AMD & NV image quality comparison

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Good luck. So far all the screen shots I've seen at OC.UK show nothing.

The source of this clusterfunk gaffed their video source so bad it became an issue with star eyed Radeon fans using this as a poster for Anti-NV smear.

The fact that people in the original source thread can't tell the difference between black crush and "it looks so much better on the AMD side" makes me ashamed to be a Radeon fan. Win with data, not emotions and desires.

As it stands outside of some bugs (as noted for GTA 5) there seems to be zero proof of an IQ issue.

Unless, you want to use a modified video as your source. If you do, let me laugh at you too.
 

Phynaz

Lifer
Mar 13, 2006
10,140
819
126
The days of differing image quality are long gone, that's why reviews don't include them anymore.
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Link to the OC.UK thread please?

There are a few threads involved, it's a nice cluster of things to through. Just think about this for a second - the way the digital age is, especially with Reddit, you think it would be some random video blogger who gaffed his captures that would be the first to bring this to attention? Not trying to slam the guy but his honest mistake has led to one hell of a witch hunt where people are essentially becoming contortionist to some how make his mistake true.

Have fun reading:
http://forums.overclockers.co.uk/showthread.php?t=18679713

http://www.overclock.net/t/1563118/...-iq-between-titanx-980ti-and-amd-kepler-cards

http://forums.overclockers.co.uk/showpost.php?p=28258333&postcount=2391
 

Black Octagon

Golden Member
Dec 10, 2012
1,410
2
81
Even if there were a difference, would a screenshot even show it? I'd have thought you'd need to take snaps with a camera
 

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Even if there were a difference, would a screenshot even show it? I'd have thought you'd need to take snaps with a camera

Glaring obvious missing textures - yes. That would show up on a screen shot. Which is what some people are screaming is happening here. If you review the source video and have knowledge about crushed blacks, you'd laugh at the people saying there is detail missing on the Titan X while ignoring MORE detail missing on the Fury X side due to limited color range.

But after watching the video a few times, reading the threads (I'm so bored today haha ), it really comes down to one guy gaffed his videos (adjust Fury X color scaled, captured Fury X side at 30mbits and Titan X side at 12mbits, I mean at this point it sounds he was intentionally trying to give the Fury X side an advantage - one mistake, sure, one deliberate modification, yeah that's malice).

Hell, trying to find something for people who seem uninformed about black crush brought me to this:

https://creativemarket.com/blog/201...ed-black-film-look-in-photoshop-and-lightroom
Is It Better?

It's important to note that, from a strictly empirical standpoint, we've actually made this photo worse, because we've removed pixel detail. However, people's personal preferences when it comes to photography have nothing to do with empirical measurements of pixel data. Some people absolutely hate this look, others can't get enough. Whether the photo is "better" or not after the adjustment in terms of aesthetics is completely in the eye of the beholder.

Keep in mind though that this does represent a popular trend. The upside of following trends is that your work stays fresh and in line with what people like. The downside is that trends change quickly, so what's popular today might make your photo look ugly and out of date tomorrow. Be especially aware of this as you build your professional portfolio. Hang on to those unedited versions of your images, they may come in handy when it's time to update your site later!

And if the guy who altered his video is of that mentality - it looked better to him. And there are tons of Xbox fans I know who will fight you tooth and nail that XBox 360 had the better color scale, well until Xbone launched.
 

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Yet another example of just how much detail black crush loses:



You can barely make out the details on the textures used on the home on the right side of the screen or the NPC literally a few feet ahead of you. But people who watched the gaffed video would say "the right looks so much better, the left looks washed out." :hmm:
 

amenx

Diamond Member
Dec 17, 2004
4,397
2,717
136
I dont doubt their may be the odd bug or two in some game that may make one card look better than the other, but 'overall IQ', no. I dont have GTA 5 yet but from what I've gathered it does seem to have a bug that affects anisotropic and texture filtering:

https://www.google.co.uk/search?q=gta+5+nvidia+blurry

Another thing that may mess with IQ (although with some its a matter of taste) is in game bloom setting. Hated it in Skyrim, FC4, Witcher 3 and a couple other games where it can give a 'washed out' sort of effect. When you max out a game, it is often enabled so people may tend to think their card has crappy IQ. Disable it and everything is clear as a bell. Whether poor implementation of game code or drivers, I dont know.

I remember last time we had this discussion someone posted 660ti SLI vs 7950cf comparison screenshots of Sleeping Dogs and was inferring lacking IQ of Nvidia:



And curious I started the game and went to same spot and got this with my Nvidia card:



Full set of pics showing supposedly "better" AMD IQ:

http://forums.anandtech.com/showpost.php?p=34849227&postcount=103

My set of pics in same locations (different times of day/night, shop lights on/off can give different results):

http://forums.anandtech.com/showpost.php?p=34851702&postcount=134
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Oh, I agree bugs do exist and they do affect IQ in some games. Those, in my opinion, are easy to notice and most importantly easy to replicate.

It's when people question the source material, who then fumbles his answers and suddenly he's (not by his own intention) lifted into the air as some hero.

Your example is jarring because that isn't a simple difference in IQ, that is either intentional turning off of features or NV had a bad driver. Which of the two, you can argue.

So far for BF everyone who took the time to post images can not recreate the things they see in the video. I bet you I could. If I had the right software.
 

railven

Diamond Member
Mar 25, 2010
6,604
561
126
There you have it. Original guy who captured the first two videos redoes it:

https://www.youtube.com/watch?v=zC-VpCnexkU

No black crush.
Suddenly "the solid white lines" look faded and blurred on both sides.
The "zebra crossing" looks identical (covered by papers and debris).

But now you got another question:
Both sides show lower frame rates than the original videos he captured? Haha.
EDIT: I will attribute the lower FPS on both due to him capturing at a higher frame rate. If he did as he said and used exact same IQ/Panel settings as his first attempt.
 
Last edited:

hawtdawg

Golden Member
Jun 4, 2005
1,223
7
81
The fact that people in the original source thread can't tell the difference between black crush and "it looks so much better on the AMD side" makes me ashamed to be a Radeon fan. Win with data, not emotions and desires.


The fact that anyone in that thread pushing this "black crush" conspiracy, thinks they even have any clue in the world what their talking about is the appalling part. I've never read so much ignorance in one thread. Confusing character shadowing, and incorrectly set gamma (and possibly video levels) for some sort of black crush conspiracy that magically hides detail, is about the most misinformed thing i've ever read.
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
The fact that anyone in that thread pushing this "black crush" conspiracy, thinks they even have any clue in the world what their talking about is the appalling part. I've never read so much ignorance in one thread. Confusing character shadowing, and incorrectly set gamma (and possibly video levels) for some sort of black crush conspiracy that magically hides detail, is about the most misinformed thing i've ever read.

Haha, I can't tell if you're agreeing with me or against me (rethinking my response to you in the other thread.)

I think this post is what set me off because it reminded me about the black crush issues in consoles and how it was defended during one generation:

This isn't about colors, gamma, blacks, vibrancy of colors on AMD vs washed out on NV (which I have experienced myself).

This is about short lod distance, poor texture filtering, missing details (like bricks on the pavement, long distance objects).

Possible reasons for the poorer rendering:

1. The guy fake his control panel, overriding settings. Such as disabling AF (which still won't explain the lod distance problem). <- possible, we can't trust randoms on the interwebz!
2. Youtube compression causes one-half of the video to be worse. <- unlikely
3. The guy actually have Titan X on High settings in-game and not Ultra. <- possible.
4. NV has a driver bug for some games. <- possible.
5. NV has a driver "feature" or "optimizations" for some games. <- possible.

The original capture video was terrible, but some people took that as "NV is cheating with IQ settings. FPS crown in question!"
 

hawtdawg

Golden Member
Jun 4, 2005
1,223
7
81
Haha, I can't tell if you're agreeing with me or against me (rethinking my response to you in the other thread.)

I think this post is what set me off because it reminded me about the black crush issues in consoles and how it was defended during one generation:



The original capture video was terrible, but some people took that as "NV is cheating with IQ settings. FPS crown in question!"

look at this image:


The XBOX pic clearly has shadowing and shading that the PS4 version does not. Look at the shadow being cast by the grenades. Look at the shadow being cast by his goggles, look at the shadow being cast by his head onto the exo-frame. You're confused into thinking that the image should be washed out enough so you can make everything out clearly. There is not detail being hidden, there are shadows being cast that aren't in the PS4 image, and better ambient occlusion being used. the XBOX image does look better, because it is better.
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
look at this image:


The XBOX pic clearly has shadowing and shading that the PS4 version does not. Look at the shadow being cast by the grenades. Look at the shadow being cast by his goggles, look at the shadow being cast by his head onto the exo-frame. You're confused into thinking that the image should be washed out enough so you can make everything out clearly. There is not detail being hidden, there are shadows being cast that aren't in the PS4 image, and better ambient occlusion being used. the XBOX image does look better, because it is better.

Okay, clearly you don't know what black crush does either, here you go:

In all, it's our first chance to see Battlefield 4's multiplayer mode in action on Xbox One, and to witness either version of the game running on final retail hardware. On the visual side, HBAO is indeed added to the Xbox One version, as promised by DICE following its omission in the Stockholm build. However, the divide in internal resolution - which remains at 900p on PS4 compared to the straight 720p on Xbox One - is the big sticking point, even with the removal of the Xbox version's sharpening filter which amplified the aliasing. Put simply, the assets and effects are identical for each next-gen platform, but the added pixel definition on PS4 spares it from the distracting edge-crawl on thin objects that proves to be an issue on the Microsoft side.

Shadows had to be updated on the Xbone version to match the PS4, but still dat black crush! It created the illusions you mistake as "better shadows"

https://www.youtube.com/watch?v=ogqpvVR4kEo

If you can't see the black crush on the Xbone version - there is nothing left for us to discuss. This is a known issue for MSFT. So bad they already tried to address it TWICE. BF4 came after the first address of the issue. [DICE] studios said it was due to the upscale algorithm MSFT was using to get from <1080p on the game render to 1080p for output.

EDIT:

Correction, it was Dice that addressed the scalar, not Respawn (they fixed the controller issue)

It's interesting to note that the majority of native 1080p titles - including Need for Speed: Rivals, Lego Marvel Super Heroes and Tomb Raider - do not have gamma issues at all. Developer contacts tell us that the phenomenon is linked to the use of the Xbox One scaler in combination with the utilisation of multiple display planes (allowing native and sub-native elements to blend in the final framebuffer). This is backed up by our experience with Battlefield 4. The pre-final code utilised the Xbox One hardware scaler and featured an unnatural gamma shift, while the release version saw the implementation of a software scaler created by DICE itself that resolved the colour balance issue in addition to removing the sharpening effect.

Note: Improved it, it didn't completely fix it, as evidence by the Youtube video above.

EDIT #2:
Before and After shots:
http://images.eurogamer.net/2013/ar...nch_2_copy.bmp.jpg/EG11/quality/90/format/jpg
http://images.eurogamer.net/2013/ar...tch_2_copy.bmp.jpg/EG11/quality/90/format/jpg

Anyways, this is getting OT. But there are people that argued tooth and nail that the Before looked better.
 
Last edited:

hawtdawg

Golden Member
Jun 4, 2005
1,223
7
81
Okay, clearly you don't know what black crush does either, here you go:



Shadows had to be updated on the Xbone version to match the PS4, but still dat black crush! It created the illusions you mistake as "better shadows"

https://www.youtube.com/watch?v=ogqpvVR4kEo

If you can't see the black crush on the Xbone version - there is nothing left for us to discuss. This is a known issue for MSFT. So bad they already tried to address it TWICE. BF4 came after the first address of the issue. [DICE] studios said it was due to the upscale algorithm MSFT was using to get from <1080p on the game render to 1080p for output.

Give me one single timestamp in that video that shows the Xbox crushing blacks.



EDIT #2:
Before and After shots:
http://images.eurogamer.net/2013/ar...nch_2_copy.bmp.jpg/EG11/quality/90/format/jpg
http://images.eurogamer.net/2013/ar...tch_2_copy.bmp.jpg/EG11/quality/90/format/jpg

Anyways, this is getting OT. But there are people that argued tooth and nail that the Before looked better.
That's an issue with the sharpening filter (and some weird shadow artifacting on one of the characters), not black crush.

I'm not sure you even understand what black crush is. You're clearly confusing the difference between a 2.2 and a 2.4 gamma (and shadow effects that are simply not existent in the PS4 version).

The logic you're using would say that the left half of this image has "black crush".

 

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Give me one single timestamp in that video that shows the Xbox crushing blacks.

Whoops, linked you the wrong video:
https://www.youtube.com/watch?v=8XqWRACk2zY

Let's try that experiment again.

That's an issue with the sharpening filter (and some weird shadow artifacting on one of the characters), not black crush.

I'm not sure you even understand what black crush is. You're clearly confusing the difference between a 2.2 and a 2.4 gamma (and shadow effects that are simply not existent in the PS4 version).

The logic you're using would say that the left half of this image has "black crush".


I'm using the term as it is described in the photography/filmography sense. There is a RGB range difference (as you said) where the console was setting the upscale end result to limited. It was truncating the end scale of the both black and white. The sharpening issue is addressed by the clear softening of the image quality. The black levels by what the Xbone original scalar was doing.

Dinner, see that youtube video, tell me there is no issue. Ciao!
 

hawtdawg

Golden Member
Jun 4, 2005
1,223
7
81
Whoops, linked you the wrong video:
https://www.youtube.com/watch?v=8XqWRACk2zY

Let's try that experiment again.



I'm using the term as it is described in the photography/filmography sense. There is a RGB range difference (as you said) where the console was setting the upscale end result to limited. It was truncating the end scale of the both black and white. The sharpening issue is addressed by the clear softening of the image quality. The black levels by what the Xbone original scalar was doing.

Dinner, see that youtube video, tell me there is no issue. Ciao!


Well yeah, in that video you would be right. clear black clipping. Still though, that's totally on the developer for releasing it that way.

The original ps4/XB1 image you posted however is not a case of black clipping. That is clearly character shadowing and ambient occlusion, and possibly a higher gamma. There are multiple shadows clearly being cast in the xbox pic that arent in the ps4 pic (grenades are most obvious, but the shadow being cast by his head onto the exo and left shoulder/arm is pretty obvious too.) There is nothing being clipped there.
 
Last edited:

Tapoer

Member
May 10, 2015
64
3
36
Gregster found out what the problem was:
http://forums.overclockers.co.uk/showthread.php?t=18679713&page=7
Gregster said:
And I think we have found out what is going on. The default profiles let the 3D application decide what to do but switching to 'prefer max quality' makes that line nice and white

The default settings/profile of Nvidia control panel don't produce the best quality, at least in BF4, only with "prefer max quality" AF was working as it should at the cost of ~10% performance hit on a Titan X.
I wonder how many more games this happens and how many reviews are done with the default setting in the control panel or profiles.
 
Last edited:

Hitman928

Diamond Member
Apr 15, 2012
6,641
12,244
136
There you have it. Original guy who captured the first two videos redoes it:

https://www.youtube.com/watch?v=zC-VpCnexkU

No black crush.
Suddenly "the solid white lines" look faded and blurred on both sides.
The "zebra crossing" looks identical (covered by papers and debris).

But now you got another question:
Both sides show lower frame rates than the original videos he captured? Haha.
EDIT: I will attribute the lower FPS on both due to him capturing at a higher frame rate. If he did as he said and used exact same IQ/Panel settings as his first attempt.

The updated video with both recorded at the same quality shows the exact same issue as before with the TitanX, a clear lack of AF, texture detail, and possibly draw distance of objects. This has nothing and never had anything to do with the color or gamma settings or "black crush". This is simply that his TitanX video shows lowered LOD and there are very clear screenshots from the video that point this out. You can watch the video and see it clearly. This could very well be an issue with his setup or how he recorded, but the difference is clear.

If you really did read the thread, then to try and call him out is flat out tasteless. The only reason he did the videos is to give back to the community and he had no intention of starting any of this, it was other posters who saw the video and noticed the quality difference. The OP didn't want to be involved in any controversy and that's why he was reluctant to add further material to the discussion.

Lastly, there are only a couple posters (as expected) who are trying to make this into a hit piece against nVidia, the vast majority of people have simply noticed what is clear in the video and are interested to look into it further. Again, this very well could be an issue with a very simple answer, but it deserves a little investigation. Those trying to throw out all these red herrings as to make it seem like it's all a farce are either being ignorant or dishonest.
 

Hitman928

Diamond Member
Apr 15, 2012
6,641
12,244
136
Gregster found out what the problem was:
http://forums.overclockers.co.uk/showthread.php?t=18679713&page=7


The default settings/profile of Nvidia control panel don't produce the best quality, at least in BF4, only with "prefer max quality" AF was working as it should.
I wonder how many more games this happens and how many reviews are done with the default setting in the control panel or profiles.

Yep, that's what many had guessed was the issue, the default setting in the nvidia control panel seems to have issues with LOD in some games and forcing it to highest quality fixes the issue. I would think nvidia users would want to know this so they can make sure to set it correctly for highest quality.

edit: from the op @ overclockers uk

Nvidia control panel default settings:

Set to highest quality:
 
Last edited:

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
Yep, that's what many had guessed was the issue, the default setting in the nvidia control panel seems to have issues with LOD in some games and forcing it to highest quality fixes the issue. I would think nvidia users would want to know this so they can make sure to set it correctly for highest quality.

is there a performance delta with these settings of just lack of visual fidelity?
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |