Do AMD graphics look better?

[papajo]

I remember that being true waaaaay back in the AGP era and popular magazines of the day e.g PC magazine had some test done on that but since then I totally just fazed out of such comparisons and was more like an FPS/$ kinda guy 

 

 

But incidentally I watched a few minutes ago a comparison between RX 5700 xt vs RTX 3080  and  I liked the collors more on the AMD clip and then I start pausing the scenes and noticed that RTX 3080 seems more blury besides having more blunt colors 

 

 

 

Then I thought it might be that the guy messed up in post-production with the nvidia clip so I took a look into other comparison videos... e.g this one..

 

 

Which is kinda the same in terms of fidelity/difference...  Did that guy mess post-production as well? what a coincidence... so I searched for more like this one... 

 

 

 

Again... It's not like nvidia graphics look bad its just that they are a "click" less pleasing to the eye compared to AMD graphics...  (obviously when you guys check use the highest resolution at full screen because as I said the difference is subtle but its there at least using my monitor and eyes )  

 

 

 

Like even in the Lara Croft scene I can see more threads on her shirt on the AMD side lol

 

I think linus should make a more "sciency" comparison in having e.g two identical monitors side by side and stopping the game at the same scene and having his friends judge which one looks nicer

 

And If this is true I wonder if nvidias bigger FPS counts have something to do with that? maybe by having blunter colors and being slightly less defined per frame you increase significantly your average FPS count... who knows

[HanZie82]

Im guessin they did not use a proper capture card but rather the graphics cards built in encoder to reocrd the video.
Thus skewing the result slightly as NVIDIA had hardware encoding specificly for that, Not sure if the newer AMD cards use hardware or thats its all software done.

Thats my guess.

[Zando_]

What recording solution are they using? Is it Shadowplay vs ReLive or are they using an external/internal capture card with the same settings for both? If AMD's RIS is on (IIRC it defaults to on in any DX11 games), that'll have a sharpening effect. 

[papajo]

1 minute ago, HanZie82 said:

Im guessin they did not use a proper capture card but rather the graphics cards

All three of them (and I looked at more I just didnt want to clutter the topic)  ? They are doing benchmarks one would think they would have used the external capturing ....

 

And that being said I think that nvidia has better/more capable capture abilities so even if that is the case then the nvidia ones should look better 

[papajo]

3 minutes ago, Zando Bob said:

What recording solution are they using? Is it Shadowplay vs ReLive or are they using an external/internal capture card with the same settings for both?

again I believe (since they are 3 different channels which I am not affiliated with and I saw other videos too I just did not want to fill the topic with videos) that shadow play is better than relive and also find it highly unlikely that ALL did the same thing 

[Mateyyy]

I remember this being a thing back in the day, but nowadays not so much, at least from what I've heard and from personal experience (swapping between an R9 290 and an RTX 2070 resulted in no change in colour reproduction whatsoever for me).

[LOST TALE]

1 minute ago, papajo said:

again I believe (since they are 3 different channels which I am not affiliated with and I saw other videos too I just did not want to fill the topic with videos) that shadow play is better than relive and also find it highly unlikely that ALL did the same thing 

I use Relive and have used shadow play until 2 years ago. What's your experience?

[Zando_]

Just now, papajo said:

again I believe (since they are 3 different channels which I am not affiliated with and I saw other videos too I just did not want to fill the topic with videos) that shadow play is better than relive and also find it highly unlikely that ALL did the same thing 

You cannot draw accurate conclusions from something with that many variables, especially if the people doing the testing didn't state their methodology. 

[Mark Kaine]

I always thought nvidia looks "better"

 

Always found it funny ps3 had nvidia and xbox had amd back in the ps360 days... 

 

https://en.m.wikipedia.org/wiki/RSX_Reality_Synthesizer

 

And this looked always better to me, colors etc... 

 

 

 

I know people always think it doesn't matter, but of course it does, image sharpening, anti aliasing, after effects, stuff like rtx, these will always make a difference. 

 

What one prefers or even cares about is up to the individual tastes (or lack thereof) 

 

[papajo]

9 minutes ago, Zando Bob said:

You cannot draw accurate conclusions from something with that many variables, especially if the people doing the testing didn't state their methodology. 

well many variables yes but all the same result ?  I mean e.g a football match has many variables but if Team 1 beats team 2 in all their games then you can say team 1 is better 

 

here is an other one 

 

 

But I agree with you that this is not like a "proof" but its a strong indication hence I suggested linus making a video about it

[papajo]

8 minutes ago, Mark Kaine said:

What one prefers or even cares about is up to the individual tastes (or lack thereof) 

 

yes but  I am talking about blunter colors, less detail or blurriness...on a tiny but still noticable scale at least when you pause the frame and make comparisons (obviously while not much movement is going on) 

 

 those are not subjective things (its only subjective to say I like it being blurier, but it being blurier is an objective thing if it is and its not just me/my monitor or whatnot ) 

[Kilrah]

11 minutes ago, papajo said:

here is an other one 

I'm picky, and I see absolutely no difference (other than game content, at times for example the AMD side has more fall leaves on the ground so it'll look more warm/full, or one side has more fog than the other).

[Zando_]

7 minutes ago, papajo said:

well many variables yes but all the same result ?  I mean e.g a football match has many variables but if Team 1 beats team 2 in all their games then you can say team 1 is better 

here is an other one

-snip-

But I agree with you that this is not like a "proof" but its a strong indication hence I suggested linus making a video about it

This is pretty different from sports which have clearly defined rules as to what constitutes a win. Answer bleeds more into this:

5 minutes ago, papajo said:

yes but  I am talking about blunter colors, less detail or blurriness...on a tiny but still noticable scale at least when you pause the frame and make comparisons (obviously while not much movement is going on) 

 

 those are not subjective things (its only subjective to say I like it being blurier, but it being blurier is an objective thing if it is and its not just me/my monitor or whatnot ) 

Those are all down to how/with what it was recorded. That's the reason I've looked at AverMedia capture cards over the Elgatos, the AverMedias capture better colors and contrast. Even on the exact same computer. The recording looks different, the actual game played on the monitor and viewed by the person playing it looks the same every time. So if it's a difference between Shadowplay and ReLive then it's just one capturing higher quality colors and detail than the other, and not any indication that the game looks any better to the actual player. 

[papajo]

1 minute ago, Zando Bob said:

Those are all down to how/with what it was recorded.

So all the channels are using a capture method (the same one) that favors AMD? I find this too far fetched. 

[Zando_]

Just now, papajo said:

So all the channels are using a capture method (the same one) that favors AMD? I find this too far fetched. 

If they're using the built in capture functionality of the cards, and ReLive captures better colors/detail, this would make sense. It's well known that Shadowplay/Nvidia's NVENC recording is lower quality than CPU encoding, it being slightly behind ReLive on colors/detail wouldn't be surprising. It's used for the very low performance impact + a lot of streaming, and livestreams have such heavy compression that a higher quality recording would be useless on the viewer's end anyways. 

[papajo]

2 minutes ago, Zando Bob said:

This is pretty different from sports which have clearly defined rules as to what constitutes a win.

well in the context of this topic its clear what constitues a win as well... is it blurrier? does it have less detail? are the colors blunt? 

 

e.g check the water or the plants 

 

(AMD left <> Nvidia Right) 

Or here check the rail  the nvidia one is blurier or the guys clothes the AMD one's are warmer in color (also the rail ) 

\

 

(AMD left <>Nvidia right) 

 

Or here check the "ball" on that gus or the detail in the corner or the detail in the cracks 

 

 

(nvidia left <>AMD right) 

 

 

Obviously it's better if you full screen and max out the resolution and watch it from the benchmark videos directly but I screenshoted them just to make a point 

[Zando_]

3 minutes ago, papajo said:

-snip-

Again. This is totally dependent on what is being used to record. You'd need to record both things with the same device to then be able to compare how they look to that same device. 

 

I agree with the idea that someone should set out to actually test this, but trying to draw any sort of accurate conclusion from the dataset you have here is utterly inane. 

[papajo]

3 minutes ago, Zando Bob said:

Again. This is totally dependent on what is being used to record. You'd need to record both things with the same device to then be able to compare how they look to that same device. 

 

And that is probably what they are doing... who uses two different capture devices for the same video? 

 

Also as I said I find it farfetched that ALL the channels used the same capture device or method. 

 

 

Last but not least PM linus to make a video about it

[Mark Kaine]

What I don't get in the last picture the colors are very different, like someone put a brown filter on or something - also as usual we have no way of knowing resolutions, settings etc... Which is why generally such youtube comparisons are rather useless to me and often outright deceiving. 

 

 

[Zando_]

Just now, papajo said:

And that is probably what they are doing... who uses two different capture devices for the same video? 

 

Also as I said I find it farfetched that ALL used the same capture device or method. 

 

 

Last but not least PM linus to make a video about it

To quote myself from 3 minutes ago: "trying to draw any sort of accurate conclusion from the dataset you have here is utterly inane. "

The addition of fucking "PROBABLY" does not in any way help your point

[papajo]

7 minutes ago, Mark Kaine said:

What I don't get in the last picture the colors are very different, like someone put a brown filter on or something - also as usual we have no way of knowing resolutions, settings etc... Which is why generally such youtube comparisons are rather useless to me and often outright deceiving. 

 

 

We have system/res etc are in the video as a layer (they are benchmark videos) or under the video description. 

 

As for the lighting in the last screenshot (which can be found watching the 3rd video in the OP it could be that the game has a clock and changes day/night depending on the clock or it could be a difference on how each architecture translates some lighting effects. 

[Mark Kaine]

2 minutes ago, papajo said:

We have system/res etc are in the video as a layer (they are benchmark videos) or under the video description. 

And I have a bridge to sell to you...

[papajo]

1 minute ago, Mark Kaine said:

And I have a bridge to sell you...

well I just picked up videos starting with RX 5700 vs 3070 or other nvidia cards from different channels you can do the same.. I doubt that there is a global conspiracy to make the AMD scenes look better infact I doubt the channels themselves noticed that since they are making FPS comparisons  I wouldnt notice it either I did it out of coincidence  

[Mark Kaine]

1 minute ago, papajo said:

well I just picked up videos starting with RX 5700 vs 3070 or other nvidia cards from different channels you can do the same.. I doubt that there is a global conspiracy to make the AMD scenes look better infact I doubt the channels themselves noticed that since they are making FPS comparisons  I wouldnt notice it either I did it out of coincidence  

That is somehow my point tho, could be they used different settings... These videos are heavily edited too, you can't say "but we saw the settings". 

 

Additionally I watched a bit (as far as my tolerance for terrible animations allowed) and I didn't really see a difference, it goes back and forth which is better, which is probably due to the horrible youtube compression more than anything. 

 

And that last picture I pointed out, the AMD side looks clearly better, and it looks also heavily tinkered with as the color differences are way too stark. 

[i_build_nanosuits]

Splitting hairs here...both cards running the same game will look exactly the same...