Advanced search

Message boards : Graphics cards (GPUs) : AMD HD7970 announced

Author Message
Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22793 - Posted: 22 Dec 2011 | 11:50:21 UTC
Last modified: 22 Dec 2011 | 12:46:32 UTC

http://www.anandtech.com/show/5265/the-radeon-hd-7970-recap

This has the new GCN cores and therefore should be optimal for Gpugrid. Although, we still have to test it...

gdf

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22794 - Posted: 22 Dec 2011 | 12:46:54 UTC - in response to Message 22793.

http://en.wikipedia.org/wiki/Comparison_of_AMD_graphics_processing_units#Southern_Islands_.287xxx.29_series

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22882 - Posted: 3 Jan 2012 | 21:04:14 UTC

It's an impressive piece of hardware. I'm curious what you can make out of it!

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22891 - Posted: 4 Jan 2012 | 20:34:23 UTC - in response to Message 22882.

Hope you get the hold of one soon and can build an app around GCN.
While the 7970 only comes with a 5% core freq. bump, there's a 33% increase in core config and it has a 63% increase in transistors, over the 6970, and double that of the 5870. Plenty of potential.
____________
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: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22899 - Posted: 5 Jan 2012 | 21:34:08 UTC - in response to Message 22891.

Not to forget the completely different core architecture.. which provides even more potential ;)

MrS
____________
Scanning for our furry friends since Jan 2002

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 42
Credit: 18,255,462
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 22923 - Posted: 9 Jan 2012 | 9:29:54 UTC
Last modified: 9 Jan 2012 | 9:30:58 UTC

I have a 7970 coming in tomorrow (Tuesday).
If you need a tester, I'm available. ;)

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 121
Credit: 59,836,411
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22924 - Posted: 9 Jan 2012 | 9:53:34 UTC

Test also in MilkyWay and CollatcConjecture and Moo!

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

Profile oldDirty
Avatar
Send message
Joined: 17 Jan 09
Posts: 22
Credit: 3,805,080
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwatwat
Message 22925 - Posted: 9 Jan 2012 | 12:18:09 UTC - in response to Message 22924.

Moo

What is Moo?

____________

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 22930 - Posted: 9 Jan 2012 | 17:18:17 UTC

Meet Moo .... :)

http://moowrap.net/index.php

Regards
Zy

Profile [SETI.USA]Tank_Master
Avatar
Send message
Joined: 8 Jul 07
Posts: 85
Credit: 67,463,387
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22931 - Posted: 10 Jan 2012 | 4:32:22 UTC

I to have ordered one, willing to test an app when it arrives! :)

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22932 - Posted: 10 Jan 2012 | 9:24:36 UTC - in response to Message 22923.

We don't have one yet.

gdf

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 45
Credit: 103,429,292
RAC: 0
Level
Cys
Scientific publications
watwatwatwat
Message 22934 - Posted: 10 Jan 2012 | 10:37:19 UTC

I will have one also, next year or maybe in two years.

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22935 - Posted: 10 Jan 2012 | 14:55:14 UTC - in response to Message 22934.

Unless the existing AMD app somehow works out of the box on an HD7970 and better than anticipated, the researchers would need to have one in the lab to build and test an app on. I'm guessing this year would be better than next year!
____________
FAQ's

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

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 22941 - Posted: 11 Jan 2012 | 3:46:12 UTC - in response to Message 22924.
Last modified: 11 Jan 2012 | 3:46:47 UTC

Test also in MilkyWay and CollatcConjecture and Moo!

:)


Moo, MW and Prime Grid dont run as is on a 7970.

Collatz appears to be running ok, WUs are completing, am awaiting validation by wingmen to see if it completed correctly.

MW and PrimeGrid PPS Sieve fall over quickly - in the first few seconds, so with luck shouldnt be too hard to find out the cause. The 7970s at least run on Collatz - await confirmation of validation - so its likely all will be well in the once there is time to focus on adjusting existing packages to keep things going until GCN can be properly brought into play.

See what the next couple of weeks brings ....

Regards
Zy

Profile [SETI.USA]Tank_Master
Avatar
Send message
Joined: 8 Jul 07
Posts: 85
Credit: 67,463,387
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 22949 - Posted: 11 Jan 2012 | 21:22:40 UTC
Last modified: 11 Jan 2012 | 21:23:09 UTC

just curious what driver you are using. Is it the one specifically for the 7000 series (only for win7 x64 atm)? - http://support.amd.com/us/kbarticles/Pages/radeon-7900-series.aspx

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22950 - Posted: 11 Jan 2012 | 22:10:57 UTC - in response to Message 22941.

Please keep us updated :)

MrS
____________
Scanning for our furry friends since Jan 2002

Werkstatt
Send message
Joined: 23 May 09
Posts: 121
Credit: 321,525,386
RAC: 177,358
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22956 - Posted: 12 Jan 2012 | 18:58:46 UTC

There is a very, very, very intresting thread in the MW forum:
http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2741#52421

Just to give you an impression:

It's interesting that even at this point the HD 7970 is running the NVidia optimized app at 3.5x faster than the top GTX580 card that I could find in the database. Pretty impressive and the HD 7970 also uses less power.

I think, it's time for action now!

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22957 - Posted: 12 Jan 2012 | 20:39:14 UTC - in response to Message 22956.

The AMD runs dp at 1/4th the sp speed, whereas GTX580 (and brothers) is artificially crippled to 1/8th the sp performance in dp. MW uses dp exclusively, whereas GPU-Grid is completely sp.

Furthermore the MW algorithm is very simple (yet efficient) in the way that it doesn't require any fancy memory managment. It runs perfect even on HD38x0, whereas HD 4xxx lacks even basic features for GPU-Grid.

Sure, the performance is impressive (as fast as my tuned HD6950 with the CAL app), but in GP-GPU programming so much depends on the actual code. So much in fact that there's not much point in saying "But look, this simple code runs great!"

MrS
____________
Scanning for our furry friends since Jan 2002

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 727,920,933
RAC: 155,858
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22958 - Posted: 13 Jan 2012 | 5:46:07 UTC - in response to Message 22891.

Hope you get the hold of one soon and can build an app around GCN.
While the 7970 only comes with a 5% core freq. bump, there's a 33% increase in core config and it has a 63% increase in transistors, over the 6970, and double that of the 5870. Plenty of potential.


What info do you have on its power requirements, and board size? Those are probably the most important items that will determine whether I get one.

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 22960 - Posted: 13 Jan 2012 | 8:54:40 UTC - in response to Message 22949.
Last modified: 13 Jan 2012 | 9:27:04 UTC

just curious what driver you are using. Is it the one specifically for the 7000 series (only for win7 x64 atm)? - http://support.amd.com/us/kbarticles/Pages/radeon-7900-series.aspx


Yes, thats the link, its the 7XXX driver that AMD point you at when going through their front page driver downloads. At present you end up on that 7970 speific download page to download it.

So far so good - its running the MW NVidia OpenCL app in +/- 52 seconds settings @1115/300. Runs nice, very quiet, temps on my box are +/- 74 degrees fan on auto, and air exiting from the box is lukewarm at worst, runs very cool. Utilisation on the app is 99% on both the 7970s running in crossfire. Its looking like the scaling is very good, early days, need some more comparitors, but its way way different from the scalling issue on 5XXX.

All in All, Its a good start right off the bat with no tweeking, running what is essentially an NVidia app.

Matt at MW will get a 7970 shortly, so we shall see how it goes once he gets a chance to get stuck in.

Regards
Zy

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 22961 - Posted: 13 Jan 2012 | 9:13:59 UTC - in response to Message 22958.
Last modified: 13 Jan 2012 | 9:46:40 UTC

Hope you get the hold of one soon and can build an app around GCN.
While the 7970 only comes with a 5% core freq. bump, there's a 33% increase in core config and it has a 63% increase in transistors, over the 6970, and double that of the 5870. Plenty of potential.


What info do you have on its power requirements, and board size? Those are probably the most important items that will determine whether I get one.


Board is nothing special - todays "standard" boards will run it. If buying new make sure you get PCI-E 3.0 lanes - not essential now, but will be useful over the life of the board. It runs fine over PCI-E 2.0, its not an issue or performance hit. I am running two 7970s in crossfire inside a small Lian-Li Armorsuit PC-P50 case, MSI 890FXA-GD70 motherboard, 1090T @3.6Ghz, 16Gb RAM @1600mhz. Runs smooth as silk, no issues.

I also have a 3960X full tower case, Rampage IV Extreme E-ATX, but its has 2x5970s in it and I am not messing with that, its slated for the 7990s when they come out. The 7970s stay in their long term home where they are.

Its drawing around 215w if going at full stretch 1125/1575 and power slider at max +20% so if any 5XXX or 6XXX or 5XX card runs on an existing board, 7XXX will run without exception, and at a very significant power saving. I would say I am running at about 180w power draw on the card as I am not hitting it full stretch at present.

The 215w is worst case no external voltage overclock, the absolute well stretched full blast all overclocked power draw with external voltage applied - the whole nine yards - is around 255-260w. Real world non overclock use around 180-215w. Budget for 300w, and that is way overkill giving a nice margin for use and max efficiency in the PSU. A bland real world budget figure for all us mere mortals not slamming with a mega overclock or LN2, is 200w leaving no margin.

Power draw is a non issue, I guarantee it draws less than existing similar class cards, so no hidden additional cost - no special motherboard either, nor CPU. Discount all that, if you are satisfied with performance levels, go buy one, there are no associated additional hardware factors lurking in the background - other than a significant power saving.

Anyone with bottle-neck worrys, read the review below which essencially says if you have a modern 4 core class CPU as minimum, real world bottle necking doesnt exist.

http://www.guru3d.com/article/radeon-hd-7970-cpu-scaling-performance-review/

EDIT: There is a case that says wait for the 1.5Gb card memory versions. The first release cards are all full blown 3Gb GDDR5 memory to give performance on very high res triple/six stacked 2550x1440 monitors for power gaming & multi panel monitor needs. If there is no requirement for that, some will not need 3Gb, so the 1.5Gb cards will be just the job, cheaper, and no performance hit. I have no idea when the 1.5Gb version are due, or which 7XXX variants they will apply to. Milegae on this is subjective of course, as its dependent on poersonal use, aims and intentions. But its a nice option for AMD to put out there.

Regards
Zy

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22965 - Posted: 13 Jan 2012 | 13:38:14 UTC - in response to Message 22961.

Lots of positives raised about the 7970. So I won't go there :)

Good point about the 1.5GB card. It would definitely be less expensive, and there is no reason why it would crunch less on a high GPU utilizing project. On projects that require you to build an app_info file to obtain higher GPU utilization the more RAM the better, but even then 1.5GB would be enough to run several tasks simultaneously.

A few more reasons to wait:

    NVidia should be releasing 600 series cards within the next few months. These should offer up alternatives to the 7970 and should force the high prices down (as will better yields).
    It's likely that AMD will have a refresh/interim series update that uses XDR2, probably after NVidia's release.
    Existing apps still run well on existing 5000 & 6000 series hardware. The apps will probably have to be developed/enhanced (mature) to gain greater performances for the 7900series.


____________
FAQ's

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

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22967 - Posted: 13 Jan 2012 | 15:52:21 UTC - in response to Message 22965.

So, AMD just told us that they cannot gives us even one card as they have sent everything to retail. So, we don't expect to have a card soon.

gdf

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22968 - Posted: 13 Jan 2012 | 17:11:44 UTC - in response to Message 22967.
Last modified: 13 Jan 2012 | 17:25:10 UTC

AMD just told us that they cannot gives us even one card

What a Pity!

At £440 ($550/499Euro) they are not cheap.

The 7950 is reportedly due out on 31st Jan, but it's still likely to cost around $449 (~410Euro). I think the lesser AMD 7000 GPU's are due out in Feb, but if they are not GCN they won't be any use.

Hope you asked NVidia too; their GK-104 Kepler isn't far off.
____________
FAQ's

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

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 22970 - Posted: 13 Jan 2012 | 18:02:08 UTC - in response to Message 22968.

Ok, good news, we seem to be able to get hold on one in a couple of weeks.

gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22975 - Posted: 14 Jan 2012 | 12:32:47 UTC

@Zy: my unlocked HD6950 @ 900 MHz ist jsut as fast at MW as your HD7970, although I'm running the "native" CAL app, of course. And I don't think I'm drawing more power while doing so, the entire rig draws < 300 W off the wall, with an i7 @ 4.0 GHz crunching on 7 other threads.

However, still very impressive performance of the newcomer, with the nVidia app! Ideally in OpenCL you program and it works on any hardware.. but we all know how fragile performance on GPUs is still. There's certainly more potential in GCN.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 22998 - Posted: 18 Jan 2012 | 14:01:36 UTC - in response to Message 22975.

Yup - as you point out its the effect of OpenCL apps. I reckon that will be more widespread as OpenCL takes hold over time - red or green team - but for now it has a greater impact on perception.

Hardcop did a no holds barred - as far as the new AMD hardware locks would allow them - voltage tweek overclocking session, and they were bowled over by the result, the untapped power still there under the hood is stunning. I've seen others do similar sessions also coming to the same conclusion. If you havent seen it, the session is at:

http://hardocp.com/article/2012/01/09/amd_radeon_hd_7970_overclocking_performance_review/1

I suspect similar levels of performance increase will come out of the green team, the effect of 28nm. As always the final conclusions and preferences will come from software developments, and the use an individual wants to put the card to. I've said it before, but .... 2012 is gonna be fun :)

Regards
Zy

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 23003 - Posted: 18 Jan 2012 | 19:45:55 UTC - in response to Message 22998.

Their OC at stock voltage is cool (but hey, anyone can do this ;). However, for 1.26 GHz they need 1.30 V - that's massive! The last time I applied that much to a chip 24/7 was at 65 nm :p And my Cayman hums along nicely at 1.10 V.

Anyway, with watercooling this might not be such a bad proposition, if you've got a limited amount of PCIe slots and want to maximize your credits. Drawing 100 W more for 33% more performance out of such a high end card.. if you calculate that as e.g. 100k RAC more for 100 W more, then it doesn't look all that bad. Which is why I'm not running my Cayman at e.g. 0.9 V and 700 - 800 MHz, even though it would be more efficient.

And achieving this high clock speed on a new design bodes well for all future GCN cards!

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 23005 - Posted: 18 Jan 2012 | 21:02:13 UTC

ack .... the designs go in leaps and bounds on every shrink. Its probably only a couple of shrinks away before people start yelling "my hologram is better than your hologram" rofl :)

Regards
Zy

Profile [SETI.USA]Tank_Master
Avatar
Send message
Joined: 8 Jul 07
Posts: 85
Credit: 67,463,387
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 23130 - Posted: 25 Jan 2012 | 20:15:27 UTC

Any update on when we may be able to start testing the 7970 here?

Profile Fred J. Verster
Send message
Joined: 1 Apr 09
Posts: 58
Credit: 35,833,978
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 23133 - Posted: 25 Jan 2012 | 22:42:28 UTC - in response to Message 23130.

Hi, after adding GPUGrid on my Q6600+GTX470, I also have an I7-2600+ 2 HD5870
GPUs.
It's my 'work puter', has 3500GByte storage and 8GByte DDR3 (1388MHz;base clock is @ 102MHz, PSU is 1KWatt, other QX9650+GTX480, has a 650Watt PSU, same as the one above. (WIN XP, x86 & x64 on the NVidia hosts and WIN 7 64bit on the I7-2600)

It's now doing Rosetta, Malaria on CPU and SETI runs 6 WUs on 2 HD5870 GPUs, using OpenCL.
1 core free to load and write the result file.
Is there already a OpenCL version for ATI (AMD) GPUs.?
I've not been able to read all forum info on this.



____________

Knight Who Says Ni N!

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 23135 - Posted: 25 Jan 2012 | 23:18:29 UTC - in response to Message 23133.

We might receive a GCN card soon to test.
gdf

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 23147 - Posted: 26 Jan 2012 | 21:05:18 UTC - in response to Message 23133.

Not yet, Fred.

MrS
____________
Scanning for our furry friends since Jan 2002

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 42
Credit: 18,255,462
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 23168 - Posted: 28 Jan 2012 | 18:03:23 UTC

Hurry up and give these guys a card, AMD.
Pleeeeeeease. :p

JLConawayII
Send message
Joined: 31 May 10
Posts: 48
Credit: 28,893,779
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 23170 - Posted: 28 Jan 2012 | 22:47:10 UTC

Or you could just buy one. Send some of the undergrads out for a weekend carwash and use the proceeds to purchase a 7970.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1957
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 23172 - Posted: 29 Jan 2012 | 9:09:32 UTC - in response to Message 23170.

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.

gdf

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 42
Credit: 18,255,462
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 23183 - Posted: 30 Jan 2012 | 8:31:57 UTC

Awesome!
Very much looking forward to it. :)

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 23185 - Posted: 30 Jan 2012 | 13:32:01 UTC - in response to Message 23183.

I read that BOINC 7.0.11 is required to identify the HD 7970.
____________
FAQ's

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

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 23187 - Posted: 30 Jan 2012 | 15:13:12 UTC

Its 7.0.3 and above. I am running 7.0.8 and its fine. I've seen 7970s at POEM running 7.0.3, so as long as 7.X.X is loaded - should be fine

Regards
Zy

Profile Fred J. Verster
Send message
Joined: 1 Apr 09
Posts: 58
Credit: 35,833,978
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 23188 - Posted: 30 Jan 2012 | 16:17:04 UTC - in response to Message 23187.

Anyway, I hope there will be an OpenCL (or BROOK?), soon ;-)
Thanks for your reply, as GPUgrid is my main project, now.


____________

Knight Who Says Ni N!

Profile Zydor
Send message
Joined: 8 Feb 09
Posts: 252
Credit: 1,309,451
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 23189 - Posted: 30 Jan 2012 | 16:43:00 UTC - in response to Message 23188.
Last modified: 30 Jan 2012 | 16:45:57 UTC

Suspect lots of us will be joining you Fred .... an AMD app at GPUGRID long been a hope in many minds - and as long as AMD come up with the goods on 79XX compared to NVidia, lots will.

If they dont, suspect many will change to NVidia in a heartbeat - I certainly will. 28nm is a defining moment, its a huge step change from 40nm, and if NVidia screw this up aka 3.XX and 4.XX .... I suspect they will be in trouble. I hope not, high prices on cards in the last two years is because basicly AMD had it all their own way for years, and that didnt change until 560/570/580 came out, they were a last min saviour for NVidia.

We need good offerings from NVidia in the markeplace ....

Regards
Zy

STE\/E
Send message
Joined: 18 Sep 08
Posts: 368
Credit: 313,909,798
RAC: 486,126
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 23190 - Posted: 30 Jan 2012 | 23:02:59 UTC - in response to Message 23187.

Its 7.0.3 and above. I am running 7.0.8 and its fine. I've seen 7970s at POEM running 7.0.3, so as long as 7.X.X is loaded - should be fine

Regards
Zy


Yes, I'm able to run both POEM & Milkyway with the 7.0.3 Client ...

____________
STE\/E

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 42
Credit: 18,255,462
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 23233 - Posted: 3 Feb 2012 | 13:49:49 UTC

Any updates? :)

peeticek_LubosKrutek
Send message
Joined: 30 Nov 08
Posts: 7
Credit: 62,377,145
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 23277 - Posted: 7 Feb 2012 | 9:27:36 UTC

Ohlalalaaaa :) - gpugrid on AMD gpu is just my DREAM..

In that moment, when you`ll release a stable gpugrid app for 7950 ill buy one!

Good luck guys!

p.

Evil Penguin
Avatar
Send message
Joined: 15 Jan 10
Posts: 42
Credit: 18,255,462
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 24014 - Posted: 17 Mar 2012 | 21:51:25 UTC

Any progress on the AMD side of things?

cristipurdel
Send message
Joined: 31 Mar 10
Posts: 45
Credit: 103,429,292
RAC: 0
Level
Cys
Scientific publications
watwatwatwat
Message 24051 - Posted: 20 Mar 2012 | 10:56:52 UTC - in response to Message 24014.

Any progress on the AMD side of things?


AMD HD7970 announced

Old man
Send message
Joined: 24 Jan 09
Posts: 42
Credit: 16,676,387
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27173 - Posted: 29 Oct 2012 | 11:51:09 UTC - in response to Message 24014.

Any progress on the AMD side of things?


Up.

I have tested donate@home but i like run my ati card here. Any news from ati app?

Werkstatt
Send message
Joined: 23 May 09
Posts: 121
Credit: 321,525,386
RAC: 177,358
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27483 - Posted: 30 Nov 2012 | 20:43:22 UTC

Hello dear all,

just want to post a link to the top hosts list @ einstein:
http://einstein.phys.uwm.edu/top_hosts.php

AMD is leading there and at positions 3,4 and 6.

And here? Still talking about poor performance of AMD openCL?

Reality has overtaken you, developers!

Alexander

5pot
Send message
Joined: 8 Mar 12
Posts: 411
Credit: 2,083,882,218
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27484 - Posted: 30 Nov 2012 | 22:59:33 UTC
Last modified: 30 Nov 2012 | 23:02:40 UTC

Lol

Top supporters != most efficient

I've run Einstein, and I can tell you that their CUDA app is more efficient.

Edit: also, Jeroen is #2 because he is running at pcie 2.0, since you cannot run 3.0 with Kepler on SB-E on Linux. The reg hack windows users can use obviously does not work for linux.

cenit
Send message
Joined: 11 Nov 09
Posts: 23
Credit: 668,841
RAC: 0
Level
Gly
Scientific publications
watwat
Message 27485 - Posted: 1 Dec 2012 | 3:55:12 UTC - in response to Message 27484.

Lol

Top supporters != most efficient

I've run Einstein, and I can tell you that their CUDA app is more efficient.

Edit: also, Jeroen is #2 because he is running at pcie 2.0, since you cannot run 3.0 with Kepler on SB-E on Linux. The reg hack windows users can use obviously does not work for linux.


anyway, even if inefficient, many users have requested an amd OpenCL app. Maybe, as in einstein, even if inefficiently, that app could take the top spot (and in the meanwhile you can always improve your code...), thanks to the extreme performances of GCN computationally wise (much more than Kepler, btw, if you're honest - just GK110 can be comparable but it does not exist in any consumer card for the moment).
But it seems that here the devs are absolutely not interested. And this is a pity.

5pot
Send message
Joined: 8 Mar 12
Posts: 411
Credit: 2,083,882,218
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27486 - Posted: 1 Dec 2012 | 4:07:14 UTC
Last modified: 1 Dec 2012 | 4:10:29 UTC

But that app has the top spot because the 1 other person there, who would blow past the other guy, is on Linux.

Im not saying that ATI GPUs are not decent, it's just that OpenCl isn't.

EDIT: And the difference between PCIe 2.0 and 3.0 on Einstein app is something like 20% better runtimes when running multiple tasks. And computational performance does not always equate to better performance as can be seen at Einstein.

cenit
Send message
Joined: 11 Nov 09
Posts: 23
Credit: 668,841
RAC: 0
Level
Gly
Scientific publications
watwat
Message 27487 - Posted: 1 Dec 2012 | 4:14:08 UTC - in response to Message 27486.

But that app has the top spot because the 1 other person there, who would blow past the other guy, is on Linux.

Im not saying that ATI GPUs are not strong, it's just that OpenCl isn't.

EDIT: And the difference between PCIe 2.0 and 3.0 on Einstein app is something like 20% better runtimes when running multiple tasks.


OpenCL is a language. It cannot be strong or weak.
The problem, maybe, is that so many very well developed libraries exist only for cuda. And many devs have an expertise only in that language.
And also the tools were much more developed for cuda, from the compiler to the debugger. Things have improved dramatically, but here it seems that they're never interested. Every gen that comes out they say: this is the perfect one, we will look into it. They also said, once, "we have an app ready, please enable beta". But nothing ever happened then. No real exchange of true informations. I'm sure that in this community there're also some good programmers. Just release the code opensource and someone can also help you.

Werkstatt
Send message
Joined: 23 May 09
Posts: 121
Credit: 321,525,386
RAC: 177,358
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27489 - Posted: 1 Dec 2012 | 7:25:07 UTC - in response to Message 27487.


OpenCL is a language. It cannot be strong or weak.
The problem, maybe, is that so many very well developed libraries exist only for cuda. And many devs have an expertise only in that language.
And also the tools were much more developed for cuda, from the compiler to the debugger. Things have improved dramatically, but here it seems that they're never interested. Every gen that comes out they say: this is the perfect one, we will look into it. They also said, once, "we have an app ready, please enable beta". But nothing ever happened then. No real exchange of true informations. I'm sure that in this community there're also some good programmers. Just release the code opensource and someone can also help you.


You got the point, cenit.
First time I asked for an app for ATI-cards the HD5xxx series was state of the art.
When the HD6xxx series was up to date, they posted, their openCL app is working, but performance is poor. But still no beta apps.
Now we have the HD7xxx in the market, next rework (of the GCN) is announced, openCL has proofed to be fast and reliable, and the reaction here: not existent.
It was the time when I installed my first HD5xxx card when they advised us to enable beta apps.

[boinc.at] Nowi
Send message
Joined: 4 Sep 08
Posts: 44
Credit: 3,685,033
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwat
Message 27490 - Posted: 1 Dec 2012 | 9:32:49 UTC

I gave up the hope for OpenCL-app since the start of donate@home. IMHO the question is: Is it wise to release an APP for AMD, when you can earn money from the performance of AMD at donate? That will canabalize donate, which is only a support project to fund GPUGrid.

This question can only be answered by the devs.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 7,520
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27492 - Posted: 1 Dec 2012 | 9:53:07 UTC - in response to Message 27487.

OpenCL is a language. It cannot be strong or weak.
The problem, maybe, is that so many very well developed libraries exist only for cuda. And many devs have an expertise only in that language.
And also the tools were much more developed for cuda, from the compiler to the debugger. Things have improved dramatically, but here it seems that they're never interested.

This is not true. Even your next statement disproves it.

Every gen that comes out they say: this is the perfect one, we will look into it. They also said, once, "we have an app ready, please enable beta". But nothing ever happened then.

While I can recall the developers said that they have an app ready, they said the same time it's not reliable and/or very slow (compared to the CUDA app), therefore it's not ready to release even as beta.

No real exchange of true informations. I'm sure that in this community there're also some good programmers. Just release the code opensource and someone can also help you.

According to this post, the code of the ACEMD client won't be open source.

Profile Christopher Herr
Send message
Joined: 1 Apr 10
Posts: 13
Credit: 106,905,353
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 27493 - Posted: 1 Dec 2012 | 10:03:27 UTC - in response to Message 27490.

I gave up the hope for OpenCL-app since the start of donate@home. IMHO the question is: Is it wise to release an APP for AMD, when you can earn money from the performance of AMD at donate? That will canabalize donate, which is only a support project to fund GPUGrid.

This question can only be answered by the devs.


Yes, that is exactly what I have been thinking since Donate was rumored and since Kepler was glorified; now they found an excuse for all AMD owners, as poor as that is, that they gave the volunteers an AMD application to support this project, well, indirectly by running none science related bitcoining at yet another project spoiling the volunteers with credits instead of doing valuable science. The developers are simply too inexperienced with AMD OpenCL, so they blame it on everything else, that is the only thing I can conclude from that.
I am not an OpenCL developer, but then again this is not my task, as I am donating my hardware, power and time to a project refusing to adapt to another programming platform than CUDA...
This is all I have to say about that.
Now the CUDA fanboys are free to call me names, insult me and bash me again, as that is what happens every time AMD owners complain rightfully... :))
____________

Profile dskagcommunity
Avatar
Send message
Joined: 28 Apr 11
Posts: 456
Credit: 817,865,789
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27495 - Posted: 1 Dec 2012 | 19:52:18 UTC

I personally would switch my 7950 instantly after the end of WCG HCC GPU to GPUGrid IF they would run OpenCL. But not to donate (only tested it there in the downtime of WCG last week). It would bring me very much Points (620k per day) but im not a credit hunter, so i want to run extra buyed dedicated BOINC Hardware to science.

Its a good way to fund the project by volunteers who has problems to donate to boinc projects when they have to much hardware lieing around ;) I tried one time to donate money over my creditcard to, i dont know was it seti or something?, and mastercard banned my creditcard (not reversible!) because they think it was a unwished transaction. i dont want to do that anymore again -_-

Pllzz make an OpenCL Support, even when it is not the best performanced one. Im struggling a little bit what i can make with the 7950 after HCC. At the time i only can use it for...MW then..I dont have the hardware to support a 7950 on POEM :/ *sign*
____________
DSKAG Austria Research Team: http://www.research.dskag.at



5pot
Send message
Joined: 8 Mar 12
Posts: 411
Credit: 2,083,882,218
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27496 - Posted: 1 Dec 2012 | 20:37:37 UTC

I would suggest Einstein.

I'm currently in a bind of wishing to buy a 7950 for HCC, but at the same time, I feel I would have no place to take said card after HCC GPU app ends or during its "dry spells"

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27497 - Posted: 1 Dec 2012 | 23:23:06 UTC

Guys.. let's stay cool, take a step back and recapitulate what we're dealing with here. GPU-Grid

- has limited ressources and needs to decide how to make the best use of them
- requires manpower on the science side of things: analyse the results, formulate new problems to solve, write publications, supervise thesis'.. all the usual stuff
- requires manpower to improve the CUDA code
- requires manpower to maintain the CUDA code (work around issues popping up with new driver releases and support of many combinations of cards, driver and OS)

They built an AMD app, but it performed poorly due to hardware and software issues. The hardware side may be solved by GCN, but the software wasn't some time ago. The software certainly has improved, but I don't know if the encountered stability issues are solved yet. And someone made something work in OpenCL - fine, but they're using different code, function calls & libraries. This is nice to know, but doesn't guarantee the GPU-Grid code will be fine.

In any way, developing and supporting an AMD code path using other development tools and a different language will require additional manpower. It even introduces much more complex cross-platform testing.

Surely this can be done - given enough manpower. But GDF has said in the past that they'd rather properly support nVidia cards than nVidia and AMD in a mediocre way.

Of course one would prefer perfect support & optimization for both.. but this is not a realistic option, as it seems. You're of course welcome to report on OpenCL progress and ask about the current state. But don't get mad at the developers if they can't realize everything you want.

MrS
____________
Scanning for our furry friends since Jan 2002

TheFiend
Send message
Joined: 26 Aug 11
Posts: 99
Credit: 2,500,112,138
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27502 - Posted: 2 Dec 2012 | 11:01:12 UTC - in response to Message 27497.

Guys.. let's stay cool, take a step back and recapitulate what we're dealing with here. GPU-Grid

- has limited ressources and needs to decide how to make the best use of them
- requires manpower on the science side of things: analyse the results, formulate new problems to solve, write publications, supervise thesis'.. all the usual stuff
- requires manpower to improve the CUDA code
- requires manpower to maintain the CUDA code (work around issues popping up with new driver releases and support of many combinations of cards, driver and OS)

GDF has said in the past that they'd rather properly support nVidia cards than nVidia and AMD in a mediocre way.

Of course one would prefer perfect support & optimization for both.. but this is not a realistic option, as it seems. You're of course welcome to report on OpenCL progress and ask about the current state. But don't get mad at the developers if they can't realize everything you want.

MrS


Well said MrS.


Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 7,520
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27508 - Posted: 3 Dec 2012 | 0:02:28 UTC

I've got two things on my mind:

1. Why does anyone think, that a cheaper product (HW&SW) with a shorter development time period can do the same work as a much more mature (and therefore more expensive) one?
2. Why do some volunteers think that the researchers volunteered to please them?

Profile MJH
Project administrator
Project developer
Project scientist
Send message
Joined: 12 Nov 07
Posts: 696
Credit: 27,266,655
RAC: 0
Level
Val
Scientific publications
watwat
Message 27550 - Posted: 4 Dec 2012 | 19:53:44 UTC - in response to Message 24051.


Any progress on the AMD side of things?


AMD just put out a new revision of their OpenCL SDK, so we'll be taking another look in the new year. However, based on past experience, I don't expect to see the performance and stability necessary to justify the additional effort.

Matt

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27557 - Posted: 4 Dec 2012 | 23:21:23 UTC - in response to Message 27550.

At POEM the OpenCL performance of HD7000 GPUs with GCN increased by ~30% with the Cat 12.11 beta drivers. POEM is probably limited by task switching and data transfers.. but there's been some progress :)

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1112
Credit: 6,162,416,256
RAC: 0
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27573 - Posted: 5 Dec 2012 | 19:06:29 UTC - in response to Message 27557.
Last modified: 5 Dec 2012 | 19:07:21 UTC

At POEM the OpenCL performance of HD7000 GPUs with GCN increased by ~30% with the Cat 12.11 beta drivers. POEM is probably limited by task switching and data transfers.. but there's been some progress :)

Was that 30% going from 12.8 > 12.11 or 12.10 > 12.11? There was a large increase in performance with the 12.9 beta drivers and consolidated in the 12.10 release, and now another performance increase with the 12.11 beta. The AMD driver guys are busy. The increases were very significant for 5xxx and 6xxx GPUs too.

It would be interesting to have a 12.12 legacy release (much like they did with the 12.6 legacy drivers) to support some of the earlier cards. The 40nm 4770 is still a relatively efficient cruncher on some projects considering it's low power draw.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27575 - Posted: 5 Dec 2012 | 19:45:34 UTC - in response to Message 27573.

From what I heard (don't have a GCN card yet) it's ~30% from 12.8 (or earlier) to 12.11 beta whatever. Personally I've seen a 10% (accumulated) increase for an HD6970 - which is still significant at POEM.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1112
Credit: 6,162,416,256
RAC: 0
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27581 - Posted: 6 Dec 2012 | 4:34:34 UTC - in response to Message 27575.

From what I heard (don't have a GCN card yet) it's ~30% from 12.8 (or earlier) to 12.11 beta whatever. Personally I've seen a 10% (accumulated) increase for an HD6970 - which is still significant at POEM. MrS

My 5850s and 5870s are all showing improvement in the 30% range. A little less with my lone 5770.

Werkstatt
Send message
Joined: 23 May 09
Posts: 121
Credit: 321,525,386
RAC: 177,358
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27589 - Posted: 6 Dec 2012 | 16:13:03 UTC - in response to Message 27508.

I've got two things on my mind:

1. Why does anyone think, that a cheaper product (HW&SW) with a shorter development time period can do the same work as a much more mature (and therefore more expensive) one?
2. Why do some volunteers think that the researchers volunteered to please them?


@1: according to performance comparison chart @einstein
http://www.dskag.at/images/Research/EinsteinGPUperformancelist.pdf
the HD7970 is in the same performance range as the GTX690. If you include power consumption to your personnel comparison it might change your mind.
@2: The question amd-app yes or no is a very old question here. The last post from a dev I can remember was: atm the amd openCL has very poor performance but stay tuned and keep the checkbox for beta apps checked. So why should I not come in from time to time and ask about the progress of this question?

As MrS posted earlier
You're of course welcome to report on OpenCL progress and ask about the current state..
I will do that in the future as I made it some days ago, and be shure, I don't get mad at the developers if no progress is there.

Alexander

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27596 - Posted: 6 Dec 2012 | 19:53:08 UTC - in response to Message 27589.

1: The hardware is powerful, I'm sure we can all agree on this. But the VLIW5 and VLIW4 cards have also been pretty powerful.. depending on which code you fed them.

MrS
____________
Scanning for our furry friends since Jan 2002

Werkstatt
Send message
Joined: 23 May 09
Posts: 121
Credit: 321,525,386
RAC: 177,358
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27597 - Posted: 6 Dec 2012 | 21:52:11 UTC
Last modified: 6 Dec 2012 | 21:59:19 UTC

Well, maybe nVidia fits the needs of this project better than AMD does; situation at other projects may be different. For shure it is changing from time to time.[edit] or better: from generation to generation.
The current situation is only a snapshot of a long way to the 'supercomputing tablet'. The next step can possibly found here:
http://www.kickstarter.com/projects/adapteva/parallella-a-supercomputer-for-everyone.
Maybe in two years many of us will crunch with a farm of small gpu-addon-cards.
I found the link in the Einstein forum.

Alexander

Old man
Send message
Joined: 24 Jan 09
Posts: 42
Credit: 16,676,387
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 33445 - Posted: 10 Oct 2013 | 12:40:31 UTC

Any news? Or do i need continue in other projects?

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2705
Credit: 1,311,122,549
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 33446 - Posted: 10 Oct 2013 | 19:02:28 UTC - in response to Message 33445.

No news on the AMD front, as far as I know.

MrS
____________
Scanning for our furry friends since Jan 2002

Post to thread

Message boards : Graphics cards (GPUs) : AMD HD7970 announced

//