• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

8800GTS 512MB Stock Cooler Ballistics Report

Soldato
Joined
11 Sep 2003
Posts
14,716
Location
London
8800gts512hsfbigwayneph7.jpg


Going back a month or two I was in despair to see that my shiny new 8800GT was hitting 95°C when really pushed hard. Maybe it was within specification for it to run that hot but being an 'overcooler' I was none to happy.

Anyway nVidia's new 8800GTS 512MB card is here and it has a big strange looking dual-slot cooler, hmm wonder if its any good? Only one way to find out. . . "Say Hello to my little friend!"



Ok if you don't know what this program above is, lets just say it is the Prime95 of the GPU world, when run for any length of time it will stress your GPU out big time, nothing else I've used can heat a graphics card up as much as this little 'furry' monster.

The Fur Test is being used as a worst case scenario and its unlikely the card will ever get as much stress from normal gaming but for the purposes of testing how well a cards cooling is I think its first rate. I set it to run for 10mins in the tests below.

I have tried to recreate an average setting for this card, its in a closed case which features 2x 80mm fans (one intake, one exhaust) so there is basic airflow but not brilliant. The room the computer is in is warm enough to wear tracksuit bottoms and a T-Shirt.

The 8800GTS 512MB is running at stock nVidia speeds . . .

gpuz8800gts512mbyw3.jpg


The results were better than I was expecting and as such I reduced the number of tests to three

36%-38% *Auto* Fan


50% *Fixed* Fan


60% *Fixed* Fan


tempsallnn4.jpg


Verdict: Yes I'm happy, looks like nVidia have done a real nice job in designing this cooler but once again there is scope to improve things a little.

The fan is pretty quiet, I can't hear it at all at stock or 50% and only just at 60%. As you can see from the results above the difference between [Auto] Fan and [Fixed 50%] fan is quite large, 15°C lower in fact and in virtual silence, nice!

Cranking up the fan to [Fixed 60%] further reduces the temps but introduces a more noticable whoosh, nothing too bad but if gaming using speakers you will notice it a little [note: not all gamers are fraggers, it may be a quiet/atmosphere game lol!).

Final Thoughts: No problems here, move along pls. Can't see anybody having overheating problems with this one, even using the stock Fan-Duty-Cycle but I personally have used RivaTuner to run my fan at 50% when the card gets over 50°C and then drop the speed back down to 36% when below 50°C.

I can live with 66°C full furry beast load, its unlikely to ever get that hot in normal gaming and 50% Fan is nice and quiet.

In the near future I will look at the stock cooling again on an overclocked card and perhaps SLI . . .that's all for now and thanks for reading! :cool:

[edit] having problems getting RivaTuner to detect your 8800GTS? the fix is listed below
 
Last edited:

C64

C64

Soldato
Joined
16 Mar 2007
Posts
12,884
Location
London
run 3d ozone for ten minutes or more with 50% fan setting i would like to see waste of time running it 1 minute.
 
Soldato
Joined
17 Dec 2006
Posts
8,197
run 3d ozone for ten minutes or more with 50% fan setting i would like to see waste of time running it 1 minute.
The tests are for 10 minutes aren't they? Or at least it mentions he ran it for 10 minutes in the tests.

Nice info on the cooler btw OP, thanks for doing it :)
 
Soldato
OP
Joined
11 Sep 2003
Posts
14,716
Location
London
Owners of the new 8800GTS 512MB will need to modify the current version of RivaTuner to work correctly, it just needs five characters (,600h) added to the config

,600h

Code:
[GPU_10DE]

G92 = 610h-61Fh


Code:
[GPU_10DE]

G92 = 610h-61Fh[COLOR="Yellow"],600h[/COLOR]
 
Soldato
Joined
30 Jul 2005
Posts
19,516
Location
Midlands
tried this fur benchmark. but the spinning cube in ati tool, or the spinning spaceship in tray tools heads the card up more and does proper artifact testing. no need to use that fur benchmark.
 
Back
Top Bottom