Advanced search

Message boards : Graphics cards (GPUs) : 9800GT v2

Author Message
MarkJ
Volunteer moderator
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 738
Credit: 200,909,904
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 6967 - Posted: 24 Feb 2009 | 13:29:00 UTC

I got myself another Palit 9800GT card today. Interesting that it looks totally different to the card I got about 3 months ago (also a Palit 9800GT). I was told that its a 2nd generation one.

It has a somewhat larger heatsink/fan combo and according to vtune is running about 8 degrees warmer than the previous model. I don't expect it will be any quicker though.

I'll post some pics on my blog in the next couple of days, so people can compare them. I might also have some wu completed to compare times against.
____________
BOINC blog

MarkJ
Volunteer moderator
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 738
Credit: 200,909,904
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 6980 - Posted: 25 Feb 2009 | 13:21:33 UTC

1st wu returned from the newer card and it looks like it might be a bit faster. Comparing a 2478 credit wu between both of my gpu-capable machines below. Both have same version of drivers (180.48) and are supposedly running at stock speeds.

Luke (new 9800GT)
Elapsed time = 42,236 secs
Time step = 84.472 ms

Qui-Gon (old 9800GT)
Elapsed time = 48,426 secs
Time step = 96.853 ms

That would look like approximately a 12% speed increase. Pics are on my blog if anyone wants to take a look. I have updated the driver on new machine tonight to 182.06, (after this wu was done) so we'll see if that makes any difference. More samples needed to get a clearer picture.
____________
BOINC blog

Scott Brown
Send message
Joined: 21 Oct 08
Posts: 144
Credit: 2,973,555
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwat
Message 6983 - Posted: 25 Feb 2009 | 15:28:10 UTC - in response to Message 6980.

Try running GPU-Z on the machine to identify differences in the cards. I suspect that the newer card my be on the smaller 55nm die and the older card on a 65nm (especially since the shader clocks are identical).



MarkJ
Volunteer moderator
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 738
Credit: 200,909,904
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 7003 - Posted: 26 Feb 2009 | 13:42:26 UTC - in response to Message 6983.

Try running GPU-Z on the machine to identify differences in the cards. I suspect that the newer card my be on the smaller 55nm die and the older card on a 65nm (especially since the shader clocks are identical).


GPU-Z screen shots on my blog. According to it they are both 65nm cards. The only difference I can see is the BIOS version.

I have updated the drivers for the old card to the latest (182.06) tonight. Once the old card has processed a few wu I might be able to compare and see if its the driver that makes it faster.
____________
BOINC blog

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 7049 - Posted: 28 Feb 2009 | 13:48:43 UTC - in response to Message 7003.

Looks like cost-optimized power ciruitry. If you take enough WU samples you should get similar performance.. they're the same chips, after all. I wouldn't expect a revolution in bios programming with a minor version update and after such a *long* time.

MrS
____________
Scanning for our furry friends since Jan 2002

Post to thread

Message boards : Graphics cards (GPUs) : 9800GT v2

//