Digital Foundry Article Technical Discussion [2021]

Status
Not open for further replies.
I think this may have been the game.
--------------------

Could this be attributed to XBSX possibly using VRS in Control or a lower precision RT reflection during a given scene? (Not my image by the way)

ySUzduJ.jpg
So this one is solved by a member in the DF discord, Cliptacular, this is a bug.
Only 1 of 2 maps reflect as shown by the video below. I apologize for it being so low res, but w/e you get the point.
 
I guess Control’s photo mode is also a best case scenario for memory contention (i.e., the CPU not stealing bandwidth from or adding latency to the GPU’s requests)?
 
Is this photomode present on the 8th gen versions? (also with unlocked framerate)
Those might be some interesting results.
 
What I found really jarring while playing the game in Graphics mode was that, since the ray traced reflections are reconstructed from multiple previous frames, you can spot it lagging behind the reflected world and cause a jelly effect, which is especially noticeable when panning the camera.

I would like to see them using the spare computation headroom on consoles to fix this, for example, adjust the number of previous frames used depending on the load.
One way to improve this is to use seperate, proper motion vectors to reflections.
Pretty sure Frostbite had something on this in their presentations.
 
Imo, the CPU sharing power with the GPU is the worst case scenario. And in areas where no CPU is required all the power draw will go towards GPU. It's unlikely to draw additional power away from the GPU in these scenes because the CPU has nothing to do, this is in fact the best case scenario for PS5 if you want have the absolutely maximum amount of frame rate. As per Cerny's words, the PS5 handles its clocking based on the code, so if it detects it's going to spin up for no reason and fly it to the moon, it will clock down to conserve that power. I don't want to have to go through the numerous amount of post history that tried to tell me for a long time that both the CPU and GPU would run at it's maximum clocks for PS5 and that' there's no way the CPU would 'ever' dip the GPU down enough to drop below 10TF. And now with this perfect scenario, no CPU, all GPU, it's going to use more power seems counter intuitive to a great deal of discussion we had by multiple members earlier on before release.

I've seen this crop up several times, cache sizes, joined caches, latency etc. Frankly I don't know what to make of it, seems too big of a deal for 2 consoles apparently both using Zen 2 + RDNA 2. But I mean, all the games in which we've seen 'PS5 Win' is basically a discussion around whose dropping below the target frame rate of 60fps more. You don't actually see how far either console goes above it. Is it possible that XSX could both be outperforming PS5 in unlocked scenarios and under stress also under performing it? Sure it can. Do we know if it is? No. But we are assuming it can't because we haven't seen many unlocked frame rate games. And not the recent ones either.

PS5 has been, as far as I can see, very consistent in it's performance since launch (BC aside). And XSX not so much.
So I'd probably say, XSX got a problem instead of PS5 having secret sauce.

imo, If you want to know what's up with the dipping on XSX, stop looking for things on PS5 as it's not dipping and look for where XSX could be doing things differently.
But that's not at all what PS4 and Pro games are teaching us. We know they have being studying countless of PS4 and Pro games power consumption with different loads in order to design the PS5 dynamic clocks. The aim of Sony was to give the max power during the most meaningful scenes (during gameplay). Cerny has even mentionned (or implied) the fan noise was a direct indicator or power consumption when playing a game and that noise was the strongest usually not during gameplay but during specific scenes like cutscenes or map screens.

All those moments are sharing a common particularity. Usually the GPU is not restrained by any game logic and can render whatever task it is doing (simple or not) the most that it can do (and sometimes it is somehow not limited by vsync limit because for instance the GPU is being uselessely rendering any number of shaders it can, I remember reading this in a presentation, I think that was either Infamous or Killzone SF).

If you owned a PS4 or Pro and studied the moments of most noisy fan spinning (which I have being doing since 2014, my poor ears...), you'll know immediately that oddly in most cases (but not all) the fan is spinning the most during non-gameplay scenes: map screens, cutscenes, photo mode or when you enter the dashboard while running the game in the background.

On the other hand games like Horizon, TLOU2, God of War or MGS5 (one of the most impressive games on PS4) are usually reasonnably noisy during gameplay. Again the fact that DF are using the start screen of God of War in order to test the max power consumption on Pro (a scene comparable to a cutscene with plenty of details but no game logic) is another proof of what I am saying.
 
Last edited:
Other pics have popped up with XBSX having missing textures or decals in other scenes, so more than likely they are bug related as well.
Yeah very much so classified under bug. There are areas on PS5 which also show the wrong or missing decal in a reflection in areas where it is not on XSX. Like this opening area of the containment sector, the shelter/vault on the far side after the initial combat scenario.
Like every game, there are unintended differences between the versions due to oversight, even though the textures, streaming settings, ray tracing settings etc. are in actuality all the same.
 
Like every game, there are unintended differences between the versions due to oversight, even though the textures, streaming settings, ray tracing settings etc. are in actuality all the same.
Even with identical code, once you feed that code through different compilers you can get different results. Not to mention they are using different graphics APIs. It almost definitely isn't worth the effort to tweak either/both to reach visual parity. It may not even be possible.
 
Another thing to note is that when comparing ray tracing performance in photo mode compared to real-time gameplay is that the BVH doesn't need to be rebuilt for static scenes in the case of the former so this comparison might not be representative of the performance you'd see in real-time gameplay between different hardware ...

I don't exactly know what heuristic Remedy is applying to determine whenever they do a BVH rebuild in Control and I don't know if they're potentially doing the optimization that I stated above ...
 
@Dictator

Weird that with some of the pictures you show in your video, that the quality of the images, has XBSX images and RT quality worse than on the PS5.

It's been pointed out in another video, so I can't claim ownership, pointing out the clear differences and others but I thought I would post your time links to the images in the video in 4K where you can clearly see the images and RT is better on the PS5. Maybe you can post a video analysing why the quality is different.

I can post the link if the guys video analysing your video if you want but I think you can find it on your own.

 
Last edited:
The PS5 Version suffers from Black crush - which will exaggerate detail at times while eliminating others. A known issue we logged with remedy while looking at the game pre-release.
Imo - it is best not to look for minor differences of colour and Variation or even pure detail as each shot I made required me to line up the Camera manually. They are not Pixel perfect across platforms, I am not a Machine sadly.
 
@Dictator

Weird that with some of the pictures you show in your video, that the quality of the images, has XBSX images and RT quality worse than on the PS5.

It's been pointed out in another video, so I can't claim ownership, pointing out the clear differences and others but I thought I would post your time links to the images in the video in 4K where you can clearly see the images and RT is better on the PS5. Maybe you can post a video analysing why the quality is different.

I can post the link if the guys video analysing your video if you want but I think you can find it on your own.


We know about the issue of the missing texture in the second link. It's a bug, and @iroboto posted a gif about it above:


What worse RT am I supposed to be seeing in the other shots though?
 
The PS5 Version suffers from Black crush - which will exaggerate detail at times while eliminating others. A known issue we logged with remedy while looking at the game pre-release.
Imo - it is best not to look for minor differences of colour and Variation or even pure detail as each shot I made required me to line up the Camera manually. They are not Pixel perfect across platforms, I am not a Machine sadly.
I don't think the differences can be attributed to the camera angles though, there's substantial differences in the images themselves.Differences.jpg
 

Attachments

  • Differences.jpg
    Differences.jpg
    346.9 KB · Views: 15
  • Differences2.jpg
    Differences2.jpg
    390.8 KB · Views: 16
  • Differences3.jpg
    Differences3.jpg
    577.9 KB · Views: 13
  • Differences4.jpg
    Differences4.jpg
    325.1 KB · Views: 14
Other than the known bug, what "worse" RT am I supposed to be seeing there?

It might be there, it's just I can't see it (I'm on a 1080p display, if that makes a difference). What is it?
Hey, you said it was a bug. I understand that now as I wasn't aware that it was a bug. Please explain away the other differences in the other pictures because different camera angle doesn't do it for me.
 
As an example - the texture Streamer actually stops streaming in photo mode at the moment you freeze the game. So in that scene for example, i started photo Mode on xsx immediately after the elevator door opened. On PS5, Jesse had already been sitting there for a while. So textures may perhaps not be fully streamed to highest mips on xsx.II have no idea.
Also, there could be z fighting differences between versions for decals.
There are thousand of reasons. IMO load the game up on both there and make some recording yourself. None of any Version differences have to Do with the game's settings which have been confirmed to be the same by... the developer.
 
As an example - the texture Streamer actually stops streaming in photo mode at the moment you freeze the game. So in that scene for example, i started photo Mode on xsx immediately after the elevator door opened. On PS5, Jesse had already been sitting there for a while. So textures may perhaps not be fully streamed to highest mips on xsx.II have no idea.
Also, there could be z fighting differences between versions for decals.
There are thousand of reasons. IMO load the game up on both there and make some recording yourself. None of any Version differences have to Do with the game's settings which have been confirmed to be the same by... the developer.
Well then explain it in the video when you are doing the comparisons. Could this also not contribute to the fps difference in a minor way. You are doing yourself no favours when showing off screenshots like these and there's differences, it just looks like you are sucking up to MS, which I am sure you are not but it looks like it. It's an impression you are giving off.
 
Well then explain it in the video when you are doing the comparisons. Could this also not contribute to the fps difference in a minor way. You are doing yourself no favours when showing off screenshots like these and there's differences, it just looks like you are sucking up to MS, which I am sure you are not but it looks like it. It's an impression you are giving off.
Seriously? You must know from watching DF that my investment in next gen consoles is purely intellectual and I quite literally only care about games on PC and the PC platform.
 
One digression about "developer said settings are the same" - I'm not saying it's not true in Control but know one example - Asseto Corsa - when devs said physics are the same on consoles and on pc and that clearly wasn't true as pc car setups doesn't works same on consoles
 
Status
Not open for further replies.
Back
Top