• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

WOZ's OC of G92 8800 GTS

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

woz101

Member
Joined
Oct 6, 2007
Location
Los Angeles
Hello, all. I am NOT really an OCer. Just want to have the biggest bang for the buck.:)

So I have just received my stepped-up G92 512MB 8800GTS(stepped up from the old 640MB G80 8800GTS), and I OCed it to this:

6q9kxs9.jpg

The load temp should be 55C rather than 51C. I waited a few seconds, after the ATI Tool scan for artifacts test stopped, so the temp dropped a few C.

Ambient temp about 18C and fan speed set at 70%. Idle temp 38C.

So far, so good.

Will report back in the next post to see how it runs Crysis.

Last but not least, I stole the OC figures/profile from Omniskillz's STABLE 24/7 gaming OC profile!:)

Hope it is OK!:)
 
Last edited:
Well, the initial OC was NOT stable in Crysis. Everything was set at Medium. The game kept crashing at those settings. So I have now backed to 770 core/1923 Shader/1056 Memory.

It is NOT the highest OC on this card, just a random decision I have made.

Surprisingly, Crysis did NOT run well on this card at the above OC. In some places outdoors, the fps was about 55 to 60. However, in some other places outdoors, the fps even dipped to 31 or 32!

Indoors, the fps was around 60 to 65 or so.

I used FRAPS.

With my old 640MB 8800GTS(OCed to 656 core/1519 Shader/1020 Memory), the fps at outdoors was about 40 to 45, while indoors the fps was about 55 or so.

I used the same driver--169.25--for both cards, and everything was set at medium.

I tried the high setting(everything) with the new GTS OCed, but I only got about 19 or 20 fps outdoors. Did not even try indoors, and I immediately switched back to medium setting.

Temperature-wise, the new GTS is much cooler than the old one. See my first post.
 
Last edited:
I can see what a popular guy I am!

Nobody is replying to my posts!:)

Anyway, one thing I forgot to mention is that the new GTS is about several mm's longer than the old G80 GTS. So on my Asus P5K E WiFi, the first two SATA ports are now blocked!:bang head

I am using the rest of the SATA ports. The OS HD is hooked onto the third SATA port. Vista reports that the disk speed is 5.6. When it was connected to Port 1, Vista reported the disk speed as 5.7. Will benchmark with HDtune or HDtach, when I get a chance!

I'd like some feedback!:)
 
I can see what a popular guy I am!

Nobody is replying to my posts!:)

I can assure you that I'm watching all the Q6600 / GTS 512 OC posts with interest considering my own thread in this section... :beer:

But I'm afraid I won't have anything in the way of experience to add until later in the week.
 
I can assure you that I'm watching all the Q6600 / GTS 512 OC posts with interest considering my own thread in this section... :beer:

But I'm afraid I won't have anything in the way of experience to add until later in the week.

Thanks for the feedback!:)

That's one!
 
Do some 3DMARK06 benchmarks at your current OC, at standard resolution and also 1680 x 1050
As far as Crysis is concerned there is a Crysis benchmark tool that you should use for bencing performance in the world of Crysis

once you get those on this thread you'll see some comments coming :)
 
Do some 3DMARK06 benchmarks at your current OC, at standard resolution and also 1680 x 1050
As far as Crysis is concerned there is a Crysis benchmark tool that you should use for bencing performance in the world of Crysis

once you get those on this thread you'll see some comments coming :)

Thanks for the suggestions!

Downloading.

The Crysis Benchmark Tool you linked is some kind of intranet thing. IE7 cannot open it, even after I have enabled intranet.

Well, anyways, not a big deal. I googled, and will be downloading it from Guru3d.

Thanks again!
 
Crysis Benchies

OC--770 core/1923 shader/1056 memory

OC Medium
12/18/2007 10:49:14 PM - Vista 64
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX10 1680x1050, AA=No AA, Vsync=Disabled, 64 bit test, Windowed
Demo Loops=3, Time Of Day= 9
Global Game Quality: Medium
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 39.52s, Average FPS: 50.60
Min FPS: 37.84 at frame 1942, Max FPS: 65.62 at frame 1743
Average Tri/Sec: 43922128, Tri/Frame: 867975
Recorded/Played Tris ratio: 1.06
!TimeDemo Run 1 Finished.
Play Time: 36.47s, Average FPS: 54.85
Min FPS: 37.21 at frame 1970, Max FPS: 65.62 at frame 1743
Average Tri/Sec: 48058328, Tri/Frame: 876254
Recorded/Played Tris ratio: 1.05
!TimeDemo Run 2 Finished.
Play Time: 36.30s, Average FPS: 55.09
Min FPS: 36.62 at frame 1967, Max FPS: 65.62 at frame 1743
Average Tri/Sec: 48296356, Tri/Frame: 876631
Recorded/Played Tris ratio: 1.05
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

12/18/2007 10:49:14 PM - Vista 64

Run #1- DX10 1680x1050 AA=No AA, 64 bit test, Quality: Medium ~~ Overall Average FPS: 54.97

OC High
12/18/2007 10:56:10 PM - Vista 64
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX10 1680x1050, AA=No AA, Vsync=Disabled, 64 bit test, Windowed
Demo Loops=3, Time Of Day= 9
Global Game Quality: High
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 72.13s, Average FPS: 27.73
Min FPS: 20.94 at frame 1943, Max FPS: 33.83 at frame 1753
Average Tri/Sec: -27117536, Tri/Frame: -977945
Recorded/Played Tris ratio: -0.94
!TimeDemo Run 1 Finished.
Play Time: 67.61s, Average FPS: 29.58
Min FPS: 20.94 at frame 1943, Max FPS: 33.83 at frame 1753
Average Tri/Sec: -28578132, Tri/Frame: -966100
Recorded/Played Tris ratio: -0.95
!TimeDemo Run 2 Finished.
Play Time: 68.06s, Average FPS: 29.39
Min FPS: 20.88 at frame 1950, Max FPS: 33.98 at frame 1757
Average Tri/Sec: -28384000, Tri/Frame: -965923
Recorded/Played Tris ratio: -0.95
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

12/18/2007 10:56:10 PM - Vista 64

Run #1- DX10 1680x1050 AA=No AA, 64 bit test, Quality: High ~~ Overall Average FPS: 29.485


Stock Medium
12/18/2007 11:02:57 PM - Vista 64
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX10 1680x1050, AA=No AA, Vsync=Disabled, 64 bit test, Windowed
Demo Loops=3, Time Of Day= 9
Global Game Quality: Medium
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 41.94s, Average FPS: 47.69
Min FPS: 37.19 at frame 1950, Max FPS: 56.99 at frame 1654
Average Tri/Sec: 41413684, Tri/Frame: 868449
Recorded/Played Tris ratio: 1.06
!TimeDemo Run 1 Finished.
Play Time: 39.92s, Average FPS: 50.10
Min FPS: 36.43 at frame 1962, Max FPS: 58.24 at frame 1729
Average Tri/Sec: 43922440, Tri/Frame: 876756
Recorded/Played Tris ratio: 1.05
!TimeDemo Run 2 Finished.
Play Time: 39.98s, Average FPS: 50.03
Min FPS: 36.43 at frame 1962, Max FPS: 59.32 at frame 1746
Average Tri/Sec: 43841584, Tri/Frame: 876393
Recorded/Played Tris ratio: 1.05
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

12/18/2007 11:02:57 PM - Vista 64

Run #1- DX10 1680x1050 AA=No AA, 64 bit test, Quality: Medium ~~ Overall Average FPS: 50.065

Did not try Stock High, because I was afraid it might crash my system.
 
Thats a damn fine 3DMark06 score, especially for vista EDIT: Didnt notice you had a quad-core. Still a nice score

So is Crysis playable on high with a decent OC?
 
Last edited:
Thats a damn fine 3DMark06 score, especially for vista EDIT: Didnt notice you had a quad-core. Still a nice score

So is Crysis playable on high with a decent OC?

No. I tried the high setting in an outdoor scene; the fps was like 19 to 20. So I immediately switched back to medium! LOL.

I guess I will have to wait till the next gen cards in order to play Crysis at High or Very High.

Not that I am bothered by it.
 
I would have expected to see a higher ave Crysis FPS on 1680 x 1050 at OC with your GTS, see my data on the GT
http://www.ocforums.com/showpost.php?p=5394251&postcount=17

Maybe the driver for the new GTS hasn't been optimized as yet?

Don't know why.

Will see if the performance improves with the future driver releases.

Thanks for the feedback.

EDIT: that 33.8 fps in your report at the res of 1680x1050 was at the Medium setting or High setting?
 
Good to hear about your new card, however, never just copy OCs from others. Make sure that you do all your own testing, as different cards in different setups perform differently. I just finished testing my XFX 8800 GTS 512 which took two full days. I usually run ATI tool for 1 hour straight (good for memory testing), then Oblivion for 1 hour (I just leave it on the 'create a character' screen which I find really heats up the card and stresses the shaders). Then I run 3dmark 05 on a loop for over 1 hour, after which I run 3dmark 06 for over 1 hour on a loop. Moreover, I do all these tests back to back as to keep my card hot and stressed. Once I find the max. stable frequencies after all these tests, I begin to run all my games. So far I am standing @ 778/1900/1080MHz core/shader/memory with the fan @ 56%.

P.S. For just one pass on 3dmark 05/06, I was able to run 800/2000/2200MHz
 
Last edited:
Good to hear about your new card, however, never just copy OCs from others. Make sure that you do all your own testing, as different cards in different setups perform differently. I just finished testing my XFX 8800 GTS 512 which took two full days. I usually run ATI tool for 1 hour straight (good for memory testing), then Oblivion for 1 hour (I just leave it on the 'create a character' screen which I find really heats up the card and stresses the shaders). Then I run 3dmark 05 on a loop for over 1 hour, after which I run 3dmark 06 for over 1 hour on a loop. Moreover, I do all these tests back to back as to keep my card hot and stressed. Once I find the max. stable frequencies after all these tests, I begin to run all my games. So far I am standing @ 778/1900/1080MHz core/shader/memory with the fan @ 56%.

P.S. For just one pass on 3dmark 05/06, I was able to run 800/2000/2200MHz

Thanks for the feedback!

Others' OC profile, though, still gives you a ROUGH idea of where you can go, no?

As you have said, it will take a while to further tweak the new card.

The bright side is that I stepped up from the old 640MB G80 8800 GTS to the new GTS with only $15!

So I am happy about that!
 
What OS are you using?

I am using Vista 64 bit Ultimate.

See sig.

Also, the score of the 8800GT is ONE single card, not SLI, right?

LOL. Not that I am comparing or anything, I just want to find out an answer to the relatively low fps in Crysis.

The Crysis Benchmarking Tool is dled from here:
http://downloads.guru3d.com/download.php?det=1791

XP-32, that probably explains why you are getting relatively low scores, Vista is a resource monger, hence the reason why I uninstalled vista and only use XP

I'm using only one card for comparing with HD 3870, the other GT is idling in another rig

:)
 
Last edited:
Back