Discussion in 'Graphics Cards' started by EastCoasthandle, Apr 12, 2010.
This video benchmark shows a run through in Crysis.
Very close without AA, but with it ATI wins.
Giving Fermi the credit it deserves, it wins hands down in Extreme Tessellation Unigine Heaven:
yup, but its the same as a 4x4 beating an equally powerful saloon on a rally track in the end fermi is designed for tessellation, whereas in normal tessellation both cards are about equal.
Very Close indeed only a few frames in it, one who wins is the one who costs the least in my eyes. Price performance that's where it is.
wow i didn't expect that, I thought for sure the GTX480 was gonna smoke a HD5870 in something like crysis, but it lost, guess not, I guess it only shines in tesselation
yup, nvidia were polishing a turd with those heaven benchmarks.
depends if you like heaven or crysis more i guess
Take a look at the detail on the mountain from 1:18 to 1:21.
I recall they were called out on this before :shadedshu
I agree and thought the same, however Crysis is just one game, you have to look across the board for more accurate results (not suggesting that across the board would favour either one in particular).
and lol, that's just terribad. So they get less performance and skimp on the detail?
Well, take a look at the frame rates during that time the 480 is at least 1 FPS higher during that scene. However, the mountain looks like a mud slide.
Yeah I saw this earlier on today! quiet interesting!
it's obviously worth it for that 1 fps
Makes you wonder how the IQ looks in other games...
Very irratic to be honest, it's strange, I was looking at a few of the reviews that had screenie's last week, some were fine and even pretty good but 2 or 3 were aweful, it was almost like there was a mist in front of the screen, kind of unclear and hazed but not actually blurry if you get my meaning of course I spose it could have just been the quality of the image, I'll see if I can find them.
nice game, that heaven's, multi later?
You can't use a compressed to hell and back youtube video to compare that kind of stuff. I'm not saying it's true or false, just saying the youtube vid cannot be used for quality comparisons.
But they were almost certainly compressed the same on the original source, and youtube would not compress and distort a mountain shaped area on the screen The difference is very noticable, whereas the rest (trees ect.) are the same. It's not compression issues, I didn't even have to put it to 720p to notice. The mountain is much worse on the gtx480. It's undeniable.
Damn if that Tessellation benchmark is accurate us 5800 series guys are in alot of trouble with DX11 games....
10.3a Preview was a FRAUD
Well, we will be in 3 years but so will the gtx480 because it'll be trying to render and tessellate using the same CUDA cores, whilst the hd5870 is only limited by the tessellation units power. The only game to use tessellation so far was dirt 2 (EDIT: and stalker), check out the difference between the two cards in that game in dx11. It isn't a lot at all.
In the end, the hd5870 will happily render all dx11 games as long as the gtx480 will, and when heavy tessellation hits, just set the tessellation to a lower setting ^^ or buy a hd7 series/gtx9 series.
Also, why was 10.3a a fraud? lol, it boosted performance quite a bit.
It's actually fully deniable. We don't know the compression methods they used for both vids, then they had to paste them both together and recompress into a single video. It is not proof of anything at all.
You need to post quality comparisons in a lossless for them to be valid, plain and simple. There is no other way to do it. Once you introduce compression, the comparison is 100% completely useless.
Now, again, it could very well still be true(and most likely is to an extent), but this video is not the proof.
As far as 10.3a boosting performance quite a bit, yeah, if you consider less than 2% in most everything quite a bit.
Yeah, it's pretty obvious.
It could be false, but it is suspect that the rest looks identical. It doesn't seem to be compression, it can't change things THAT much while keeping everything else the same. Still, maybe they purposefully colored it in differently to make the gtx480 look bad. Either way, it's not great for the gtx480, and even worse for the consumer if it's true. Maybe we can get someone with both cards to verify it for us ^^
Drivers generally don't boost every game, but for the amount of boosts it did give, it was pretty substantial for just a single driver, no? definitely the biggest changes i've seen in a while.
Fermi was designed for Tessellation, so i'm not surprised when i see fermi twice faster than HD 5870
but when gaming, ATI still a bit faster than Fermi, but fermi cost more 100 bucks than ATI card. also fermi release 6 months after HD 5870
Separate names with a comma.