[Mini-Review] BFG 9800GTX 512MB

[Lee13168]

Just got my 8800gts today and i got 38fps with all high 1650x1050 on xp. Thats with 3.4 c2d and 750 core / 1025 memory.

[mascaras]

9800GTX @ 800/1200 in windows XP 1680x1050 High settings = ~43 Fps


regards

[Chosen.]

this is the same speed as a GTS 512 exactly

Not sure that's the case. I have a GTS-512 and I ran my 3dM06 at the closest settings possible:
Both use a Wolfdale at 4GHz, Vista 64-bit
My GTS-512 was clocked at the exact same frequencies (800/2000/2200), even though I ran my benchmark 2 days before mascaras' review:
Mine: http://www.hwbot.org/result.do?resultId=717331

Mascaras:


Only difference I can see, is that he is using the new beta drivers, while I used the 169.25 for my GTS-512.

Look at the difference at the individual scores:
Cards________8800GTS__9800GTX
3DMark Score__15921____15182
SM 2.0 Score___7299_____7118 (!)
SM 3.0 Score___7286_____6578 (!!!)
CPU Score_____3706_____3700 (normal)

Makes me wonder...

[bito]

Maybe he has slacker timings on his system ram.....

4Gb Gskill PQ 8000

[jas420221]

Mascaras... can you please check this out...

http://www.xtremesystems.org/forums/...d.php?t=182833

[Family man 72]

Not sure that's the case. I have a GTS-512 and I ran my 3dM06 at the closest settings possible:
Both use a Wolfdale at 4GHz, Vista 64-bit
My GTS-512 was clocked at the exact same frequencies (800/2000/2200), even though I ran my benchmark 2 days before mascaras' review:
Mine: http://www.hwbot.org/result.do?resultId=717331

Mascaras:


Only difference I can see, is that he is using the new beta drivers, while I used the 169.25 for my GTS-512.

Look at the difference at the individual scores:
Cards________8800GTS__9800GTX
3DMark Score__15921____15182
SM 2.0 Score___7299_____7118 (!)
SM 3.0 Score___7286_____6578 (!!!)
CPU Score_____3706_____3700 (normal)

Makes me wonder...

Maybe it's because his PCI-E is 8x

[Rattle]

i used beta driver 174.70 when i ran it.... wolfy at 4150ish but my gts 512's ram was 1100, no os tweaks whatsoever I even left sidebar going...

[Linchpin]

I think it all depends on how clean your OS install is when you bench.

[Vipeax]

Here are my results with my Asus EN9800GTX





1-4-2008 15:20:52 - XP
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=8x, Vsync=Disabled, 32 bit test, FullScreen
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: 81.93s, Average FPS: 24.41
Min FPS: 16.06 at frame 1956, Max FPS: 33.80 at frame 1004
Average Tri/Sec: -22971368, Tri/Frame: -941023
Recorded/Played Tris ratio: -0.97
!TimeDemo Run 1 Finished.
Play Time: 79.79s, Average FPS: 25.06
Min FPS: 16.06 at frame 1956, Max FPS: 34.10 at frame 994
Average Tri/Sec: -23263578, Tri/Frame: -928130
Recorded/Played Tris ratio: -0.99
!TimeDemo Run 2 Finished.
Play Time: 79.83s, Average FPS: 25.05
Min FPS: 16.06 at frame 1956, Max FPS: 34.10 at frame 994
Average Tri/Sec: -23243670, Tri/Frame: -927786
Recorded/Played Tris ratio: -0.99
TimeDemo Play Ended, (3 Runs Performed)
================================================== ============

Completed All Tests





3DMark:
http://service.futuremark.com/compare?3dm06=6056939
(Motherboard makes it impossible to clock the quad above 3.9GHz, increasing the clocks from 3.81 to 3.835 gove me 3 fps more at the beginning of the 1st benchmark in 3dmark (in the spaceship), so the card is being hold back with my quad on 3.8+)



PS:
The "Default Clock" in GPU-Z are higher, because of my bios mod .

[railmeat]

very solid review,thank you very much for the info.guess im grabbing another 8800gtx for the amazing prices there going for now....


so the bottom line is the 8800gtx\ultra will flat out beat the 9800gtx in higher res easily cause of the 384\768 at thee same clocks...what was nvidia thinking realeasing a "new" 9800gtx with lesser physical specs 256\512?


i agree %1,000 with this guy,facts are facts.i quoted him.
post #33
http://www.ocforums.com/showthread.php?p=5559166

"but i saw this this comment coming a mile away

"The GeForce 9800 GTX couldn't pull ahead of the GeForce 8800 GTX at 1920x1200 resolution with 4x AA enabled. The GeForce 8800 GTX was found to be 16% faster at 1920x1200!"

that was said for call of duty 4. i bet this is true for MANY other games when you add Anti Aliasing to the picture

ill stick with my 8800gtx for now"

post#34
"I will wait for upgrade till I see results from 9900gtx"

agree %100 -
8800gtx\ultra for the win still in any game at higher res,i run 2048x1536 in games so the 9800gtx cant touch my aging 8800gtx.not 1 single reson to downgrade to it(9800gtx).i was so excited to get this card too.i knew something was fishy when i read those specs and seen it was less then a 8800gtx and just came out?wtf???nvidia?

and i must admit the funniest thing i keep seeing everyere on the web is the "8800gt 512 vs 9800gtx 512 threads" STOP right there...it should be the 9800gtx vs 8800gtx threads,i mean they are the 2 king of cards for newest generations.who cares about the little price difference.ppl want bottom line performance at hi res in the newest games.the 8800gtx\ultras are better imo bar none.i really wish ppl would stop comparing the 8800gt 512 to it and start looking at the $300 8800gtx,s when comparing the 9800gtx to other cards.the extra $75 is worth every single penny for the 384\768 vs 256\512.

[boqonx]

So disappointing that I have to put my hope down, this is none other than a recycled GTS..this is really disturbing the market..