• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Digital Foundry: Alan Wake 2 Xbox Tech Review - Excellent On Series X, But What About Series S?

mejin

Member
This is a thread example of why I don't really fuck wit the green rats

When it performs better on ps5, they all pop in talking bout it's some BS like tools, marketing and shit.

When xsx does better - best believe it's time to retire the playstation, all game from now on - future technology - will run better on XSX and it's alright, just accept it, xsx the goat, ps weaker like we was saying back in 2020....

I say shite

PS5 is consistently beating X versions. Let them have their fun.
 

adamsapple

Or is it just one of Phil's balls in my throat?
I wonder what performance difference is produces and if it is even worth using, most people don't see this at all as they play with DLSS.


The team saw similarly large perf gains from VRS Tier 2 – up to 14%! – this time with no noticeable visual impact. See for yourself if you can tell which side of the first image in the blog has VRS enabled to get a perf boost, and which side doesn’t.


Xbox version Cyberpunk had noticeable image deficiency with VRS and run worse than PS5 version.

Again, bit of an exaggeration. Cyberpunk has supported VRS on Xbox since 2021 and no one noticed it outside of the one 400% zoom Oliver pointed out in the Phantom Liberty video.

After which the very next line he says is:

hqqo6cb.jpg





As for it still performing worse, well it's a performance gaining feature. If VRS was disabled, it would likely run a few frames worse. Not all engines universally play well with either console. This iteration of Red Engine runs better on PS5, but in all the new Phantom Liberty areas, they run identically otherwise.
 
Last edited:

Bojji

Gold Member






Again, bit of an exaggeration. Cyberpunk has supported VRS on Xbox since 2021 and no one noticed it outside of the one 400% zoom Oliver pointed out in the Phantom Liberty video.

After which the very next line he says is:

hqqo6cb.jpg





As for it still performing worse, well it's a performance gaining feature. If VRS was disabled, it would likely run a few frames worse. Not all engines universally play well with either console. This iteration of Red Engine runs better on PS5, but in all the new Phantom Liberty areas, they run identically otherwise.

I was talking about CP implementation specifically, I tested Gears myself and difference was there with more aggressive settings.

But in many games VRS produces zero or close to zero difference and you can see it in Starfield, Dead Space etc. That's why I doubt using VRS in CP2077 was worth it at all.
 

Bogroll

Likes moldy games
XSX is noticeably ahead on this one apparently, no question. PS5 version needs some additional work to stabilise the performance.
How come you don't have this view when Xbox is not performing well (off the top of my head Atomic Heart) Blaming split memory etc, now don't get me wrong XSX would be better without split memory speeds etc. But Atomic Heart as been patched a few times and runs great now on Xbox. You know at lot more than me about computers but in reality like most wannabe experts on here. You know fuck all.
 

JackMcGunns

Member
How come you don't have this view when Xbox is not performing well (off the top of my head Atomic Heart) Blaming split memory etc, now don't get me wrong XSX would be better without split memory speeds etc. But Atomic Heart as been patched a few times and runs great now on Xbox. You know at lot more than me about computers but in reality like most wannabe experts on here. You know fuck all.


I haven't seen his views, but I've seen many head to heads and seen people making a big deal and filling the thread with ANOTHER ONE.gif when a PS5 version of a game is locked at 60fps and the SX is occasionally dropping to 59, 58fps... THAT right there shouldn't be a big deal, but no one can say anything because it's immediately pointed out that SX is the more powerful machine and shouldn't be the one dropping the frames, and I agree with that by the way, but in this comparison I'm seeing the PS5 version drop frames into the low 50s while the SX is hitting 60 pretty consistantly. Also he only said that the PS5 needs additional work to stabilise the performance, he's not even saying they can't get there due to the hardware difference. Sheesh!
 

Mister Wolf

Gold Member
How come you don't have this view when Xbox is not performing well (off the top of my head Atomic Heart) Blaming split memory etc, now don't get me wrong XSX would be better without split memory speeds etc. But Atomic Heart as been patched a few times and runs great now on Xbox. You know at lot more than me about computers but in reality like most wannabe experts on here. You know fuck all.

Because we don't give a shit about Atomic Heart. People give a shit about this game because it's the most technically impressive game released this generation. It's like bragging about winning your local Golden Gloves tournament then when you get to the Olympics you lose in the first round.
 

Bogroll

Likes moldy games
I haven't seen his views, but I've seen many head to heads and seen people making a big deal and filling the thread with ANOTHER ONE.gif when a PS5 version of a game is locked at 60fps and the SX is occasionally dropping to 59, 58fps... THAT right there shouldn't be a big deal, but no one can say anything because it's immediately pointed out that SX is the more powerful machine and shouldn't be the one dropping the frames, and I agree with that by the way, but in this comparison I'm seeing the PS5 version drop frames into the low 50s while the SX is hitting 60 pretty consistantly. Also he only said that the PS5 needs additional work to stabilise the performance, he's not even saying they can't get there due to the hardware difference. Sheesh!
Yeah there's a few threads that poster pops up in and another poster Onq123 or something like that churning out the same stuff when Xbox is worse. They have good points but it's never the reason what they said about PS5 AW ie it needs a few patches which I believe the PS5 AW will get a up to the performance of the XSX version.
 

Bogroll

Likes moldy games
Because we don't give a shit about Atomic Heart. People give a shit about this game because it's the most technically impressive game released this generation. It's like bragging about winning your local Golden Gloves tournament then when you get to the Olympics you lose in the first round.
Shit analogy and I wasn't talking about you.

Edit I apologize I didn't read what you said probably. I get what you are saying now.
 
Last edited:

JackMcGunns

Member
I was talking about CP implementation specifically, I tested Gears myself and difference was there with more aggressive settings.

But in many games VRS produces zero or close to zero difference and you can see it in Starfield, Dead Space etc. That's why I doubt using VRS in CP2077 was worth it at all.


"VRS Tier 2 offers a 14% GPU performance gain in Gears 5 with no perceptible difference in image quality"

No one ever complained about Gears 5 visuals on Series X, so a 14% GPU performance gain is pretty damn good, so why not use it? I mean unless you're doing it wrong, but the wrong implementation is not the fault of the feature.
 

Darsxx82

Member
This is reality and I know it. But what to expect, Xbox as a whole is outsold by PS5 in 2:1 ratio and then you have Xbox Series S that is dominating Xbox sales. I suspect developers pay more attention to Series S version sometimes and Series X "just works". That's the reality, I also doubt it produces any big differences, if game is unoptimized it is unoptimized in core code and both consoles outside of Apis are essentialy the same hardware.

Between consoles so par in power, you're rarely going to see significant differences if there's no reason beyond the hardware. But they can happen when the problem stems from the software and the optimization time. And that's been seen on several occasions where the XSX version was blatantly left on the back burner while the PS5 version was favored. That's why it's not an "excuse" and it does have relevance. Otherwise would be to accept that cases like Calisto Protocol were the result of advantages in the PS5 hardware and limitations of XSX (as many here tried to defend) while it was proven that it had nothing to do with it.

Being a base platform for development and being the platform that STudios prioritize has great benefits. It is naïve to believe that if XSX were the base development platform and also the priority platform it would have no effect on the results of the comparisons vs PS5.
I wonder what performance difference is produces and if it is even worth using, most people don't see this at all as they play with DLSS. Xbox version Cyberpunk had noticeable image deficiency with VRS and run worse than PS5 version.
LOL. what a hyperbole. It's funny because 99.9% of people who say such a thing are then unable to detect whether VRS is being used or not and only speak based on 400% captures of specific elements and areas.

VRS"degradation" is for specific examples with zoom and concrete elements, no one appreciates its use, which is also only implemented in certain situations. Not to mention, on the other hand, you gain some overall sharpness improvement. The performance of both versions is the same in the new area DLC. That is already relevant. Apart from that, I would like to know if the latest patch has improved the performance in XSX in the old zone as the description of the patch indicates.

VRS has as many good examples of great implementation as it does examples of poor implementation. Cataloguing and valuing only bad cases is not valuing fairly.
 
I was talking about CP implementation specifically, I tested Gears myself and difference was there with more aggressive settings.

But in many games VRS produces zero or close to zero difference and you can see it in Starfield, Dead Space etc. That's why I doubt using VRS in CP2077 was worth it at all.
Try Doom Eternal, in the console space it produces a significant advantage for Xbox (only consoles that has hardware based VRS).
 

Radical_3d

Member
I don't think there will be pro consoles, just because game releases that fully support the current non pro consoles have been very slow coming.
I dont think a pro console sits right at the moment, as much as I'd like to see them personally.
You think the guy who predicted the PSPortal and the new PS5 Slim months beforehand is wrong just in that bullet point?
 

DanielG165

Member
hopefully we see this same energy when a Series X version of a game is doing worse.
We already do? Hasn’t that been the case already for whenever a SX version that does slightly worse than the PS5 version, that it needs more time and additional updates? Here, the SX is simply performing better at the moment, and the S is performing well also. The Xbox consoles, particularly the SX, have a great port that’s leveraging their feature set; that should be something people interested in video games tech should be celebrating.
 
Last edited:

Bojji

Gold Member
"VRS Tier 2 offers a 14% GPU performance gain in Gears 5 with no perceptible difference in image quality"

No one ever complained about Gears 5 visuals on Series X, so a 14% GPU performance gain is pretty damn good, so why not use it? I mean unless you're doing it wrong, but the wrong implementation is not the fault of the feature.

Gears is exception to the rule, no other game is close to having this good implementation of VRS. Plus to have that much performance gain you have to use "performance" setting that IS NOTICEABLE.

Between consoles so par in power, you're rarely going to see significant differences if there's no reason beyond the hardware. But they can happen when the problem stems from the software and the optimization time. And that's been seen on several occasions where the XSX version was blatantly left on the back burner while the PS5 version was favored. That's why it's not an "excuse" and it does have relevance. Otherwise would be to accept that cases like Calisto Protocol were the result of advantages in the PS5 hardware and limitations of XSX (as many here tried to defend) while it was proven that it had nothing to do with it.

Being a base platform for development and being the platform that STudios prioritize has great benefits. It is naïve to believe that if XSX were the base development platform and also the priority platform it would have no effect on the results of the comparisons vs PS5.

LOL. what a hyperbole. It's funny because 99.9% of people who say such a thing are then unable to detect whether VRS is being used or not and only speak based on 400% captures of specific elements and areas.

VRS"degradation" is for specific examples with zoom and concrete elements, no one appreciates its use, which is also only implemented in certain situations. Not to mention, on the other hand, you gain some overall sharpness improvement. The performance of both versions is the same in the new area DLC. That is already relevant. Apart from that, I would like to know if the latest patch has improved the performance in XSX in the old zone as the description of the patch indicates.

VRS has as many good examples of great implementation as it does examples of poor implementation. Cataloguing and valuing only bad cases is not valuing fairly.

You don't need 400% zoom to see deficiencies, you just don't have to be blind and use 4K screen. There are many games using VRS and in most of them performance difference is almost non existent while you can see IQ degradation. So far it's a failure tech.

Try Doom Eternal, in the console space it produces a significant advantage for Xbox (only consoles that has hardware based VRS).

It has higher resolution overall but actually worse image quality in some places.
 
Last edited:

Vergil1992

Member
PS5 is consistently beating X versions. Let them have their fun.
It isn't true. But hey, it's really hard to "quantify" this. What I will add is that the most GPU-demanding games often tend to have slightly better versions in XSX, like this Alan Wake 2, A Plague Tale: Requiem, Dead Space (Remake), Dying Light 2, The Witcher 3 Ray Tracing mode...



____________________________________________________________________________________________________



I don't think it's honest to say that "the version of Cyberpunk is better on PS5", if nothing has changed compared to the previous version (looking at DF's analysis it seems not, the minimum resolution is 1008p on PS5), XSX has always had higher resolutions and has also always used VRS.

VRS is not something you can notice when playing, at least if it is implemented correctly. The fact is that Cyberpunk had been using VRS for a long time and no one had noticed until Oliver took a screenshot finding an exact moment where it was in operation.

PS5 has slightly better performance than XSX in the base game, in the most current DLC the performance is identical, XSX has higher resolution and uses VRS. It's more of a tie.
 

shamoomoo

Member
That on many occasions the Studios prioritize the PS5 version and that it is also usually the base development platform is not a excuse, is a reality. On such evenly matched hardware, this is always going to have an effect

Especially when you see how in many cases how performance improves in a matter of less than 1 month or weeks on Xbox with patches that are more and more frequently arriving later on Xbox compared to PS5.

And the optimization time dedicated to each platform and what is the basiedevelopment. In hardware (XSX and PS5) that is so on par and also sharing technology, this is always key. At least for the results on launch day because we've already seen how these change with successive patches. For example, RE4 Remake was quite significant.
Lol? What about" Ghostwire Tokyo?" It came out a year later and was more unstable than the PS5,that wasn't the case with "The Medium."
 
You think the guy who predicted the PSPortal and the new PS5 Slim months beforehand is wrong just in that bullet point?
Yes I think the rumours wont come to anything, maybe the rumours came from internal testing which i guess is always happening.
Happy to be wrong in this case as I would be in on getting pro consoles, but I just think the higher ups at Sony/MS must be thinking the current base machines haven't been utilised yet so releasing a new model probably also wont be for a good few years maybe, next gen will arrives before pro consoles can be exploited. I think all pro consoles might do is just brute force higher resolutions of current gen games.
With games being in development for so long now, I wonder if generations will cease to be a thing and more powerful hardware will come along, as and when its ready and the software on the devices will make use of extra grunt for higher res or fps as developers decide.
 

Darsxx82

Member
Gears is exception to the rule, no other game is close to having this good implementation of VRS. Plus to have that much performance gain you have to use "performance" setting that IS NOTICEABLE.
No, it is not the exception, it is more that you are guided by the captures with zoom at 400% with VRS vs. non-VRS and you stay with that impression.

You don't need 400% zoom to see deficiencies, you just don't have to be blind and use 4K screen. There are many games using VRS and in most of them performance difference is almost non existent while you can see IQ degradation. So far it's a failure tech.
In the mentioned games, you do need comparative screenshots and zoom to identify their implementation. The fact that even VGtech has difficulty ensuring 100% whether a game uses VRS or not tells you everything. As if to come and say that anyone with two eyes can detect it...
It has higher resolution overall but actually worse image quality in some places.

The important thing is the whole and the improvement in resolution is vast compared to the degradation of detail in specific objects in darkened areas or outside the player's main focus. In fact that is the description and purpose of VRS technique.
 

Radical_3d

Member
Yes I think the rumours wont come to anything, maybe the rumours came from internal testing which i guess is always happening.
Happy to be wrong in this case as I would be in on getting pro consoles, but I just think the higher ups at Sony/MS must be thinking the current base machines haven't been utilised yet so releasing a new model probably also wont be for a good few years maybe, next gen will arrives before pro consoles can be exploited.
Enterprise executives give no fucks to which extend a hardware is utilised. They care about sales, market share, momentum and all those sort of stuff that is for what a Pro console is. The leaked plans for MS to wait for 2028 is more of an incentive to green light the project. You know how companies hate competition.
I think all pro consoles might do is just brute force higher resolutions of current gen games.
And I ask for nothing more. Even if I get the XSX version now the image is not pleasant. As another user pointed we’re used to cleaner images now. Alan Wake II with sightly better internal resolution would be just perfect. You can even keep the RT features. Just cleaner image. And I’m willing to gambit waiting one year to get that. And a physical release ffs.

With games being in development for so long now, I wonder if generations will cease to be a thing and more powerful hardware will come along, as and when its ready and the software on the devices will make use of extra grunt for higher res or fps as developers decide.
Middleware and tools must step up to reduce development times but that’s another discussion.
 

Darsxx82

Member
Lol? What about" Ghostwire Tokyo?" It came out a year later and was more unstable than the PS5,that wasn't the case with "The Medium."
You should educate yourself better on this topic. You have been proposing example after example of games showing that you are not up to date.

GWTokio received more updates weeks after launching on XSX announcing substantial performance and bug improvements which says a lot about the state in which the port to XSX was released




The fact is that there has been no subsequent comparison to know to what extent it improved vs PS5 but I can assure you that it has improved.

That said, you are giving the example of a port of a game created based on PS5, done in a hurry and whose sales prospects were zero. You can also use Deathloop as an example and you have the opposite result.

Finally Medium.... It's shocking to see people even using it as an example of something without mentioning the whole thing. Medium wasn't better on PS5, it was rather "different." The XSX version uses RT reflections +AO compared to PS5 which prioritized resolution and reduced the use of RT to reflections to a lesser extent.
I mean, I don't know where Medium is an example of what you're trying to argue.
 
Last edited:

Vergil1992

Member
Lol? What about" Ghostwire Tokyo?" It came out a year later and was more unstable than the PS5,that wasn't the case with "The Medium."
If what he says doesn't seem right to you, what is your theory? that PS5 is actually much more powerful and is capable of having a better RT implementation than XSX, higher resolution and better framerate?
The most similar case would be Deathloop and it works better in XSX. The point isn't just that it's the "favorite" version's fault, it's that the software is everything. GhostWire Tokyo is (or was, we currently don't know) much worse optimized on Xbox. And I say "was" because it was known that there were updates expressly to improve performance, but I have not seen comparisons of how it currently works compared to PS5.

The reasons? Well, there could be many. Maybe it was a quick port for GP with little investment, maybe Tango GameWorks is more efficient with the PS5 API...

But blaming the hardware here would be stupid. There is no way to justify such wide differences in a cross-platform game. PS5 on average runs 5-10fps better, has higher resolution, better shadows (in Xbox have bugs) and better Ray Tracing. This is completely impossible to be due to each one's hardware. It's a poor port, I don't think there's much to debate here.
 

poodaddy

Member
Is this Xbox play anywhere by chance? Meaning, if I bought it on Series would I then have the PC version or no?
 

Fafalada

Fafracer forever
SFS is still pure mystery.
It's not really a mystery - it's just that you need to move to a completely different data paradigm for it to meaningfully contribute anything.
The same data paradigm shift is also needed for SSDs to really be brought to bear fully - but that's predictably taking a long time to materialise because middleware engines can't make that dramatic a shift anytime soon, and proprietary engine-devs have been busy doing cross-get titles until this year, and 'gen exclusive' codebases won't shed all that legacy overnight.
 

adamsapple

Or is it just one of Phil's balls in my throat?

I know lol

but I guess to their greater point, this is a PC version that came in an ideal state, without any PC specific messes that have permeated almost all the major releases these days.

Dead Space on PC is *still* not fixed, for example.
 
And this is what confuses me. If this were true, then the 5700 XT wouldn't perform so poorly. Remedy said verbatim that it's an unsupported GPU but if viewed from DX12, Primitive Shaders are designed to work as Mesh Shaders, why isn't it the case with RDNA1 cards?
RDNA1 and older GPU's can't use Direct X 12 Ultimate,without this they have no ability to use Mesh Shaders.All AMD GPU's from Vega to RDNA3 have the hardware to use Primitive Shaders,this includes the Series X/S,but they have to be able to use D12U in order to use Mesh Shaders.
 
Or maybe what that Mr Chang said was just wrong, and RDNA2 and 3 do have Mesh Shaders and not Primitive Shaders.
Mesh Shaders in hardware do not exist,they are a API function dependent on drivers and software,Primitive Shaders exist in hardware in all AMD GPU's from Vega to RNDA3,it's only thru DX12U that Primitive Shaders are converted into Mesh Shaders,RDNA1 and older GPU's can't use DX 12U.
 

winjer

Gold Member
Mesh Shaders in hardware do not exist,they are a API function dependent on drivers and software,Primitive Shaders exist in hardware in all AMD GPU's from Vega to RNDA3,it's only thru DX12U that Primitive Shaders are converted into Mesh Shaders,RDNA1 and older GPU's can't use DX 12U.

Really, then why doesn't it work on Pascal and RDNA1.
 

Gaiff

SBI’s Resident Gaslighter
Mesh Shaders in hardware do not exist,they are a API function dependent on drivers and software,Primitive Shaders exist in hardware in all AMD GPU's from Vega to RNDA3,it's only thru DX12U that Primitive Shaders are converted into Mesh Shaders,RDNA1 and older GPU's can't use DX 12U.
Are you sure?











There's minor HW differences, which can be seen in Vulkan discussions.

There's a whole Twitter thread with lots of comments and insight from developers that shed light on the confusion.
 
Last edited:
Top Bottom