View Full Version : guru3d will have a x800 review tomorrow.
Pandora's Box
06-15-04, 12:53 PM
22 pages so far. will include x800pro and x800xt.
stay tuned to guru3d.com!
dan2097
06-15-04, 01:01 PM
What Nvidia cards have you found for comparison?
Pandora's Box
06-15-04, 01:05 PM
lol im just a user in the know at guru3d hehe. but the admin says he will do a comparison of x800 vs 6800 ultra. and then maybe a comparison with all optimization off (including brilinear)
Blacklash
06-16-04, 02:30 AM
Would someone please do a review with an OCed ultra vs. an OCed X800PRO for the entire review. It would be very interesting to see. The Pro is extremely underclocked.
r2d2d3d4d5
06-16-04, 06:30 AM
Would someone please do a review with an OCed ultra vs. an OCed X800PRO for the entire review. It would be very interesting to see. The Pro is extremely underclocked.
Only if those OCs can be guaranteed on every retail card. Otherwise it would be interesting from a technical point of view but there would not be much relevance to the consumer.
Omega53
06-16-04, 06:46 AM
Only if those OCs can be guaranteed on every retail card. Otherwise it would be interesting from a technical point of view but there would not be much relevance to the consumer.
The lowest OC ive seen anyone get is 500/500. Me personally I can get mine up to 535/535 without a prob but backed it down to 525/525 just to be on the safe side. 95% of all people I know that have bought a x800 pro hit a minimum OC of 525/525
Pandora's Box
06-16-04, 06:52 AM
well the review is up,
hah! x800xt pe cant stand up to 6800Ultra even when overclocked to 600MHZ memory. 560MHZ core.
:lame:
hah! x800xt pe cant stand up to 6800Ultra even when overclocked to 600MHZ memory. 560MHZ core.
:lame:
Cause both cards are using brilinear this time around and being compared on equal ground :)
Wow no Far Cry benchmarks - at least Quake 3 runs at a zallion frames per second.
Cause both cards are using brilinear this time around and being compared on equal ground :)
I thought he used 60.72? Hmm, i must have missed something. Too tired to read every page. :o
r2d2d3d4d5
06-16-04, 08:46 AM
Cause both cards are using brilinear this time around and being compared on equal ground :)
Not sure if they are being compared on equal grounds without picture comparisons though. The two brilinear implementations are not necessarily equivalent.
For some reason the charts are not showing up on Firefox 0.9 for me. Works OK with IE though.
I thought he used 60.72? Hmm, i must have missed something. Too tired to read every page. :o
He said he enabled brilinear on the 6800U - you can enable it on 60.72 if you choose to.
Halo looks like the only GPU limited test. However, that could be because of bad coding.
Pandora's Box
06-16-04, 10:04 AM
Halo looks like the only GPU limited test. However, that could be because of bad coding.
he's aware of this and thats why amd is shipping him a 3800+.
:clap:
DivotMaker
06-16-04, 10:21 AM
well the review is up,
hah! x800xt pe cant stand up to 6800Ultra even when overclocked to 600MHZ memory. 560MHZ core.
:lame:
Huh? I just read the entire review as well as studied the benchmark charts. I see both high end products within 1-2% of each other in most benchmarks. Makes me wonder if we are reading the same review....
Heck, I will quote Hilbert Hagedoorn:
Conclusion
ATI once again managed to strike back sufficiently towards NVIDIA's Series 6 of GPU. Both cards have their advantages and disadvantages. Shader Model 2 is slightly faster on the x800 series, not in extreme amounts though. When we look at overall performance the GeForce 6800 was a tad faster here and there but really... so marginal. ATI still has an advantage performance wise when it has enabled AA and AF.
I see both high end product being VERY competitive with each other. Not sure how you arrive at your conclusion however.....
Pandora's Box
06-16-04, 10:31 AM
we will have to see what the results are with the amd 64 3800.
dan2097
06-16-04, 10:50 AM
That temperature was measured on the outside of the GPU, meaning that core is really hot. The PC case was open btw. If you get an x800 with reference cooling then you better be sure that your PC is cooled sufficiently as all that heat will warm up the ambient temperature of your PC's inside and cascade the effect by warming up other components.
As the 6800 ultra uses more power than the X800 XT wouldnt that put out even more heat into the case? I mean just because a card is hot doesnt mean its putting a lot of heat into a case. If anything wouldnt it mean the opposite i.e. its cooling is insufficent to dissipate enough of the cards heat into the air of the case so the card remains hot, I know thats certainly true of my passively cooled gf2mx. That gets very hot but it defininitly releases far less heat than my r9700 into the case.
The halo results are a bit different from what other sites have shown.
Yet again I wonder why they did not include Far Cry as it is one of the main reasons people wanted or needed to upgrade their video cards in the first place.
EMunEeE
06-16-04, 12:15 PM
Why do they not go above 16XAF in some tests...or above 4XAA??
EMunEeE
06-16-04, 12:22 PM
Why do they not go above 16XAF in some tests...or above 4XAA??
Why do they not go above 16XAF in some tests...or above 4XAA??
Can you go higher than 16xAF?
fleshonbone
06-16-04, 03:00 PM
If this is being done for the consumer/gaming enthusiast - what I would've found helpful in comparison studies such as this is a compatibility test - at least it would've helped me 1000%.
I posted a whole thread on this a couple days ago, and I agree that the X800's IQ is excellent, but that doesn't help much when there are numerous little driver inconsistencies, abnormalities, etc.
I don't honestly care if the X800 gets 267fps in QuakeIII and the 6800U only gets 254fps.
To me, most of the IQ scrutiny between the two card mfg's is almost unrecognizeable - I mean, if you are going to examine stills blown up in photoshop where one card has .00001 more of pixelated aliasing, I'd rather have the card that runs more games without problems.
So, what I would personally like to see with these things is for the testor to run both cards on 50 different games. Then report which card ran the most number of games correctly, which card had more errors, driver problems, graphical problems, etc.
I realize the 6800U does not have cert. drivers yet, so I think these comparisons are a bit premature, BUT, if I were to lay money down, I would say that the IQ for both cards is probably very very close - what I'm slowly finding out (something which I took for granted before), is that IQ doesn't do diddly when you can't run the game, or there are driver problems etc - I mean Splinter Cell looks great on the X800, until you notice that all the shadows for the game have been removed!
I'm not picking on either card - I would just like to see some compatibility/stability tests one day.
fivefeet8
06-17-04, 01:06 PM
Maybe they didn't do FAr Cry Benches because the game doesn't run properly on the Nv40 yet. Things could and probably will change for the Nv40 once the 1.2 patch is released.
Everyone also seems to think that Nvidia's brilinear isn't as good IQ wise as Ati's method of optimization. I haven't seen any current proof one way or the other. I have however, tested the Full Trilinear options with the new forceware 61.34 on my Fx5900u. I can't find any difference at all with Brilinear and Full Trilinear at 4xAF or higher in the games I play. Even using Af testing programs the difference is miniscule at best. Maybe I just haven't run into a game that will show the weekness of the optimiation. Atleast I'll have the option to turn off the Optimization if I ever do run into a situation where Brilinear isn't enough.
Guru3d should have Overclocked the 6800u too. But for the most part, it's a good review.
vBulletin® v3.7.1, Copyright ©2000-2012, Jelsoft Enterprises Ltd.