Itchy for this card now, should i buy HTPC or buy this 8800GT + PSU upgrade?

This post has been edited by Bliz: Oct 25 2007, 11:00 PM

just simply put 333X9 only, wont harm your processor.
But, i also can't force you because that is your own desicion

---

Added on October 25, 2007, 11:19 pmlet him be, if he don't want overclock..i dont think the cpu 100% utilize 8800gt.

This post has been edited by 168257061: Oct 25 2007, 11:19 PM

hmm.. despite the hypes goin on bout the 8800GT, im starting to have doubts on it.

Quite some time ago, i posted my 3dm06 scores for my 2600XT crossfired, the 3dmark06 results were simply amazing, a blazing 9k on an AM2 rig. Then again, on real world performance, its not even close to 8800GTS, in fact, its losing out ALOT despite being ahead of 8800GTS on 3dm06. Basically, the problem was the random stuttering and slowdowns during texture load.

By referring to the link on gpureview.com, im suspecting the 8800GT might have the same issue due to slower texture fillrate and lower bandwidth compared to the 8800GTS, just 1 piece of advise to current 8800GTS users, dont jump to conclusion so fast, hold on to your horse and wait for the real thing to appear and see how it REALLY perform.

As i noticed, the shader clock on the 8800GT is much higher than of 88GTS, then again, 88GTS dont have any problems at all running the same speed as 88GT, given the same core/shader/mem speed, the GTS still wins.

Ive also have doubts on the single slot cooling solution there, on stock speed as reported by accs_centre, it runs even hotter than my overclocked 88GTS (650/1625/1050@60c on load).

I have strong doubts on 3dmark06 as i have personally tasted the flaws of it. Then again, the specs on the 88GT seem really tempting... guess we'll just have to wait till its officially out and more results are known.

i changed my 8800gts to 8800gt orders from strongbaby oredy... anyways, doesn this card support normal old pcie 1.0..?

anyone already knows wat temps ur getting on this card ?, in relation to ambient temp.. thx

yes..backward compatibility..higher can go lower..but lower cant go higher..

the socket of PCIE 2.0 same with PCIE 1.1 ?

if like this then i dun have to worry.. coz changed the card then thought bout the compatibility.. too happy oredy..

no problem la..using it now..my board is matx g33 ds2r..

hmm.. mine is abit kn8 ultra wor.. confirm can..? really scared buy but not compatible..

just oc my E4400 @ 2.66GHz, here's some pictures
[attachmentid=317884] [attachmentid=317885]
i really doubt it that this card can really go higher with the appropriate setting

yep i agree 3d mark is not a good way to prove the card strength only running real games comparing then only will know the result....

some of the score are from oc=ed CPU score,bro,try to make it to 3dmark06 and run at 1280*1024 and we'll know the real performance.

y not doing some in game benchmark there?

3dmark06 configs should be left default for more accurate comparison(1280x1024). The benchies that were posted accs_centre is 3dm05 and not 06. 3dm05 default config is @ 1024x768 so its fair enough.

In regards to OC-ed CPU, if all test are performed on the same CPU clockspeed side by side then i think it is fair enough to be used as comparison between the GFX. Of course.. the GFX has to be runned on stock speed.

Then again as i mentioned previously.. 3dm really fails me. 2600XT Xfire gave me 9k+ while 8800GTS on stock speed gave me 8k+, but on real gaming.. do trust me when i say the difference is SIGNIFICANT

Real performance has to be seen with your own eyes and not synthetic benchmarks, given the same framerates it doesnt mean it would be smoother. IMO, i noticed that sometimes 3dmarks doesnt detect spikes or certain lags, even if its spiking the framerates are still above 50-80fps.

My earlier beachmark with 3dmark06 can be found here:

http://forum.lowyat.net/index.php?showtopi...post&p=13789972

well unless it can surpass 20k in 3dmark05
and 13k for 3dmakr06 on c2d
on 3ghz
then this card is not worth it

also take note its ok to compare same gpu of the same type for performance since 8800gt is a evolution of the G80 .

so 3dmark06/3dmark05 would roughly reflect real world perormance on openg/shader games

its stupid to compare two different cards like HD2X00's and nvidia 8X00's on 3dmark05/06 as this will not reflect real world performance as each games are optimized for certain cards and driver enchancements also do play a role. Gone are the days where u can keep one driver and keep gaming on new games.. nowadays new games mean new drivers.

Well after november will be getting either a XFX 8800gt/gts to sli or a HD38x0's to crossfire.. my bet will be the second choice on the Maximus..

This post has been edited by cstkl1: Oct 26 2007, 02:31 AM

i think its becoz inmature driver or perhaps, the games u play not support crossfire m0od, thats y the performance worst than 3dmark, IMO

anyone can get 8800GT heatsink removed picture , i want see inside !hiak hiak ..

here is another review from taiwan review ! click here
because lack of software to oc the new 8800GT (G92) , so decide oc 8800GTS 320 only , same clock speed , clock : 600MHz , memory : 1800MHz

8800GT SLI mode


8800GT◎600/1800


8800GTS@600/1800


GT win GTS 2037


8800GT SLI@600/1800


8800GTS SLI@600/1800


GT win GTS 993

This post has been edited by likito: Oct 26 2007, 03:03 AM

i'm thinking of buying a p35 mobo.
but heard pcie 2.0 is way faster than pcie 1.1 (double if i'm not mistaken).
do you think i should hold my fire first and wait for mainstream pcie 2.0 mobo?