• 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: Assassin's Creed Mirage - DF Tech Review - PS5 & Xbox Series X/S Tested at 30FPS/60FPS

Fake

Member
I didn't hear any mention from DF of the apparently excessive use of Chromatic Aberration. Anyone know what that's like on console?

Is really dark times we live that we are in need for sites like Digital Foundry to address things that fans can report anytime. If those sites don't mention, they think this is not happening.
Consoles have tools to screenshot and capture videos, just for in the end nothing of this matter to the devs until youtubers/sites point out the bug for them.

I like having a fight with the NightDive Studios dev because if nobody reported the bug I mention, they think is a problem with my console and asked me to call Sony support. He even suggest I buy a preview build from Playstation Store with is absurd.
 

Vergil1992

Member
DF conspires against PlayStation again. They are hiding from us that it is better on PS5.

Please, it is clear that it is an identical game, that the DRS works a little faster on PS5 and prevents a frame drop more often is somewhat irrelevant, even Elden Rings which has a performance advantage on PS5 has a problem with the DRS that sometimes makes its framerate worse than in XSX. It's not even that it happens on XSX and not on PS5, it happens with some frequency on both, but more en XSX.

This is an identical game. There's little to say here to debate performance.
Not even NX Gamer/IGN, which usually makes differences where there are none, say that there are no differences. 2% in favor of PS5 in cutscenes or shot changes, and it is probably because it does not lower the resolution as quickly as PS5.

It's something anecdotal. In fact, the one who could shed light on this is VG Tech, because DF and NXGamer often get it wrong with native resolutions, especially if they are dynamic. It's a great job for both, and considering that AC: Valhalla had clearly better performance on PS5, its optimization on XSX has greatly improved.
 

King Dazzar

Member
Is really dark times we live that we are in need for sites like Digital Foundry to address things that fans can report anytime. If those sites don't mention, they think this is not happening.
Consoles have tools to screenshot and capture videos, just for in the end nothing of this matter to the devs until youtubers/sites point out the bug for them.

I like having a fight with the NightDive Studios dev because if nobody reported the bug I mention, they think is a problem with my console and asked me to call Sony support. He even suggest I buy a preview build from Playstation Store with is absurd.
Been there and done that with AC Valhalla and Ubisoft with regards the adaptive triggers and haptics bug - Ubisoft did diddly squat about it. There was even a iirc 15 page plus thread on their forums about the bug too. Lets hope they fix the CA - which I'm optimistic they will.
 

Darsxx82

Member
PS5 lead platform is probably why, this engine is a last generation engine that's not using any next-gen graphics technology.
PS5 main platform and the XSX version extremely or well.... no, more than extremely on par with that. Add to that a very well finished XSS version and you already have how a 3rd party Studio that seeks to balance development cost vs possible income works in a logical and general way.

You can be clear that if XSX (being the platform with the smallest user base) were not so capable, you would not see such equality in the comparisons with PS5 but quite the opposite.
 

adamsapple

Or is it just one of Phil's balls in my throat?
It seems to be going a little better on PS5 and without tearing.

All of them tear, PS5 too.

Both PS5 and Series X have issues with this, though I've noticed tearing more frequently on Series X

1_yoSbSvW.png


2_FVzK4Sf.png
 
Last edited:

SKYF@ll

Member
Even though the PS5 version is better, the differences are minimal.
The 3rd year AC is comparable in terms of performance.
Great optimization and everyone wins.
 
There's a noticeable uplift in geometry and textures based on my playthrough, and I've slinked 200 + hours into Odyssey.
Interesting you have a different opinion to the poster below. Still would you agree its not looking as improved as it should so many years later?
 
Interesting you have a different opinion to the poster below. Still would you agree its not looking as improved as it should so many years later?

I'll share some screen shots later today to show you guys what I'm talking about.

I think the games visuals could do with some refinement, most notably the removal of the awful chromatic aberration.

It's not perfect, the facial animations and facial detail could do with some work but I think the environments and details are pretty solid for a cross-gen game.
 
I'll share some screen shots later today to show you guys what I'm talking about.

I think the games visuals could do with some refinement, most notably the removal of the awful chromatic aberration.

It's not perfect, the facial animations and facial detail could do with some work but I think the environments and details are pretty solid for a cross-gen game.
I'm very interested to see those, thanks.
 

adamsapple

Or is it just one of Phil's balls in my throat?
Is this much better graphically than Assassin's Creed Oddessy/Origins? Certainly doesn't look it to me.

I would say its a mixed bag. Smaller scope means more detail crammed but generally the tech hasn't evolved much beyond Odyssey or Valhalla, considering this was in development as a DLC for Valhalla, they weren't going to push much in terms of new rendering techniques anyway.
 

Ev1L AuRoN

Member
Usually tearing is more of a GPU thing but I could be wrong. I know the two GPUs are different so maybe that's why the engine doesn't tear as much on the PS5 for this title.

But overall I consider both versions identical if that makes any sense. That tiny difference doesn't make one vastly superior than the other.
I think that, and that is my theory, that the vsync situation has to do with controller response.

Instead of using triple buffer to ensure no screen tearing, Microsoft games tend to prefer adaptive sync, that introduces tearing when the frame drops, but the benefit is that the latency is greatly reduced, since the game is 99% lock, they prefer to introduce some tearing on the rare occasions the frame drop, with the benefit that for the 99% of the time the game is more responsive.
 
I think that, and that is my theory, that the vsync situation has to do with controller response.

Instead of using triple buffer to ensure no screen tearing, Microsoft games tend to prefer adaptive sync, that introduces tearing when the frame drops, but the benefit is that the latency is greatly reduced, since the game is 99% lock, they prefer to introduce some tearing on the rare occasions the frame drop, with the benefit that for the 99% of the time the game is more responsive.

Well this isn't a Microsoft game.

Elder Legend Elder Legend
 
Last edited:

Ev1L AuRoN

Member
Well this isn't a Microsoft game.
But it uses Microsoft SDK, don't quote me on that since I'm not 100% sure, I remember reading about Xbox having more latency than the PS5, maybe is the hypervisor introducing some latency or even the controller, that's the only reason I can come up with to explain why games on Xbox (and that's even on the 360 era) usually has more tearing than PlayStation games. This has nothing to do with performance, it's a choice the devs make disproportionately more on Xbox.
 

Lysandros

Member
I think that, and that is my theory, that the vsync situation has to do with controller response.

Instead of using triple buffer to ensure no screen tearing, Microsoft games tend to prefer adaptive sync, that introduces tearing when the frame drops, but the benefit is that the latency is greatly reduced, since the game is 99% lock, they prefer to introduce some tearing on the rare occasions the frame drop, with the benefit that for the 99% of the time the game is more responsive.
So you are saying this is a Microsoft policy thing and not a deliberate developer choice to mitigate performance loss on XSX in multiplatform titles? To reiterate, as per this analysis both versions use adaptive sync, XSX simply tears more indicating a performance differencial. By the way, is there any evidence PS5 offering higher input latency compared to XSX across the games as a general trend?
 
Last edited:

Ev1L AuRoN

Member
So you are saying this is a Microsoft policy thing and not a deliberate developer choice to mitigate performance loss on XSX in multiplatform titles? To reiterate, as per this analysis both versions use adaptive sync, XSX simply tears more indicating a performance differencial.
I don't think it's a policy, just the studio trying to make things as equal as possible, even latency. Again, that's just my theory, that can as well not be the case. But if not for latency, why not turn on triple-buffer vsync?
 
Last edited:
Well this isn't a Microsoft game.

Elder Legend Elder Legend

There has been a long standing difference in the general vsync choice between Xbox and PS for years (at least since the start of the X1/PS4 generation). PS4 often used a buffered vsync (no tearing, but often quite noticeable vsync stutter) where X1 would use adaptive sync and get noticeable tearing but no vsync stutter.

Here no one is using a buffered vsync, I assume because the performance was so solid anyway.
 

Ev1L AuRoN

Member
Performance cost of full Vsync.
That would indicate that on PS5 devs have no problem dropping frames to ensure consistency with vsync engaged, whereas on Xbox the same developers prefer to introduce tearing while trying to maintain responsiveness even when the frames drop.

That don't make sense to me, AFAIK triple-buffer don't impact performance at all, it just a limiter, of course you can have bad implemented vsync, also it can drop single frames, different from double-buffer that tend of half the fps when the frame drops, that's why adaptive vsync was created, to produce more responsive games without the tradeoffs of the double-buffer vsync.
It must have some kind of payoff to go either route on each platform, I'm just speculation what it can be.
 

adamsapple

Or is it just one of Phil's balls in my throat?
That would indicate that on PS5 devs have no problem dropping frames to ensure consistency with vsync engaged, whereas on Xbox the same developers prefer to introduce tearing while trying to maintain responsiveness even when the frames drop.

That don't make sense to me, AFAIK triple-buffer don't impact performance at all, it just a limiter, of course you can have bad implemented vsync, also it can drop single frames, different from double-buffer that tend of half the fps when the frame drops, that's why adaptive vsync was created, to produce more responsive games without the tradeoffs of the double-buffer vsync.
It must have some kind of payoff to go either route on each platform, I'm just speculation what it can be.

I think we're all overthinking it, more than likely it's just a leftover engine niggle, they're using a nearly decade old engine with modifications made to it over time.

Multiple videos have shown that both versions perform within 99.8% of each other with only a couple of extra tears and a one or two extra frame drops on SX. Hardly a bother or any kind of meaningful performance cost.
 

Ev1L AuRoN

Member
I think we're all overthinking it, more than likely it's just a leftover engine niggle, they're using a nearly decade old engine with modifications made to it over time.

Multiple videos have shown that both versions perform within 99.8% of each other with only a couple of extra tears and a one or two extra frame drops on SX. Hardly a bother or any kind of meaningful performance cost.
You're right, of course, I was just extrapolating on why a lot of Xbox games tend to go that route vs PlayStation. Without a developer perspective, we might never know for sure. Sometimes optimizations made for the Series S sneak into the XSX code and made it worst that it needs to be, that could be the case here as well.
 

Vergil1992

Member


XSX have highter resolution (average).

DF not only does not conspire against PlayStation, but even favors it. Although I don't believe in the conspiracy theory, I think VG Tech simply counts pixels better and when it comes to dynamic resolutions it's really difficult. But again this is a case where XSX generally has a higher average resolution.
 
Last edited:

adamsapple

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


XSX have highter resolution (average).

DF not only does not conspire against PlayStation, but even favors it. Although I don't believe in the conspiracy theory, I think VG Tech simply counts pixels better and when it comes to dynamic resolutions it's really difficult. But again this is a case where XSX generally has a higher average resolution.



0.5% difference in performance, NXG gave it a roughly 2%.

Higher DRS resolution and better fire effects too.

They've certainly made a huge turnaround on this engine on Xbox compared to Valhalla.

Bodes really well for AC: Red next year.
 

Vergil1992

Member
0.5% difference in performance, NXG gave it a roughly 2%.

Higher DRS resolution and better fire effects too.

They've certainly made a huge turnaround on this engine on Xbox compared to Valhalla.

Bodes really well for AC: Red next year.
Yes, the difference really is anecdotal. Considering that their previous work with the same engine had a relatively large advantage of PS5... here the improvement in XSX has been very considerable. Additionally, it is much more stable than Valhalla on both systems.


VG Tech is the one that makes the least comparisons, but it seems to me by far the most reliable. They almost always have DF or NX Gamer retract or "update" their pixel count to match yours.


This is also useful for what I said in another thread. When dynamic resolutions and temporal reconstruction are involved, very thorough pixel counting is needed to know if a platform has a resolution advantage.
 

SKYF@ll

Member
DF "The PS5 version has less tearing"
IGN, GamingTech and Open Surprise “PS5 version frame rate is stable”
VGTech “The DRS resolution of the XSX version is higher”
(Total Torn Frames) PS5:10,XSX:130 Both almost 60fps lock

It is natural that the results will be slightly different because the scenes analyzed are different.
The differences were very small and both versions had good optimizations.
 
I'm very interested to see those, thanks.

Here are some screen shots. (Under spoiler tags because I didn't want to take up the entire screen with captures, there's no actual spoilers in the images).

AC Mirage :
SnPcmzA.jpg


AC Odyssey :

Btd47Ml.jpg

AC Mirage :

dP3das3.jpg


AC Odyssey :

M2IcqJG.jpg

AC Mirage :

rC1JPRr.jpg


AC Odyssey:
aIT1wTb.jpg

Both were captured on my PS5, Mirage in Performance Mode.

Overall I think Mirage looks better in certain aspects, the environment seems more dense and objects look more detailed. Lighting and shading also seems to look better as well. That being said I do think AC Odyssey looks better from an artistic point of view but that's purely subjective on my part, Odyssey is one of my favourite games, it also had a significantly better HDR implementation as well and one of the best I've seen to date.
 
Last edited:
Here are some screen shots. (Under spoiler tags because I didn't want to take up the entire screen with captures, there's no actual spoilers in the images).

AC Mirage :
SnPcmzA.jpg


AC Odyssey :

Btd47Ml.jpg

AC Mirage :

dP3das3.jpg


AC Odyssey :

M2IcqJG.jpg

AC Mirage :

rC1JPRr.jpg


AC Odyssey:
aIT1wTb.jpg

Both were captured on my PS5, Mirage in Performance Mode.

Overall I think Mirage looks better in certain aspects, the environment seems more dense and objects look more detailed. Lighting and shading also seems to look better as well. That being said I do think AC Odyssey looks better from an artistic point of view but that's purely subjective on my part, Odyssey is one of my favourite games, it also had a significantly better HDR implementation as well and one of the best I've seen to date.
The lighting definitely looks better in Mirage, other than that I'm not seeing a great difference if I'm honest. Thank you for taking the time, probably grab Mirage early next year.
 


XSX have highter resolution (average).

DF not only does not conspire against PlayStation, but even favors it. Although I don't believe in the conspiracy theory, I think VG Tech simply counts pixels better and when it comes to dynamic resolutions it's really difficult. But again this is a case where XSX generally has a higher average resolution.


They actually said some scenes and the difference is pretty small between the two. That's performance mode. Quality mode the two are the same.
 
Last edited:
Top Bottom