Bug is a generous descriptor. Why would an alleged world class driver team make such a mistake -assuming this is legit of course ?
That's what its supposed to do. Let app decide. Not override it. On AMD default, you control it in-game. So Ultra = Ultra.
At least [H] confirms they bench with everything on default in CC and NVCP. I agree with that because its what I would have done as well to ensure its "fair". Except with this bug, it's very far from fair.
So the old saying that AMD has better image quality still holds candle to this day:
-In the past AMD had better VGA quality because they used better RAMDAC
-They started using 10bit internal LUT from 5xxx series onwards which gives you better monitor picture quality with calibrated monitors because you get less banding. Nvidia still uses 8bit LUT on their consumer cards.
-In 2014 they upped the internal LUT to 12bit, meaning even less banding with calibrated monitors. I can now calibrate my 10bit Direct Drive HP ZR30w monitor (no internal LUT / scaling board) and not get banding at all.
-And now it seems they even have better texture quality at default settings.
No wonder why all my nvidia cards have looked like shit :hmm:
Did I just say that out loud :awe:
Bug is a generous descriptor. Why would an alleged world class driver team make such a mistake -assuming this is legit of course ?
There is a very obvious reason most reviewers don't do comparative iq tests, they look the same. Not quite sure about your assertion that nv looks terrible.
It's a bug until more info suggests otherwise.
I'm known to have an anti-NV stance here (not unwarranted, it's been the GW thing that led me to boycott NV), but I will consider it a bug that needs fixing. If its widespread and happening for so long, then I will consider it cheating, something that is anti-gamer because you pay for expensive hardware to crank settings up and you get ass "donkey" quality if you left it on driver default.
It would also be anti-competitive since its cheating IQ to get faster performance in benchmarks.
Bollocks. There is no content in gaming, movies or general apps that can make use of 10-bit. Only Adobe Premier Pro/CS5 use 10-bit OpenGL which AMD support but with Nvidia you need a Quadro card. Nvidia Geforce are capable of 10-bit on DirectX full screen surfaces (since 200 series) but again, there is nothing that can make practical use of it either for Nvidia or AMD yet. ArgyllCMS calibration software is 10-bit capable, but again no point if just about everything outside of it is not 10-bit (gaming, programs, general use). So banding can only be an issue if you have Adobe Premier/CS5 to show it on 10-bit monitors.So the old saying that AMD has better image quality still holds candle to this day:
-In the past ATI had better VGA quality because they used better RAMDAC
-They started using 10bit internal LUT from 5xxx series onwards which gives you better monitor picture quality with calibrated monitors because you get less banding. Nvidia still uses 8bit LUT on their consumer cards.
-In 2014 they upped the internal LUT to 12bit, meaning even less banding with calibrated monitors. I can now calibrate my 10bit Direct Drive HP ZR30w monitor (no internal LUT / scaling board) and get no banding at all.
-And now it seems they even have better texture quality at default settings.
No wonder why all my nvidia cards have looked like shit :hmm:
Did I just say that out loud :awe:
Bollocks. There is no content in gaming, movies or general apps that can make use of 10-bit. Only Adobe Premier Pro/CS5 use 10-bit OpenGL which AMD support but with Nvidia you need a Quadro card. Nvidia Geforce are capable of 10-bit on DirectX full screen surfaces (since 200 series) but again, there is nothing that can make practical use of it either for Nvidia or AMD yet. ArgyllCMS calibration software is 10-bit capable, but again no point if just about everything outside of it is not 10-bit (gaming, programs, general use). So banding can only be an issue if you have Adobe Premier/CS5 to show it on 10-bit monitors.
Here's a small sample of the start of my monitor color profile (it's actually a text file!):
0.00000000 0.00000000 0.00000000 0.00000000
0.00392160 0.00115970 0.00012207 0.00000000
0.00784310 0.00402840 0.00308230 0.00296030
0.01176500 0.00695810 0.00608830 0.00610360
0.01568600 0.00993360 0.00915540 0.00929270
0.01960800 0.01294000 0.01228400 0.01252800
0.02352900 0.01600700 0.01545700 0.01583900
0.02745100 0.01912000 0.01867700 0.01919600
0.03137300 0.02226300 0.02197300 0.02262900
0.03529400 0.02548300 0.02531500 0.02612300
This is called a gamma ramp / curve.
Range: 0 --> 1, for the sake of simplicity I will assume in my examples that the range is 0 --> 255 (8-bit per color channel).
The graphics card loads up the matrix in a Look-up Table, where it basically converts the input color signals to the corrected output signals on-the-fly.
Now here's where it gets confusing:
AMD's consumer cards support 10-bits per color channel. Nvidia's cards don't; they only support 8-bits per color channel. Now, how is this relevant when you have an 8-bit per color channel monitor, you might ask?
Look up, at the matrix. Notice how many digits there are after the decimal point. These number contain enough precision that even 16-bits per color channel do not cover them. So, that means 8 or 10 bits are taken, and the rest are dumped (not considered).
Here's how it looks when mapped from 0 --> 255:
0 0 0 0
1 0.295724 0.0311279 0
2 1.02724 0.785987 0.754876
3 1.77432 1.55252 1.55642
4 2.53307 2.33463 2.36964
5 3.2997 3.13242 3.19464
6 4.08179 3.94154 4.03895
7 4.8756 4.76263 4.89498
8 5.67707 5.60311 5.77039
9 6.49816 6.45533 6.66137
Here's what it looks when only 8 bits are taken (after rounding is performed):
0 0 0 0
1 0 0 0
2 1 1 1
3 2 2 2
4 3 2 2
5 3 3 3
6 4 4 4
7 5 5 5
8 6 6 6
9 6 6 7
Duplicate numbers in sequence? Inability to display certain shades? Banding? Whoops.
Here's how it looks like when 10 bits are taken (after rounding is performed):
0 0 0 0
1 0.25 0.00 0.00
2 1.00 0.75 0.75
3 1.75 1.50 1.50
4 2.50 2.50 2.50
5 3.25 3.25 3.25
6 4.00 4.00 4.00
7 5.00 4.75 5.00
8 5.75 5.50 5.75
9 6.50 6.50 6.75
10 7.50 7.50 7.50
Less gaps, smoother curve, less banding. Awesome.
BUT...monitor is still 8-bit...sooo what AMD does is use dithering from the 10-bit LUT to the 8-bit monitor. Banding is gone!
Re the AF and texture filtering, AF has minimal impact on performance, maybe 1 or 2% if that. Texture filtering when going from Quality (default) to High quality there is an impact on Nvidia cards, and that I believe is around 10% or even more. My comment is only based on personal testing in 3dmark and observing resulting scores with such settings enabled. Even that was a few years ago, not sure of current impact, but will test it later on.
Not sure, may be best to ask Florian Hoech, author of DispcalGUI. He is always helpful and can give you a far more informed answer than I can.Yes we all know windows can't do more than 8-bit system wide so its all candy and dandy but how do you explain this:
EDIT: why SPOILER tags do not work. Me not understand :|
EDIT2: Sorry for slight offtopic.
Screenshot provided by GTX 980 user on NCP at default settings. Looks fine on this.
Benchmarks should be run with both vendors set to the highest quality level in the drivers. What one vendor calls default may not be (and probably isn't) the same as another vendor. On the other hand, with everything set to the highest quality there shouldn't be any performance optimizations that noticeably reduce IQ (at least not in this day and age).
Of course, looking at IQ under a magnifying glass is something sites should still do now and then to make sure everyone is playing honest.
NVIDIA forced AF
That picture only furthers the problem
1. The photographer is closer in the picture you posted, the AF cut off is further away but it's still there
2. There are multiple missing objects in the picture you posted. Take a look at the yellow stop lights, the missing lamp poles and that yield sign
This situation is kind of ridiculous.
Brent over at HardOCP is discussing the matter. And his comments leave a lot to be desired about their reviews.
http://hardforum.com/showthread.php?t=1867421&page=5
Starts on post #77
Some choice quotes, which I preface with his first quote:
Then the kicker for me.
What is the point of Apples to Apples comparisons if the picture does not look exactly the same or even as close as they can get? I want to say that this one case from OCUK can be the outlier and I want to give Brent the benefit of the doubt, and his reviews are trustworthy. But this really needs more investigation.
I do agree, lots of reviews are just graphs and words, no real comparative pictures, or diving into impacts of different settings.
NVIDIA forced AF
That picture only furthers the problem
1. The photographer is closer in the picture you posted, the AF cut off is further away but it's still there
2. There are multiple missing objects in the picture you posted. Immediately you notice the missing paper decals on the road but take a look at the yellow stop lights, the missing lamp poles and that yield sign in the distance
That's a good way to free up draw calls.
Well he is sort of right. I'll effect on quality, yes if you take the time and pay really close attention you will find it. But how come no one has notice before? I would say because it is very minute.
I am not trying to defend the practice FYI.
:thumbsup:Because at the upper levels of IQ settings, the returns diminish. In most games, going from High to Ultra tanks performance, but provides very little in terms of IQ gains. I've seen this in many recent titles, hence I tweak games to find the best sweetspot on IQ/perf.
All it takes is for one vendor to cheat and make "Ultra" slightly less "Ultra" and most wouldn't notice it, except for the faster performance, because a raw number is more obvious.
This used to happen a long time ago, both ATI & NV did cheating optimizations but we thought the call out back then, meant it no longer happens.
http://www.geek.com/games/futuremark-confirms-nvidia-is-cheating-in-benchmark-553361/
NVIDIA forced AF
That picture only furthers the problem
1. The photographer is closer in the picture you posted, the AF cut off is further away but it's still there
2. There are multiple missing objects in the picture you posted. Immediately you notice the missing paper decals on the road but take a look at the yellow stop lights, the missing lamp poles and that yield sign in the distance
That's a good way to free up draw calls.
For comparison this is default settings by the same user as your screenshot (should take care of issue 1):
It's safe to say that there is a clear difference in AF, and also a clear difference in performance (roughly 10%)