Topic: General Xbox Series X|S Thread

Expand

@Insightful Pretty much all games are running in a B/C mode - not taking advantage of many (if any) new technological improvements to the hardware - just relying on the 'brute force' of these consoles as opposed to being fully tuned and optimised utilising all (or at least most) of the features.

The vast majority of games right now are running the Series X like a XB1X+ and the PS5 like a PS4 Pro+ - just using the raw processing power these consoles have. Its not surprising as these games are built with the older gen in mind. Things like the decompression of assets may well still be done by the CPU because that's what the older gen consoles have to use and because the CPU is much faster and more efficient.

Hitman 3 doesn't look 'way' better on Series X. It has a small resolution advantage with a slight boost in Shadow Quality that you only notice in a Side by Side. Even the difference between the Series S, with its 'lower' texture quality as well doesn't look 'way' worse - especially not on a HD TV without the added 'upscaling' blur a 4k TV adds. The difference between most PS5 and Series X versions is less than the difference between XB1X and PS4 Pro and less than most of the XB1 and PS4.

Back Compat games are different again because for a lot of games, they were optimised for XB1X/PS4 Pro. The settings they used for the older hardware seems to be the limiting factor so we see a bigger difference.

On paper, the RAW specs of the PS5 is closer to the Series X than the Pro was to the X or the XB1 was to the PS4. Of course there is a LOT more to a system than the RAW specs - how all the components synergise with each other as well as how 'easy' it is for the Devs to optimise their games for the hardware, use all the features and schedule tasks to be most efficient.

Of course I expect the Series X to have 'an' advantage more often than not in numerous metrics - like resolution or visual settings. Performance, if the devs do their Job right, should be very similar as both systems will target the same frame rate so if the devs do their job right, the performance 'should' be the same. On paper, the Series X has quite a lot more processing power so could put that to use on maintaining a slightly higher resolution or other visual settings.

Of course, that's on 'paper'. What we don't yet know is how all those 'parts' of the system synergise, how easy it is for devs to utilise the hardware to maximise its potential etc. We don't yet know how the Geometry Engine compares to the Velocity Architecture for example or whether 'different' approaches to design (smaller but faster GPU vs larger but slower GPU) will make a 'big' difference - at least in 'some' if not all games. We don't know if Sony/MS have other 'built-in' tricks to offset perceived advantages either - for example, Sony went 'big' on Data throughput with their SSD speed, MS's approach was to look at the Data and be much more efficient on what Data is actually needed (Sampler Feedback) so only stream that Data.

Both have Pro's and Con's of course but right now, we cannot say categorically that Sony's decision to go with more Data bandwidth is a 'better' choice than MS's - after all, that Bandwidth is always there where as MS's solution relies on the system/devs to be much more efficient on what Data is needed. For example, an area may have 10GB of Assets in it and on Sony's system, that 'could' be transferred in under 2secs compared with over 4secs on Xbox. However, in that area, 5GB of textures etc are not visible - either because they are obscured by other objects or only part of the texture file is being used - so Xbox only streams the 5GB of data and thus negates Sony's advantage - but only if the Devs utilise that feature. That's just one example of how a 'different' approach to the same issue of transferring Data quickly enough could end up giving a very similar 'end user' experience. People expect Sony's system to load faster, to stream assets in quicker etc (minimise texture pop-in) because it has a much faster SSD but the difference could be 'minimal' if Sampler Feedback is utilised fully.

As someone with a PS4 Pro and XB1X (as well as PS5 and Series X, PS4 and XB1), I am also less bothered about higher actual pixel counts than I was back in 2013. Over the course of the generation, we have seen more methods of improving image quality above the actual running resolution of the game. DLSS2.0 is simply incredible - how it can look like 4k from just 1080p. Also Chequerboard or temporal reconstruction can look 4k too. I'd take a 1080p or 1920x2160p with DLSS2.0 or CB reconstruction over a Native 1800p - even Native 4k in a lot of cases - especially if that enables higher frame rates and/or higher visual settings. Even if the Series X has a 'higher' average pixel count, doesn't mean it will necessarily look better either - a CB4k may well have a lower average pixel count per frame than say 1620p but looks better on a 4k TV assuming all other settings are the same. Point is, some 'advantages' can be offset, or even bettered. I'd rather have a DLSS 1440/60 with all the visuals turned up to max than a native 4k, unstable 60 and some visual settings turned down to try and push the frame rate up a bit more. Technically, the Native is pushing out more than double the pixel count every frame and therefore would appear to have the advantage but overall, the visual quality of the DLSS 1440p would exceed that.

Anyway, I look forward to the upcoming years as it will be fascinating to see what devs create with these new consoles and technology in mind. It will be interesting to see how the Hardware holds up to different game designs and whether or not, the game design is built around a specific feature - like Data throughput - how they compare on both systems. Its entirely possible that some games could be better on PS5 or have some advantages that are 'more' important to you as a gamer than just resolution...