HardOCP jump the gun?
- Thread starter SsP45
- Start date
digitalwanderer said:
yes and the resolutions are either the same with the nv40 having higher filtering levels or the nv40 is operating @ a higher resolution...
I wonder if the full/final release from [H] will contain direct compares (ie same res/filtering) for the frame-rate...
its highly unfortunate for [H] that someone has leaked this online... lost hits for him... gained for someone else...
but good info none-the-less...
not quite as fast as the rumours had placed it at but by far ahead of the game v/s today's cards... and nice AA too
It might actually help them get hits by building publicity for the final review, ya never know.
Yeah, but Brent has been trying to find a better way to show how the cards will benefit the gamer for a long while now...what better time to show off his new approach than the first Graphic Card Super Bowl? :|AlphaWolf said:
yah... I'm not quite sure what the philosophy for comparing cards by keeping the fps as close as possible was...
i am sure there was a method to the madness but its just beyond me...
however I guess it is unfair to judge w/o seeing the final review and benches that are published...
i am sure there was a method to the madness but its just beyond me...
however I guess it is unfair to judge w/o seeing the final review and benches that are published...
I don't get it either and almost all my initial reactions aren't all that hot, but we are dealing with only a small and distorted look at what Brent is doing and I'd rather wait until he gets a chance to present it the way he wants to before accusing Brent of anything.
But again I say, tomorrow morning is going to be an absolute blast for me! 8)
But again I say, tomorrow morning is going to be an absolute blast for me! 8)
Rev. Night
Newcomer
while it is nice to see that at identical fps, what is each card is capable of doing (at 60 fps, card A will have this AA/AFF, Card B will have this, etc..). but reall tho, that is just secondary. I want to see as primary how each card does at the exact same res, aa/AF setting. All things being equal, i want to see how each card performs.
And enough with this squiggly shit. The min,max, average is good, but those worms are killing me.
And enough with this squiggly shit. The min,max, average is good, but those worms are killing me.
It does make sense, hards reviews show the real benefit you actually would get from changing cards. The settings are chosen purely based on what is playable on each respective card. For example you can deduce that you will be able to set the aa higher, or get hiugher resolution and more aa if you change card.
The other method of benchmarking where cards can get ridiculously high results or all get unplayable low results can give meaningless results. i.e. fx 5700 ultra is faster than a 9600XT at 1600x1200... but both cards were unplayable at that resolution so the point was mute. Or you get quake 3 benchmarks where all cards are over 200fps and so are all effectively equal.
The other method of benchmarking where cards can get ridiculously high results or all get unplayable low results can give meaningless results. i.e. fx 5700 ultra is faster than a 9600XT at 1600x1200... but both cards were unplayable at that resolution so the point was mute. Or you get quake 3 benchmarks where all cards are over 200fps and so are all effectively equal.
MrGaribaldi
Regular
dan2097 said:
Yes, and that is interesting. However, what I think most people here would like is to see the regular way to review too since it's a new architecture.
As it is now, we don't see how much each card drops when you enable/disable FSAA and how it drops with diff resolutions.
dan2097 said:
When just interesting in current games, those results are quite useless as you say, but when trying to figure out how it performs in general, what strengths and weaknesses it is much better as here we'd get to see how it performs over a spectre of settings, and not just one setting in each game.
Äs it stands now, it's quite hard to see what is keeping the GF6800Ultra @ 9800XT levels when it comes to games like Halo and FarCry. Is it the shader performance? And if so, is it running 3.0, 2.0 or 1.1?
Thus I think most people here would like to see "regular" review first to understand the card, and then see a review like Brent has done later.
(But I don't really see any reson to complain as much as they do about it. The review in istelf isn't bad (judging from pics), but targetted at another audience than B3D)