is Windows 7. Xp and Linux give the best performance.

gdf

dyeman
Send message
Joined: 21 Mar 09
Posts: 34
Credit: 376,487,001
RAC: 900,429
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 19405 - Posted: 10 Nov 2010 | 9:16:45 UTC - in response to Message 19158.

For me, SWAN_SYNC increases GPU usage from 62% or so to 67% (Win 7 64 bit)

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19202 - Posted: 2 Nov 2010 | 22:14:14 UTC - in response to Message 19201.

http://www.gpugrid.net/forum_thread.php?id=2123

Profile Fred J. Verster
Send message
Joined: 1 Apr 09
Posts: 58
Credit: 35,833,978
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19201 - Posted: 2 Nov 2010 | 21:49:09 UTC - in response to Message 19158.
Last modified: 2 Nov 2010 | 21:56:52 UTC

If you set the environment variable SWAN_SYNC=0, in most tasks the use is around 97% in GTX480 in GPUGRID.


A command line parameter? I'm not very skilled in programming. More hardware.
(Old fashioned tubes(!)/transistor/mosFET/(Amstrad CPC128, first PC, BASIC & CPM1.1)(Nice)
But one is never too old to learn :)
Where do I put this? BOINC (Program)dir?
____________

Knight Who Says Ni N!

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 19158 - Posted: 31 Oct 2010 | 23:38:35 UTC - in response to Message 19155.

If you set the environment variable SWAN_SYNC=0, in most tasks the use is around 97% in GTX480 in GPUGRID.

Some month back, a few people from the Lunatics Crew, made several app.'s
for computing AstroPulse tasks and later MultiBeem, too. A CUDA version is already working for G80, G92 & G100 ands later GPU's.
I used the 2.1 SDK, it worked, but took awhile to get a decent GPU troughput, like =>50% or better.(On my 480 GPUgrid uses 50%) Einstein less then 5%! SETI 96%
Have a rig with an EAH4850 and a EAH5870, which is about 3 times faster.(Collatz C.,MW and DNETC)



Profile Fred J. Verster
Send message
Joined: 1 Apr 09
Posts: 58
Credit: 35,833,978
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19155 - Posted: 31 Oct 2010 | 22:46:55 UTC - in response to Message 19091.

Some month back, a few people from the Lunatics Crew, made several app.'s
for computing AstroPulse tasks and later MultiBeem, too. A CUDA version is already working for G80, G92 & G100 ands later GPU's.
I used the 2.1 SDK, it worked, but took awhile to get a decent GPU troughput, like =>50% or better.(On my 480 GPUgrid uses 50%) Einstein less then 5%! SETI 96%
Have a rig with an EAH4850 and a EAH5870, which is about 3 times faster.(Collatz C.,MW and DNETC)



____________

Knight Who Says Ni N!

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19091 - Posted: 29 Oct 2010 | 7:39:16 UTC - in response to Message 19083.

/me 2

Sadly that's not something GDF can influence.. and if he knew about it, he wouldn't be allowed to tell us ;)

MrS
____________
Scanning for our furry friends since Jan 2002

M J Harvey
Send message
Joined: 12 Feb 07
Posts: 9
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 19083 - Posted: 28 Oct 2010 | 22:14:36 UTC - in response to Message 19075.

Also, we want to see what's new in the 69xx series...

MJH

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19075 - Posted: 27 Oct 2010 | 19:48:25 UTC - in response to Message 19073.

Dec-2010

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 19074 - Posted: 27 Oct 2010 | 19:47:31 UTC - in response to Message 19073.

it should be before the end of the year I guess.
gdf

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 19073 - Posted: 27 Oct 2010 | 19:21:59 UTC - in response to Message 19032.

Have AMD given any kind of Indicative timeframe for 2.3? They'll never give a date, thats for sure, but they do at times give a vague Q1/Q2 etc kind of response.

Regards
Zy

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 103
Credit: 1,109,932,927
RAC: 10,480,238
Level
Met
Scientific publications
watwatwat
Message 19032 - Posted: 25 Oct 2010 | 9:54:23 UTC - in response to Message 19019.

The 2.2 SDK is still a bit too ropey. We're deferring any public app until 2.3 is available.

MJH

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19019 - Posted: 23 Oct 2010 | 18:52:41 UTC - in response to Message 19016.

what's up?


Sit back, relax and don't expect anything too soon.

MrS
____________
Scanning for our furry friends since Jan 2002

refla
Send message
Joined: 12 Feb 09
Posts: 9
Credit: 385,357
RAC: 0
Level

Scientific publications
watwatwatwat
Message 19016 - Posted: 23 Oct 2010 | 1:25:38 UTC

what's up?

when should ATi begin?

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 19012 - Posted: 22 Oct 2010 | 20:20:41 UTC - in response to Message 19008.

I dont know of any reason to require double precision for here, so only single precision would be required.


Agreed. Otherwise almost none of the nVidias would work here, or would have trouble keeping up with CPUs :p

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 19008 - Posted: 22 Oct 2010 | 15:26:17 UTC - in response to Message 19004.

I dont know of any reason to require double precision for here, so only single precision would be required.

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 19004 - Posted: 22 Oct 2010 | 9:43:42 UTC - in response to Message 18964.
Last modified: 22 Oct 2010 | 9:44:19 UTC

Sorry for asking again, but are you going to reguire single or double precision for ATI client?
Just read that they killed FP64 in 6850/6870
http://www.anandtech.com/show/3987/amds-radeon-6870-6850-renewing-competition-in-the-midrange-market/2

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 303
Credit: 52,518,184
RAC: 21,933
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 19001 - Posted: 22 Oct 2010 | 3:04:06 UTC
Last modified: 22 Oct 2010 | 3:11:59 UTC

An idea on how to produce a GPUGRID version for the AMD/ATI HD4nnn family, once you get suitable software support:

See if you can divide the tasks between upper level processors which can reach all of the graphics card's memory, and lower level processors which can't. Tell the lower level processors to do most of the calculations, but send their requests for access to memory they can't reach to the upper level processors.

Note - I'm NOT familiar enough with the AMD/ATI GPU architectures to tell if this is possible, or even what you've already tried.

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 18973 - Posted: 19 Oct 2010 | 12:15:42 UTC - in response to Message 18964.

Good News :)

Regards
Zy

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18964 - Posted: 18 Oct 2010 | 12:57:58 UTC - in response to Message 18953.

Yes, the level of stability is still not sufficient for the use in GPUGRID.
We waiting for the SDK2.3 and in the meanwhile optimizing the software.

gdf

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 119
Credit: 28,568,536
RAC: 25,059
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 18953 - Posted: 17 Oct 2010 | 18:20:03 UTC - in response to Message 18785.

Any progres?
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!

pvh
Send message
Joined: 17 Mar 10
Posts: 14
Credit: 177,591,256
RAC: 307,784
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 18785 - Posted: 29 Sep 2010 | 20:18:12 UTC - in response to Message 18752.

For those interested in protein research: Poem@Home is starting testing of GPU versions of their program, for both Nvidia and ATI cards. Name is POEM++; current version is 0.01.


Poem@Home currently only has POEM++ CPU clients out. Once they are convinced that these are stable they will start rolling out the GPU clients. Current estimate is that this will happen in a month or so.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 303
Credit: 52,518,184
RAC: 21,933
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 18752 - Posted: 22 Sep 2010 | 18:07:41 UTC - in response to Message 17199.
Last modified: 22 Sep 2010 | 18:14:21 UTC

From skgiven:

I think Folding just does protein folding, so it is different to this project; GPUGrid is unique in that it looks at super-structures (complexes of large molecules, such as lipid bylayer spanning pore proteins). Folding is more akin to Foldit in its research.


For those interested in protein research: Poem@Home is starting testing of GPU versions of their program, for both Nvidia and ATI cards. Name is POEM++; current version is 0.01.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 303
Credit: 52,518,184
RAC: 21,933
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 18751 - Posted: 22 Sep 2010 | 17:59:04 UTC - in response to Message 13540.

From Paladin*:

PS: As a side note Subscribing to a Thread is Pointless for me as I never receive a Notice by E-Mail even though I have my Preferences set to do so ... ???


Some of the ISPs in Germany are using a rather obsolete list of which other ISPs to block email access to - for example, it includes mine, whose outsourced newsgroups server USED to be a common posting point for newsgroups spam. In the years since then, it brought the newsgroups server in house with a newsgroups person more active in controlling spam, then was acquired by another ISP and dropped newsgroups entirely.

You may want to check if whatever ISP you have your email account on is on one of those obsolete lists.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 18746 - Posted: 22 Sep 2010 | 14:15:17 UTC - in response to Message 18743.

[quote
At any rate my 5870 will accept beta work units, if/when some are issued. It's been crunching a long time now, and curious about GPUgrid.

M.[/quote]

+1

Alexander

MAD!
Send message
Joined: 9 May 10
Posts: 6
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 18743 - Posted: 22 Sep 2010 | 3:38:47 UTC - in response to Message 18724.

Performance of a top ati card however is 3 times slower than a fermi at the moment but at least it runs.


Yay! I'll take it, back to basics. First make something work, then fine tune the process.

At any rate my 5870 will accept beta work units, if/when some are issued. It's been crunching a long time now, and curious about GPUgrid.

M.

_hiVe*
Send message
Joined: 18 Feb 09
Posts: 12
Credit: 2,109,577
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 18738 - Posted: 21 Sep 2010 | 20:30:01 UTC

Gosh, I came back to see if there was something new in here, but that wasn't the case :)
Well, somebody please wake me up from my cold slumber once a finalized app for ATI is out...perhaps by then, my short lived 5870 will come back from RMA :D

toms83
Send message
Joined: 12 Oct 09
Posts: 3
Credit: 3,176,377
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 18725 - Posted: 20 Sep 2010 | 20:20:53 UTC

Thanks for the info.

I'm glad, that finally will work on ati. Let us hope, the performance will be better in future. :)

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18724 - Posted: 20 Sep 2010 | 14:40:55 UTC - in response to Message 18655.

Yes,
we just presented the first results at a conference in USA.
I'll post it soon.

As I said before, probably there will be a beta release in Autumn and a production application from SDK2.3.

Performance of a top ati card however is 3 times slower than a fermi at the moment but at least it runs. We hope that with 2.3 will get better.

gdf

toms83
Send message
Joined: 12 Oct 09
Posts: 3
Credit: 3,176,377
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 18655 - Posted: 13 Sep 2010 | 13:29:35 UTC

Hey.

Any progress in the work?

M J Harvey
Send message
Joined: 12 Feb 07
Posts: 9
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 18348 - Posted: 14 Aug 2010 | 19:16:27 UTC - in response to Message 18330.

I think the researchers are down to a skeleton team during the summer holidays


Yep! Can't be wearing the white coats every day of the year...


MJH

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18330 - Posted: 12 Aug 2010 | 11:16:51 UTC - in response to Message 18329.

I think the researchers are down to a skeleton team during the summer holidays; enough staff to keep the tasks flowing, but little or no development work gets done.

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 18329 - Posted: 12 Aug 2010 | 6:49:02 UTC - in response to Message 18328.

sdk 2.2 out...

Mike
Send message
Joined: 8 Jun 09
Posts: 2
Credit: 26,447
RAC: 0
Level

Scientific publications
wat
Message 18328 - Posted: 12 Aug 2010 | 4:09:56 UTC - in response to Message 18313.

ATI 2.2 fixes most serious bugs and we will probably put out a test application for ATI. The performance is still poor for several reasons. ATI on one side and us on the other, will work to make it faster. Most likely, SDK 2.3 could be a better release performance wise, now that bugs are under controls.

gdf


I have a 5870..do you still plan on doing a beta release the end of summer or is it sooner?

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 18315 - Posted: 10 Aug 2010 | 9:01:49 UTC - in response to Message 18313.

ATI 2.2 fixes most serious bugs and we will probably put out a test application for ATI.
gdf


I've set my mainsys (with two ATI-cards) to accept GPUGRID wu's. If something more to do is required, I expect instructions somwhere.

Regards,

Alexander

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18313 - Posted: 9 Aug 2010 | 21:12:29 UTC - in response to Message 18310.

ATI 2.2 fixes most serious bugs and we will probably put out a test application for ATI. The performance is still poor for several reasons. ATI on one side and us on the other, will work to make it faster. Most likely, SDK 2.3 could be a better release performance wise, now that bugs are under controls.

gdf

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 18310 - Posted: 9 Aug 2010 | 18:50:11 UTC - in response to Message 18303.

Thank you for the update.

Alexander

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18303 - Posted: 9 Aug 2010 | 16:13:20 UTC - in response to Message 18302.

Perhaps the scientists will have another look at ATI cards in September or October, but they have a lot to do; optimize the GTX460, and forthcoming GTX475, try to get working apps for the GTS450, GTS445 and GTS440 (due out on the 12th), as well as the GF108 cards, also due out soon.

Last I heard they were perhaps about 50% of where they wanted to be WRT performance, but there were still bugs in the ATI code, that would require fixing before they could progress. If ATI have fixed their compiler and libraries then I would expect GPUGrid would be keen to restart work on an ATI application quite soon after the summer holidays.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 18302 - Posted: 9 Aug 2010 | 15:41:22 UTC - in response to Message 18264.

Does this mean that the new SDK is available and testing of the ATI-apps is going on?
I have 2992 ATI GFLOPS currently crunching mainly for milkyway and I would assign some of them to GPUGRID.

Regards,
Alexander

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 18264 - Posted: 7 Aug 2010 | 7:16:28 UTC
Last modified: 7 Aug 2010 | 7:17:22 UTC

http://support.amd.com/us/kbarticles/Pages/OpenCL11ATICat107UpdateDriver.aspx
Can someone confirm that Cat 10.7b with ATI Stream SDK 2.2 has support for OpenCL 1.1 for 4xxx series?

chumbucket843
Send message
Joined: 22 Jul 09
Posts: 21
Credit: 195
RAC: 0
Level

Scientific publications
wat
Message 17972 - Posted: 15 Jul 2010 | 1:19:23 UTC

i guess you could say that. local data share on rv770 was too restricted to meet opencl requirements.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 17938 - Posted: 12 Jul 2010 | 22:00:28 UTC

GPU-Grid doesn't need DP and this is hopefully not going to change anytime soon. At least one important difference between the 4000 and 5000 series is flexibility in memory management, if I understood correctly.

MrS
____________
Scanning for our furry friends since Jan 2002

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 17936 - Posted: 12 Jul 2010 | 17:00:14 UTC - in response to Message 17934.
Last modified: 12 Jul 2010 | 17:01:06 UTC

Why?


The design of the R700 processor used in the 4xxx series cards prevents the complete implementation of some features of OpenCL that are of critical importance for our application. (It's yesterday's technology, sorry!)

MJH


5xxx series has OpenCL 1.1 support. If you can get the 57xx series to work, at least you aren't going for DP :) .

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 103
Credit: 1,109,932,927
RAC: 10,480,238
Level
Met
Scientific publications
watwatwat
Message 17934 - Posted: 12 Jul 2010 | 14:04:40 UTC - in response to Message 17933.

Why?


The design of the R700 processor used in the 4xxx series cards prevents the complete implementation of some features of OpenCL that are of critical importance for our application. (It's yesterday's technology, sorry!)

MJH

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 748
Credit: 831,683,992
RAC: 2,262,206
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwat
Message 17933 - Posted: 12 Jul 2010 | 13:25:26 UTC - in response to Message 17928.

[quote]
And the 4xxx is right out.

MJH

Why? The 4850, 4870, 4890 & 4770 are generally faster than the 57xx and support DP to boot. The 57xx does not.

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17929 - Posted: 12 Jul 2010 | 6:40:17 UTC - in response to Message 17924.

Hi,

thank's to your and GDF's efford to explain, whats going on.

[quote}
Well, it probably would work, but it wouldn't be doing any GPU-Grid science ;)
[/quote]
I really like these words! Yes of course, beside the results in the project itself there should be a progress in gpu-computing as well!

I am looking forward to the first apps for my 5830 and maybe also sometimes for my old GTX260 too ...

Regards,
Alexander

MJH
Project administrator
Project developer
Project scientist
Send message
Joined: 12 Nov 07
Posts: 382
Credit: 27,130,153
RAC: 0
Level
Val
Scientific publications
wat
Message 17928 - Posted: 11 Jul 2010 | 22:01:01 UTC - in response to Message 17924.


may I ask why? To keep things as simple as possible initially or is there any important technical difference for GPU-Grid between e.g. HD58x0 and HD57x0?


The 58x0 is what we work with here. A 57x0 app may follow, probably, if all goes well and they prove to be fast enough. Lower 5xxx series cards will definitely be too slow to be useful. And the 4xxx is right out.

MJH

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 17924 - Posted: 11 Jul 2010 | 13:39:01 UTC

@MJH: may I ask why? To keep things as simple as possible initially or is there any important technical difference for GPU-Grid between e.g. HD58x0 and HD57x0?

@Werkstatt: I'd put it this way: imagine some guy compiling a "hello world" program in C++. Now there's another guy who's trying to build the perfect operating system. He may also be using C++, but can not get it quite as bug free and stable within a finite development time.

It's not that the other projects would be trivial (*), but they're using different functions, libraries and API calls. Apart from the fact that they're using a different API these differences in the code alone would be enough so that you can not simply say "Just do the same stuff the others did and it should work". Well, it probably would work, but it wouldn't be doing any GPU-Grid science ;)

Are these apps producing wrong results?


That's not the point (and is probably not the case). For GPU-Grid the problem is that using the current SDK the app is not stable. And if I understand correctly the problems have been traced down to API calls (functionality provided by ATI), so it's up to them to fix it.

MrS

(*) the code of CC and MW is not very complex, but certainly not trivial either
____________
Scanning for our furry friends since Jan 2002

M J Harvey
Send message
Joined: 12 Feb 07
Posts: 9
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 17921 - Posted: 10 Jul 2010 | 22:43:59 UTC - in response to Message 17919.

Hi,

We'll probably be in a position to start testing of an ATI application after SDK 2.2 is released in a month or so. It will require a 5800-series card initially.

MJH

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17919 - Posted: 10 Jul 2010 | 20:33:17 UTC - in response to Message 17918.

I dont think double precision is on the cards (it is not needed now), but that does not mean that a 5770 would not work. If ATI GPUs do eventually work here, that card should be a good one.

lkiller123
Send message
Joined: 24 Dec 09
Posts: 22
Credit: 15,875,809
RAC: 434
Level
Pro
Scientific publications
watwatwatwatwatwatwatwat
Message 17918 - Posted: 10 Jul 2010 | 20:03:25 UTC

If there will ever be an ATi support for GPUGrid, will double precision be required for the card? In other words, will I be able to run GPUGrid on 5770s?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17914 - Posted: 9 Jul 2010 | 10:57:07 UTC - in response to Message 17913.

CAL is the old programming model. It is deprecated. If you have an old application it makes sense to use it still, for a new one, it's better to wait for OpenCL.


gdf

Hi,
I really don't want to bother anyone, but I don't understand the troubles, that araise in this project. So let me please ask one question:
Seti has working apps for cuda and cal
Milkiway has apps for cuda and cal
collaz conjectur has apps for cuda and cal
einstein has apps for cuda
and they all work with my old gtx260/ATI-GPU's, some of them with my 250-GPU.
I understand, that the 260-192 has (build-in)troubles, but all ATI-cards work without any problem in all related projects.
Are these apps producing wrong results?

I'm not so happy to cruch for the search for numbers (collatz) and I'm waiting for a more intresting project (that may help more people than a handful of math freaks), that use my gpu's. Doesn't it make sense to try to connect the programmers of these project, just to exchange experience? They should have the same problems and they seem to have a solution.
Well, I'm not in the process of developing these apps, so my thoughts may be completely wrong. If so, then please forget my question.

Regards,
Alexander

Werkstatt
Send message
Joined: 23 May 09
Posts: 90
Credit: 9,064,060
RAC: 27,181
Level
Ser
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17913 - Posted: 9 Jul 2010 | 8:40:33 UTC

Hi,
I really don't want to bother anyone, but I don't understand the troubles, that araise in this project. So let me please ask one question:
Seti has working apps for cuda and cal
Milkiway has apps for cuda and cal
collaz conjectur has apps for cuda and cal
einstein has apps for cuda
and they all work with my old gtx260/ATI-GPU's, some of them with my 250-GPU.
I understand, that the 260-192 has (build-in)troubles, but all ATI-cards work without any problem in all related projects.
Are these apps producing wrong results?

I'm not so happy to cruch for the search for numbers (collatz) and I'm waiting for a more intresting project (that may help more people than a handful of math freaks), that use my gpu's. Doesn't it make sense to try to connect the programmers of these project, just to exchange experience? They should have the same problems and they seem to have a solution.
Well, I'm not in the process of developing these apps, so my thoughts may be completely wrong. If so, then please forget my question.

Regards,
Alexander

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 17912 - Posted: 8 Jul 2010 | 10:47:17 UTC - in response to Message 17749.
Last modified: 8 Jul 2010 | 10:49:51 UTC

We are waiting for ATI to fix their compiler and libraries.
gdf

From this post http://oscarbg.blogspot.com/2010/07/ati-stream-sdk-roadmap.html are you going to send some beta test in aug?

MAD!
Send message
Joined: 9 May 10
Posts: 6
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 17844 - Posted: 3 Jul 2010 | 6:38:17 UTC - in response to Message 17749.

Thank-you for the update. Good to read where developement is at and moving along...

M

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17749 - Posted: 29 Jun 2010 | 8:42:50 UTC - in response to Message 17745.
Last modified: 29 Jun 2010 | 8:49:01 UTC

We are waiting for ATI to fix their compiler and libraries. There is no much we can do. Some new tests are in progress these days.

Looking at the progress we are moving, I would expect to be in a stable, production situation with a OPENCL ATI application for GPUGRID before the end of the year. Of course beta work will start several months before. Maybe even just after the summer. All depends on them.

If they do a good job, I would expect an ATI 5870 to be within a factor 2 slower or faster than a GTX480. I know that the two extremes are quite different, but remember that few months ago, an ATI 5850 was 10 times slower than a GTX275 with ACEMD...

gdf

Profile liveonc
Avatar
Send message
Joined: 1 Jan 10
Posts: 292
Credit: 41,567,650
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwat
Message 17745 - Posted: 29 Jun 2010 | 4:21:18 UTC - in response to Message 17706.

Is this a possible way of cutting corners to getting something to work for ATI GPU's? http://www.fixstars.com/en/solutions/opencl/compiler.html
____________

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 17706 - Posted: 26 Jun 2010 | 20:23:20 UTC - in response to Message 17703.

Rest assured that if there is no update, it's not there yet ;)
(of course it would be nice to know whether there was progress or not)

MrS
____________
Scanning for our furry friends since Jan 2002

MAD!
Send message
Joined: 9 May 10
Posts: 6
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 17703 - Posted: 26 Jun 2010 | 5:37:15 UTC - in response to Message 17685.

Hmm, an update would be nice, it's been a few weeks since anything was said.

Profile rebirther
Avatar
Send message
Joined: 7 Jul 07
Posts: 53
Credit: 3,048,781
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 17685 - Posted: 21 Jun 2010 | 19:13:21 UTC - in response to Message 17683.

Any news on ATI cards yet? Did you get them working well? I have a couple 5970's I would like to use but...


Yes, I cannot wait to help this project with my ATI.

Brandon
Send message
Joined: 21 Apr 10
Posts: 1
Credit: 315,000
RAC: 0
Level

Scientific publications
watwat
Message 17683 - Posted: 21 Jun 2010 | 12:32:37 UTC

Any news on ATI cards yet? Did you get them working well? I have a couple 5970's I would like to use but...

sgh
Send message
Joined: 3 Feb 09
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 17477 - Posted: 31 May 2010 | 9:53:49 UTC - in response to Message 17248.

I have a 5750 ATI GPU running on linux, I would be happy to help with some testing if needed.

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 17248 - Posted: 24 May 2010 | 8:31:01 UTC

In so much as PoorBoy is loyal to Boinc, others are loyal to Folding. Perhaps if Folding and Boinc colaborated in terms of recognition (points/tasks/achievements) it would be more favourable, but Folding is a very big project, with hundreds of thousands of present crunchers.
They dont need you as much as GPUGrid does.


I may be wrong but it seem to me at one time Folding & BOINC tried to get together but never could Iron out the Details. From what I think I remember Folding wanted everything to much their way and the Deal never went thru ... Like I said I think I remember something like that taking place in the past ...
____________
STE\/E

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 17240 - Posted: 23 May 2010 | 22:42:43 UTC - in response to Message 17235.

It's a bit off topic, but ATI for SETI? I didn't hear about that heard yet. How's performance?

MrS

Still in Beta and it is for the AP tasks only as far as I have been able to find out. Again, the preliminary reports are that it is somewhat like the EaH application in the it uses more CPU than most would like though that is supposedly falling as they have worked the code.

The main coders are on Lunatics and the beta is not a full open beta yet ...

The hybrid version (for 2xxx and up) yes, still using too much CPU, like the EaH.
The OpenCL+Brook+ (for 4xxx and up) is using 98% GPU + ~5% CPU.

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwat
Message 17235 - Posted: 23 May 2010 | 17:23:58 UTC - in response to Message 17230.

It's a bit off topic, but ATI for SETI? I didn't hear about that heard yet. How's performance?

MrS

Still in Beta and it is for the AP tasks only as far as I have been able to find out. Again, the preliminary reports are that it is somewhat like the EaH application in the it uses more CPU than most would like though that is supposedly falling as they have worked the code.

The main coders are on Lunatics and the beta is not a full open beta yet ...

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 17230 - Posted: 22 May 2010 | 23:43:16 UTC - in response to Message 17225.

It's a bit off topic, but ATI for SETI? I didn't hear about that heard yet. How's performance?

MrS
____________
Scanning for our furry friends since Jan 2002

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 43
Credit: 199,994
RAC: 0
Level

Scientific publications
watwat
Message 17225 - Posted: 22 May 2010 | 17:05:10 UTC - in response to Message 17216.

I agree with the wrapper for folding, the one for dnetc looks good. The fah client is not so user friendly in comparison with boinc. I stopped folding a month ago due to the poor optimization of the ati client, and also the reluctance to give feedback to the 'new client'. I'm using my ati cards for seti and collatz (the low-level ones).

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 17216 - Posted: 22 May 2010 | 11:41:55 UTC

Would be nice if F@H introduced a BOINC wrapper, like e.g. yoyo@home. But their entire infrastructure is probably not very compatible with BOINC and they looked into it some time ago - and found it not be stable enough (or whatever other reason).

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17199 - Posted: 21 May 2010 | 11:11:05 UTC - in response to Message 17196.

I think Folding@home is a good project, and I have run a few hundred tasks there.
Their cutoff times are longer, so people that struggle to finish here, or use older cards might want to try folding. The GPU utilisation tends to be higher, so watch out for the heat.
My general opinion is that if the card cant work here it's not worth having, so I get rid of it, or just dont use it - it's down to project preferences, and I'm not interested in some sciences.

I think Folding just does protein folding, so it is different to this project; GPUGrid is unique in that it looks at super-structures (complexes of large molecules, such as lipid bylayer spanning pore proteins). Folding is more akin to Foldit in its research.

In so much as PoorBoy is loyal to Boinc, others are loyal to Folding. Perhaps if Folding and Boinc colaborated in terms of recognition (points/tasks/achievements) it would be more favourable, but Folding is a very big project, with hundreds of thousands of present crunchers.
They dont need you as much as GPUGrid does.

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 17196 - Posted: 21 May 2010 | 9:18:15 UTC - in response to Message 17081.

Reason enough for sure.

The reason I asked is that you seemed to have a particular animus towards Folding as indicated by your statement:

they'll sit Idle before I run Folding@home ...


Folding seems like a worthwhile project to me so I was wondering if you had a bad experience or something?


No, I've never had any dealings with Folding & from reading about it I'm sure it's a worthwhile project. But BOINC was what got me started (SETI Classic) & BOINC is what I will stay with. If Folding ever came under the BOINC Umbrella I would run it but until it does I won't a long as there are BOINC Projects around to run ...

____________
STE\/E

chumbucket843
Send message
Joined: 22 Jul 09
Posts: 21
Credit: 195
RAC: 0
Level

Scientific publications
wat
Message 17146 - Posted: 18 May 2010 | 21:13:37 UTC

just out of curiosity have you tested your openCL code on a CPU yet? AMD's opencl cpu compiler should be more mature and less buggy than their gpu compiler. it would also be a nice performance reference.

bigtuna
Volunteer moderator
Send message
Joined: 6 May 10
Posts: 80
Credit: 98,784,188
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17081 - Posted: 16 May 2010 | 0:19:49 UTC - in response to Message 17046.

Reason enough for sure.

The reason I asked is that you seemed to have a particular animus towards Folding as indicated by your statement:

they'll sit Idle before I run Folding@home ...


Folding seems like a worthwhile project to me so I was wondering if you had a bad experience or something?

STE\/E [BlackOps]
Send message
Joined: 18 Sep 08
Posts: 354
Credit: 250,952,560
RAC: 0
Level
Asn
Scientific publications
watwatwatwatwatwatwatwatwat
Message 17046 - Posted: 14 May 2010 | 17:44:53 UTC - in response to Message 17016.

poorboy
...I sure hope you get this going soon as Collatz & Milkyway are Complete Flops a lot of times like right now. Most of my ATI Cards are just sitting Idle & have been for quite awhile now because Collatz & Milkyway are down and may be down for ???


You could use them on Folding@home



I only run the BOINC Projects, they'll sit Idle before I run Folding@home ...


Really? Why is that?


Their my Box's and I choose to Support the BOINC Projects only, that's enough reason for me ... :)

____________
STE\/E

bigtuna
Volunteer moderator
Send message
Joined: 6 May 10
Posts: 80
Credit: 98,784,188
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17016 - Posted: 14 May 2010 | 0:24:08 UTC - in response to Message 13655.
Last modified: 14 May 2010 | 0:24:39 UTC

poorboy
...I sure hope you get this going soon as Collatz & Milkyway are Complete Flops a lot of times like right now. Most of my ATI Cards are just sitting Idle & have been for quite awhile now because Collatz & Milkyway are down and may be down for ???


You could use them on Folding@home



I only run the BOINC Projects, they'll sit Idle before I run Folding@home ...


Really? Why is that?

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 17014 - Posted: 13 May 2010 | 23:01:28 UTC

I second ETAs comment. I also became a recent convert, after 15 years with NVidia ever since NVidia started on the consumer end of things. The ATI performance on the non BOINC tasks I do has been amazing, it far outweighs expectation let alone hard comparative performance figures. On the BOINC ATI Projects I am now running, it flys ....

Kicking myself I didnt make the move earlier end of last year, but we all live and learn I guess - I had let the last 15 years colour my judgement. I certainly add my lowly voice to the chorus of thanks to AMD for the assist, lets hope it remains over time, and is substantive not transient.

Regards
Zy

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwat
Message 16997 - Posted: 13 May 2010 | 14:12:44 UTC - in response to Message 16996.

Great news, indeed! If you haven't done already, tell AMD that their help and hardware are greatly appreaciated in the world of BOINC :)

MrS

Even better point out that the more projects that run ATI applications that work decently, the more cards they will sell to this small, but intense market ... I know I am solidly in the ATI camp and am down to my last couple of Nvidia cards as we type/read ... and if they stay ahead of Nvidia will stay ATI for some time ...

I can see that some of the bigger ranchers have 20-30 systems and they tend to stick with the same cards if they can find one that works well ...

We are not a big as the gamers segment (yet), but 300,000 users world wide is not a trivial market to ignore either ...

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 16996 - Posted: 13 May 2010 | 13:33:44 UTC

Great news, indeed! If you haven't done already, tell AMD that their help and hardware are greatly appreaciated in the world of BOINC :)

MrS
____________
Scanning for our furry friends since Jan 2002

MAD!
Send message
Joined: 9 May 10
Posts: 6
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 16977 - Posted: 12 May 2010 | 5:05:08 UTC - in response to Message 16959.

The SDK2.1 is better but still has several bugs in parts which are required to run the code reliably. We must wait for the next release.

In the meanwhile we are working on the performance with the help of AMD.

gdf



New here with a 5870 open for testing if required. Awesome to read of the work going on here by everyone involved, and the help from AMD!

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 748
Credit: 831,683,992
RAC: 2,262,206
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwat
Message 16962 - Posted: 11 May 2010 | 15:18:46 UTC - in response to Message 16959.
Last modified: 11 May 2010 | 15:19:26 UTC

In the meanwhile we are working on the performance with the help of AMD.

Very good news. Thanks!

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwat
Message 16960 - Posted: 11 May 2010 | 14:53:42 UTC - in response to Message 16959.

The SDK2.1 is better but still has several bugs in parts which are required to run the code reliably. We must wait for the next release.

In the meanwhile we are working on the performance with the help of AMD.

gdf

Thats is good news once again ... and if they are helping, maybe they will see first hand the problems with the SDK and that will improve the chances that that will be fixed as well ...

Thanks for the update ...

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1861
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 16959 - Posted: 11 May 2010 | 14:38:44 UTC - in response to Message 16915.

The SDK2.1 is better but still has several bugs in parts which are required to run the code reliably. We must wait for the next release.

In the meanwhile we are working on the performance with the help of AMD.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3490
Credit: 830,811,232
RAC: 1,203,121
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 16915 - Posted: 8 May 2010 | 20:45:14 UTC - in response to Message 16913.
Last modified: 8 May 2010 | 20:48:05 UTC

... :)
That is funny. No, hilarious.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2450
Credit: 154,912,466
RAC: 266,900
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 16913 - Posted: 8 May 2010 | 20:25:10 UTC - in response to Message 16788.

Ugh, I would expect based on other projects that those numbers would be more like reversed ... :(


Just don't forget that ATIs 5-wide execution units give you more raw compute horse power per transistor, but are also harder to use efficiently. So you can't expect ATI to be faster for every algorithm.. otherwise nVidias design would be really stupid ;)

MrS
____________
Scanning for our furry friends since Jan 2002

X-COM
Send message
Joined: 5 May 10
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 16833 - Posted: 5 May 2010 | 15:52:00 UTC

Hi I have a Win 7 64bit computer with Radeon 5870. I would like to help if it is needed.

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwat