64xQ FSAA with GTX 295 quad SLI Possible

[SNiiPE_DoGG]

well its not really trolling if I called their BS and it turns out 64-sample AA doesn't work at all..... is it?

[generics_user]

seriously, anything more than 8xAA is a waste of time, especially on lcd displays with high pixel density.

nvidia and ati better focus on making better sli and cf drivers instead of some marketing crap like this is (not that cf drivers are bad, but they could be better)

[Natalia]

4x is enough for me.

Yeah, I have a hard enough time seeing past 4x at the high resolutions we play at these days.


64xQ ... who cares. Just make 4x or 8x faster.

[Caparroz]

This is as usefull as a 9400GT with 1GB of frame buffer. And people actually buys 9400GT's with 1 GB of ram...

[cegras]

He voiced the sentiments of almost everyone out there, what's wrong with what Snipe said?

Even 8xCSAA / Q / whatever it's called is not noticeable from 4 on 1440 x 900, and it's low resolutions where AA matters.

[STaRGaZeR]

Aside from powerlines in some games being done better than NVIDIA, there is no perceptible difference. You're only exaggerating reality.

This is the same as "people can't see more than 60FPS", go figure.

[MrHydes]

well tommarow's the big day. I can't believe there have'nt been any leaked photo's of EVGA's or XFX's cards or any other company.

well here they are...

complete PC with GTX 295 consuming only 120 a 130 Watt IDLE 138W instaling vista

Vista 64 with drivers 185.20

Setup:
- I7 920 @ 4.2GHz Air Cooled (True)
- EVGA X58 SLI
- 2 x EVGA GTX295 @ 690-1200


Vantage with PhsyX Disable :




Crisys 1920x1200 VeryHigh






Average FPS = 72 / Max FPS: 112 a 1920x1200 VeryHigh

[Final8ty]

This is the same as "people can't see more than 60FPS", go figure.

By the people who have never had more than 60fps on a screen that can show more than 60fps.

[Cybercat]

This is the same as "people can't see more than 60FPS", go figure.

The difference is that while the framerate people can see can't be summarily quantified scientifically with any sort of consistency or credibility, anti-aliasing can be analyzed pixel-by-pixel within each frame using simple software techniques available to anyone, effectively discarding the fallible human element of perception from the equation.

[STaRGaZeR]

The difference is that while the framerate people can see can't be summarily quantified scientifically with any sort of consistency or credibility, anti-aliasing can be analyzed pixel-by-pixel within each frame using simple software techniques available to anyone, effectively discarding the fallible human element of perception from the equation.

You're forgetting that analyzing any form of video using individual screenshots is useless. Even if you compare them one by one in sequence, the human perception changes with movement and focus. Something a machine can't do. And there you have the real and visible (for humans) difference.

[Cybercat]

You're forgetting that analyzing any form of video using individual screenshots is useless. Even if you compare them one by one in sequence, the human perception changes with movement and focus. Something a machine can't do. And there you have the real and visible (for humans) difference.

It's useless huh?

In that case, perhaps you should start a petition to end the taking of screenshots altogether. Obviously we aren't benefiting from them in any way.

[alexio]

It's useless huh?

In that case, perhaps you should start a petition to end the taking of screenshots altogether. Obviously we aren't benefiting from them in any way.

I'm not sure if you're being sarcastic, but infact he is right. It is the same thing like when you compare different video codecs. Individual images may look sharper with one codec, but it's more about fluid motion. That is also one of the reasons why some forms of exotic AA actually look no better than just plain ol' 4xAA when they are compared in reviews using screenshots.

Anyway, 64xAA is ridiculous

[Kingcarcas]

It's useless huh?

In that case, perhaps you should start a petition to end the taking of screenshots altogether. Obviously we aren't benefiting from them in any way.

We should, everyone will criticize without taking into account their own monitor and how it's calibrated. Not to mention the person who took the screenshots display.

[koei]

I have a feeling this 64/64Q is more for bragging rights than anything useful in both quality and performance.

[STaRGaZeR]

It's useless huh?

In that case, perhaps you should start a petition to end the taking of screenshots altogether. Obviously we aren't benefiting from them in any way.

Yes, it's useless. And that petition is even more so

What you suggest is the best "technical" aproach you can make about something as subjetive as the human vision. But you can't rely on it to decide which one is better.

[Levish]

OMG, can't wait for pics of games running high res and 64xQ AA. That must look good, and with quad sli and a core i7, a lot of titles should be playable with 64xQ, on some native resolution of 14" - 15" desktop monitors.

FTFY

1024x768 4TL

[RejZoR]

Anyway, 64xAA is ridiculous

Most of ppl were saying the same thing when i was running 16xFSAA on GeForce 2 Pro. V-Rally 2 worked pretty well with that (by "pretty well" i mean the framerate was actually still playable) setting and quality of edges was insane back then. Not a single jaggie. Not sure what res i used with it. Probably 800x600 or 1024x768. Unfortunatelly they removed the support for 16xFSAA in later Detonator drivers.

For example with HD4850 i used 8xFSAA Edge Detect and 16xFSAA on Max Payne 2 and framerate was still so high i could use 3 times the sampling rate and anisotropic multi and framerate would probably still be far over 60fps.
So yeah it's cool to use insane FSAA on older games and finally get graphics you could just dream off when the game came out.

[Manicdan]

the only reason for the 64xAA, and i know nvidia knows it, is cause they will sell a few more units, to the people with too much money and no understanding of video cards, but somehow believe it means its better. its a sales tactic, nothing more.

[dan7777]

the only reason for the 64xAA, and i know nvidia knows it, is cause they will sell a few more units, to the people with too much money and no understanding of video cards, but somehow believe it means its better. its a sales tactic, nothing more.

yup your defo right there i will never pay £400 for that card rip off if ya ask me price drop soon enuf lol

[Shintai]

I have a feeling this 64/64Q is more for bragging rights than anything useful in both quality and performance.

Couldnt have said it better. I would almost say that for anything over 8.

[ToTTenTranz]

well here they are...

Hey... what's with the excellent multi-gpu scaling with Crysis?! It didn't use to be like that!

[EniGmA1987]

ATI's Edge-detect AA at its minimum setting of 12x is plenty fine for me. It looks great even on a 90" monitor.

[Xello]

ATI's Edge-detect AA at its minimum setting of 12x is plenty fine for me. It looks great even on a 90" monitor.

You would see a vast improvement of 64x vs 12x on a 90" monitor

Bigger pixels need higher levels of AA for the same image quality of smaller pixels.

[EniGmA1987]

Well I know a difference would be much more easily seen between 12x and 64x on a 90" screen, but the 12x edge-detect still looks great. So why would anyone need to run 64x? It is completely pointless, and as others have said it probably wouldnt even be capable of actually playing a decent res game at those levels.

[Sly Fox]

You can always run Doom with 64X.

At 2560++ res I doubt even quad SLI is going to handle any of the best looking games to run smoothly.