Digital Foundry Article Technical Discussion [2021]

[Globalisateur]

Here, I don't think I agree with this quote.



I'd imagine that the checkerboard rendering is the same on the PS4 Pro as it is on the PS5, however, when run at 30-40 FPS rather than 60 FPS, the temporal accumulation will be more prone to breaking down due to the greater time difference between temporal frames/data.

Basically, it could be an apple to apples comparison of why high FPS is needed for any temporal reconstruction (like checkerboard rendering) technique rather than evidence of "lower grade" checkerboarding.

I would postulate that if the title was capable of running at 120 Hz, that even the checkerboarding artifacts on vegetation would mostly disappear as there would be far more temporal data available in a short period of time for our brain to process.

Regards,
SB

Why would you imagine such a thing? DF are not the only ones with that claim. VGTech also noticed some noticeable ghosting only on the Pro version.

There are very different ways of doing CBR reconstruction. From bad to very good implementation often depending of how much ressources you allocate into it.

 

[Silent_Buddha]

Why would you imagine such a thing? DF are not the only ones with that claim. VGTech also noticed some noticeable ghosting only on the Pro version.

Yes, and ghosting is a temporal artifact, not an implementation artifact. It exists because there are too few frames being displayed in a given period of time. As frame rate increases temporal artifacts decrease.

Same checkerboard rendering implementation, the difference being the rate the data is being presented to our brain.

Regards,
SB

 

[snc]

Yes, and ghosting is a temporal artifact, not an implementation artifact. It exists because there are too few frames being displayed in a given period of time. As frame rate increases temporal artifacts decrease.

Same checkerboard rendering implementation, the difference being the rate the data is being presented to our brain.

Regards,
SB

nxgamer guessing its not cb on ps5 "ps5 looks to be rendering at native resolution and saving that cost in terms of inner triangles and alpha effects in terms of artifacts around the edges thats why its cleaner sharper image than ps4pro"

 

[snc]

4k on ps5 and xsx, better performance on ps5
xss 1080p
xox and ps4pro 1440p, better performance on xox
loading times ps5 8.4s, xsx 11.8s, ps4pro 20.5s

 

[BRiT]

DF Article @ https://www.eurogamer.net/articles/...-multi-platform-first-party-sony-title-tested

MLB The Show 21: the first multi-platform PlayStation Studios title tested
PS5, Xbox Series X/S, PS4 Pro and Xbox One X under the microscope.

'Sony Interactive Entertainment presents'. We've all seen that splash screen, the white type on the black background transitioning into a polished branding sequence, culminating in the familiar cross/square/triangle/circle iconography - only this time it's not just manifesting on a PlayStation game, it's also happening on Xbox. All of them in fact, from Xbox One to Xbox Series X. This is MLB The Show 21, the first multi-platform release from PlayStation Studios. I've played the game across three Xbox machines and two PlayStations and the results are fascinating.

Quite how we've reached this situation is something we can only really speculate about, but it's likely that Major League Baseball itself - the publisher - opted to move the franchise into the multi-platform arena. It's also likely it's the same outfit that brokered the deal to bring MLB The Show 21 to Game Pass on day one, which not only makes this title an unlikely battle of the consoles but also a stark contrast in the business models that Sony and Microsoft have opted for. From my perspective, MLB The Show 21 was available to download on any of my Microsoft consoles and it was just there as part of my subscription - a couple of button presses to download as I was good to go. Meanwhile, on the PlayStation side, I had to pay £75 for the cross-gen version of the game. Yes, remarkably, there's a £15 price-premium for buying the SKU that allows access to both PS4 and PS5 versions of the game. In terms of value, Game Pass clearly comes out on top - particularly when this is very much a cross-gen game with only relatively minor upgrades separating PS4 Pro and PlayStation 5.

...

 

[BRiT]

Seems like the cutscenes push graphics harder than actual gameplay.

 

[snc]

Seems like the cutscenes push graphics harder than actual gameplay.

thats actually normal, in many games (fifa21, nba2k21, mortal kombat 11) they push graphics harder during cutscenes, but capping at 30 here they don't use cap

 

[BRiT]

thats actually normal, in many games (fifa21, nba2k21, mortal kombat 11) they push graphics harder during cutscenes, but capping at 30 here they don't use cap

Ah, that kind of explains it. I'm not a sports focused gamer, open world racing doesn't count. The last ones I can remember really playing was "Tecmo Bowl" for NES and "One on One" for Atari 8Bit. I'm not counting my current tinkering with Madden '21.

 

[snc]

Ah, that kind of explains it. I'm not a sports focused gamer, open world racing doesn't count. The last ones I can remember really playing was "Tecmo Bowl" for NES and "One on One" for Atari 8Bit. I'm not counting my current tinkering with Madden '21.

I had long break from sports game but bought fifa21 and nba2k21 because of nextgen versions

 

[mr magoo]

IMO Rich analyse is right on point with studio who shipped first multiplat game performance is somewhat expected to not be top notch on X platforms (as he pointed out its clearly visible in xox and ps4pro versions). As a non american dood i couldn't care less for baseball but apparently people are bonkers.

 

[Deleted member 86764]

4k on ps5 and xsx, better performance on ps5
xss 1080p
xox and ps4pro 1440p, better performance on xox
loading times ps5 8.4s, xsx 11.8s, ps4pro 20.5s

Xbox One X outperforming Xbox Series S in the 1080p mode is surprising.

 

[snc]

Xbox One X outperforming Xbpx Series S in the 1080p mode is surprising.

probably game is not cpu demanding but rather gpu

 

[Jay]

probably game is not cpu demanding but rather gpu

Also as mentioned very much a cross gen game.
So raw gpu to gpu 1X is slightly stronger than XSS.
But if it was using current gen features VRS, mesh shaders, XVA then the XSS would shine.

This is the case for all current gen platforms.

 

[Deleted member 86764]

Also as mentioned very much a cross gen game.
So raw gpu to gpu 1X is slightly stronger than XSS.
But if it was using current gen features VRS, mesh shaders, XVA then the XSS would shine.

This is the case for all current gen platforms.

Yeah, I suspect you're right. Also with the developer being PlayStation familiar doesn't help the Xbox side.

Bit of a weird one, but actually nice to see a PlayStation developed game release on Xbox. I've no issues with seeing more of that.

Edit: and Spencer's tweet was kinda cool.

 

[Globalisateur]

Xbox One X outperforming Xbox Series S in the 1080p mode is surprising.

In a game not limited by CPU, this should be the normal outcome. 6TF Polaris > 4TF RDNA (and XSS has no Infinite cache to compensate for the quite reduced bandwidth compared to XBX so this is why we can't use the RDNA2 pure level of performance IMO).

 

[BRiT]

Come on folks, console wars is unbecoming for technical sites.

 

[Panino Manino]

In a game not limited by CPU, this should be the normal outcome. 6TF Polaris > 4TF RDNA (and XSS has no Infinite cache to compensate for the quite reduced bandwidth compared to XBX so this is why we can't use the RDNA2 pure level of performance IMO).

I don't understand how the Series S can perform worse than the P4Pro in any game.
The GPU have around the same teraflops, but it's RDNA2 vs CCN3, plus a much stronger CPU, more bandwidth... so how can the S perform worse than a PS4P? Makes no sense for me. It's just as mysterious as the Series X not showing it's GPU advantage.

 

[jlippo]

I don't understand how the Series S can perform worse than the P4Pro in any game.
The GPU have around the same teraflops, but it's RDNA2 vs CCN3, plus a much stronger CPU, more bandwidth... so how can the S perform worse than a PS4P? Makes no sense for me. It's just as mysterious as the Series X not showing it's GPU advantage.

APIs are different, perhaps some things done on GNM may have to be done differently on DX12.

Developers also had lot more time to work with the other.

 

[Panino Manino]

APIs are different, perhaps some things done on GNM may have to be done differently on DX12.

Developers also had lot more time to work with the other.

I understand that, but don't they have enough to brute force to at least equal performance? Are the Microsoft APIs so much worse? It's hard to believe.

 

[BRiT]

It's hard to believe.

It's easier to believe the developers simply need more time with 3/4 entirely new consoles and tools. They haven't had much time at all to learn the ins and outs of each individual console -- OneS, OneX, and Series S and X. They'll do better next year.