• 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.

[Vg Tech] Call of Duty Modern Warfare 2 PS5 vs Xbox Series X|S Frame Rate Comparison

Lunatic_Gamer

Gold Member



Call of Duty Modern Warfare 2 PS5 vs Xbox Series X vs Xbox Series S frame rate comparison comparing the games framerate/fps in the campaign using the 60fps and 120fps modes.

Timestamps:
00:00 120fps Mode
09:49 60fps Mode

PS5 and Xbox Series X in the 60fps mode use a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 1920x2160. PS5 and Xbox Series X in the 60fps mode seem to often render at 3840x2160 and appear to rarely drop down to 1920x2160. On PS5 and Xbox Series X in the 60fps mode a form of temporal upsampling is used that can reconstruct a 3840x2160 resolution when rendering below this resolution.

Xbox Series S in the 60fps mode uses a dynamic resolution with the highest resolution found being 2560x1440 and the lowest resolution found being 1280x1440. On Xbox Series S in the 60fps mode a form of temporal upsampling is used that can reconstruct a 2560x1440 resolution when rendering below this resolution.

Xbox Series S in the 120fps mode uses a dynamic resolution with the highest resolution found being 1920x1080 and the lowest resolution found being 960x1080. Xbox Series S in the 120fps mode seems to often render at 1920x1080 and appears to rarely drop down to 960x1080. On Xbox Series S in the 120fps mode a form of temporal upsampling is used that can reconstruct a 1920x1080 resolution when rendering below this resolution.

PS5 and Xbox Series X in the 120fps mode render at a resolution of approximately 2730x1536.

PS5 and Xbox Series X render the UI at 3840x2160 in both modes.

All three consoles in all modes all appear to be using a form of Variable Rate Shading. Temporal upsampling is used that seems to reconstruct the parts of the frame with a reduced shading rate from VRS.

Cutscenes are letterboxed which results in a lower effective resolution during these scenes.

PS5 and Xbox Series X have improvements to draw distance compared to Xbox Series S.

PlatformsPS5Xbox Series XXbox Series S
Frame Amounts
Game Frames619496381441042
Video Frames705027050270502
Frame Tearing Statistics
Total Torn Frames054846116
Lowest Torn Line-237207
Frame Height144014401080
Frame Time Statistics
Mean Frame Time9.48ms9.21ms14.32ms
Median Frame Time8.33ms8.33ms16.67ms
Maximum Frame Time16.67ms58.33ms51.17ms
Minimum Frame Time8.33ms7.37ms8.33ms
95th Percentile Frame Time16.67ms16.37ms16.67ms
99th Percentile Frame Time16.67ms16.67ms16.67ms
Frame Rate Statistics
Mean Frame Rate105.44fps108.61fps69.86fps
Median Frame Rate105fps109fps67fps
Maximum Frame Rate120fps120fps112fps
Minimum Frame Rate74fps81fps57fps
5th Percentile Frame Rate88fps90fps61fps
1st Percentile Frame Rate81fps83fps59fps
Frame Time Counts
0ms-8.33ms0 (0%)94 (0.15%)0 (0%)
8.33ms53396 (86.19%)51871 (81.28%)6537 (15.93%)
8.33ms-16.67ms0 (0%)9072 (14.22%)10988 (26.77%)
16.67ms8553 (13.81%)2774 (4.35%)23319 (56.82%)
16.67ms-25ms0 (0%)0 (0%)187 (0.46%)
25ms0 (0%)2 (0%)10 (0.02%)
50ms-58.33ms0 (0%)0 (0%)1 (0%)
58.33ms0 (0%)1 (0%)0 (0%)
Other
Dropped Frames000
Runt Frames000
Runt Frame Thresholds20 rows20 rows20 rows
 
Interesting results for 120fps mode. The PS5 hits 120 more often that the XsX however the XsX averages a higher framerate overall because it dosen't stay at the lower bounds as often as the PS5.
 

Arioco

Member
Interesting results for 120fps mode. The PS5 hits 120 more often that the XsX however the XsX averages a higher framerate overall because it dosen't stay at the lower bounds as often as the PS5.


And apparently one uses VSync and the other does not. I'm just pointing that out because Vsync takes some resources, so the comparison is not exactly like for like.
 
Last edited:

Md Ray

Member
VRR on neither console? Why.

Surely they got the memo?

Those 5000+ torn frames would lead to an fps advantage.

No point really arguing about it though, both versions should have vrr.
They do VRR on a VRR display.

Tearing occurs on Xbox on a non-VRR display due to adaptive Vsync implementation. Vsync turns on when the console hits its target 120fps, when the frame-rate goes below that -- Vsync turns off to preserve input latency at the cost of screen-tearing.
 
Last edited:

MikeM

Member
Can someone explain to me why ps and xbox use different vsync options? Why wouldn’t they both use the same one?
 

Arioco

Member
its as like for like as we gonna get. good results for both though


Yes, I know. And if the rest of the settings are the identical (which apparently they are) that's the only real difference that could any impact on perforce.

By the way, I didn't know this game uses a form of Variable Rate Shading. I take it it's a custom technique created by the devs, it would've been nice to see VRS Tier 2 on Series consoles, to know how both forms of VRS compare in term of IQ and performance gains.
 

Mr Moose

Member
120fps mode isn't the best 120fps mode, but a decent HFR mode. Tearing on the Xbox consoles but you should be using a VRR TV because fuck tearing. Series S is just pointless in that mode, but all have a solid 60fps mode. Maybe the DRS is a bit faulty in the 120fps mode since it's not dropping often, and not at all on the big brothers.
VRR on neither console? Why.

Surely they got the memo?

Those 5000+ torn frames would lead to an fps advantage.

No point really arguing about it though, both versions should have vrr.
VRR is there if your TV supports it. You can't see VRR in a video.
Yes, I know. And if the rest of the settings are the identical (which apparently they are) that's the only real difference that could any impact on perforce.

By the way, I didn't know this game uses a form of Variable Rate Shading. I take it it's a custom technique created by the devs, it would've been nice to see VRS Tier 2 on Series consoles, to know how both forms of VRS compare in term of IQ and performance gains.
They've been using software VRS for years in CoD, and it looks pretty good.
 
Last edited:

phil_t98

#SonyToo
Yes, I know. And if the rest of the settings are the identical (which apparently they are) that's the only real difference that could any impact on perforce.

By the way, I didn't know this game uses a form of Variable Rate Shading. I take it it's a custom technique created by the devs, it would've been nice to see VRS Tier 2 on Series consoles, to know how both forms of VRS compare in term of IQ and performance gains.

the games practically identical on both systems, yeah a few torn frames and a few dropped FPS but pretty much the same. which ever console its bought on everybody be happy
 

DJ12

Member
They do VRR on a VRR display.

Tearing occurs on Xbox on a non-VRR display due to adaptive Vsync implementation. Vsync turns on when the console hits its target 120fps, when the frame-rate goes below that -- Vsync turns off to preserve input latency at the cost of screen-tearing.
So I guess it's left to nxg Gamer to hive us the vrr perofmance then
 

DaGwaphics

Member
Dj Khaled GIF


Never saw a 3fps advantage coming, PS5 basically a slideshow in this.
 

Darsxx82

Member
By the way, I didn't know this game uses a form of Variable Rate Shading. I take it it's a custom technique created by the devs, it would've been nice to see VRS Tier 2 on Series consoles, to know how both forms of VRS compare in term of IQ and performance gains.

The COD graphics engine has been using a in house form of software VRS since the last generation. According to them, its use is more convenient for them because it is applicable to any platform, unlike VRS tier2.
The result is more than satisfactory.

The possibility of using both techniques at the same time could be given, as shown in the technical article of VRS tier 2, but I suppose that the performance achieved is sufficient for them. It is not an exclusive game where you can focus on a certain platform either.
 

SpokkX

Member
PS5/Series X are really close in all tests. This is a good generation - no console gets shit versions like used to happen on PS3/Xbox One
 
Interesting results for 120fps mode. The PS5 hits 120 more often that the XsX however the XsX averages a higher framerate overall because it dosen't stay at the lower bounds as often as the PS5.
Indeed XSX has screen-tearing that significantly improves the average framerate. But the game is actually quite more stable on PS5 with 86% vs 81% (XSX).

Stats don't lie here and we know vsync has a performance cost on PS5 (as in any other game).
 
Last edited:

Mr Moose

Member
No I do, but as he's the only one with a vrr capture card, he's the only one that can test performance in vrr.
You can't capture VRR on video unless you point a camera at the screen, which is pointless because who wants to read a TVs screen display instead of seeing the full game in high quality capture.
 

DenchDeckard

Moderated wildly
Now we trying to spin v sync to gain some credit or something lmao? Is this 2002?

If you don't have vrr with your 120hz tv you doing it wrong. I don't get why you would have that....
 

Clear

CliffyB's Cock Holster
Now we trying to spin v sync to gain some credit or something lmao? Is this 2002?

If you don't have vrr with your 120hz tv you doing it wrong. I don't get why you would have that....

Frame tearing is a sacrifice made to boost performance, so it affects the metrics. Whether you have a means to work around it, or can happily ignore it, it isn't what's relevant.
The point is that the difference in operation is detectable.
 

DenchDeckard

Moderated wildly
Frame tearing is awful for sure. I agree. But why would you have v sync on in a 120hz mode it doesn't make sense....because surely you would have vrr...that removes tearing.
 

DJ12

Member
?? Performance is what you see here...difference is with vrr the drops are not as perceptible and there would not be screentear.
One has vsync one doesn't therefore claing one version performs better or not bas3d on this dataset is flawed.

How do you know that in VRR mode without vysnc enabled ps5 does perform equally as well as Series x.

You don't is the answer.
You can't capture VRR on video unless you point a camera at the screen, which is pointless because who wants to read a TVs screen display instead of seeing the full game in high quality capture.
No but with a vrr capture card he can run the game and framecount the game with vrr enabled. Clearly as they detect tearing in the series version df and vg don't have this capability.
 

xion4360

Member
One has vsync one doesn't therefore claing one version performs better or not bas3d on this dataset is flawed.

How do you know that in VRR mode without vysnc enabled ps5 does perform equally as well as Series x.

You don't is the answer.

No but with a vrr capture card he can run the game and framecount the game with vrr enabled. Clearly as they detect tearing in the series version df and vg don't have this capability.
How much better do you think it would perform with vsync off? Plus with VRR its even harder to tell any difference. Seriesx and ps5 are effectively tied here.
 

Mr Moose

Member
One has vsync one doesn't therefore claing one version performs better or not bas3d on this dataset is flawed.

How do you know that in VRR mode without vysnc enabled ps5 does perform equally as well as Series x.

You don't is the answer.

No but with a vrr capture card he can run the game and framecount the game with vrr enabled. Clearly as they detect tearing in the series version df and vg don't have this capability.
The game still tears with or without VRR.
The Elgato HD60 X has VRR pass-through so you still get the benefits of being able to use VRR on your screen, that's all (or in cases that have a 40+/unlocked modes, it'll be able to capture that versions performance).
 
And apparently one uses VSync and the other does not. I'm just pointing that out because Vsync takes some resources, so the comparison is not exactly like for like.
Actually good they disabled vsync no need for it if you have vrr. Is v sync also disabled in the 60fps mode?
 

avin

Member
Considering the fact that PS5 hits its target of 8.33ms 86.16% of the time with full vsync versus XSX' 81.28% in 120 FPS mode and the state of 60 FPS mode (absolutely locked on PS5), i would say that PS5 performs ever so slightly better on this at the same resolution/settings.
Looking at the 120 fps mode, if the PS5 hits its target more often, and you end up with a lower average, that means when it does drop, it drops deeper.

For me, the real bottom line is how poorly these comparisons seem to be done. I actually don't know how hard these are to do, so I shouldn't criticize, but there's no way to repeat the same run three times, get some measure of scatter? Present an average and a deviation?

avin
 
Last edited:

Lysandros

Member
Looking at the 120 fps mode, if the PS5 hits its target more often, and you end up with a lower average, that means when it does drop, it drops deeper.

For me, the real bottom line is how poorly these comparisons seem to be done. I actually don't know how hard these are to do, so I shouldn't criticize, but there's no way to repeat the same run three times, get some measure of scatter? Present an average and a deviation?

avin
All i can say is that i regard the raw data provided by Vgtech very highly as to accuracy and objectivity, especially compared to let's say DF. Now, no comparison is perfect and accuracy can only increase with more/unlimited time at hand.
 

I Master l

Banned
Considering the fact that PS5 hits its target of 8.33ms 86.16% of the time with full vsync versus XSX' 81.28% in 120 FPS mode and the state of 60 FPS mode (absolutely locked on PS5), i would say that PS5 performs ever so slightly better on this at the same resolution/settings.
ElAnalistaDeBits says Series X maintains a higher average resolution than PS5 and have better shadows
 

M1chl

Currently Gif and Meme Champion
VRR on neither console? Why.

Surely they got the memo?

Those 5000+ torn frames would lead to an fps advantage.

No point really arguing about it though, both versions should have vrr.
I don't think you can turn off VRR on Xbox, it's baked into drivers.
 
ElAnalistaDeBits says Series X maintains a higher average resolution than PS5 and have better shadows
According to NXGamer the PS5 has actually slightly better quality shadows (but it's hard to see) and both versions should have the same framerate if both were vsynced. Resolution is exactly the same in both versions and it's most of the time at max resolution for both.
 

Darsxx82

Member
ElAnalistaDeBits' findings are highly irrelevant, he always manages to invent better results for one specific platform. There is a reason for that channel being banned here. He is simply untrustworthy.

He also could say that NXG is not the best example of impartiality in its analysis either, and we could also say that its tendency is towards a certain side.

In the end, how much or not you want to believe remains in the personal appreciation of each one. Personally, I give the same value to all analyzes and my conclusion comes from all of them.
 

Mr Moose

Member
He also could say that NXG is not the best example of impartiality in its analysis either, and we could also say that its tendency is towards a certain side.

In the end, how much or not you want to believe remains in the personal appreciation of each one. Personally, I give the same value to all analyzes and my conclusion comes from all of them.
Or we could stop using that fraud ElAnalistaDeBits as a reliable source and giving them clicks?

wz0sisO.jpeg

uSOQZuU.jpeg


Let's play a game of spot the difference.
 
Last edited:

Ezekiel_

Banned
Considering the fact that PS5 hits its target of 8.33ms 86.16% of the time with full vsync versus XSX' 81.28% in 120 FPS mode and the state of 60 FPS mode (absolutely locked on PS5), i would say that PS5 performs ever so slightly better on this at the same resolution/settings.
Nodding Yes GIF by Big Brother
 

Darsxx82

Member
Or we could stop using that fraud ElAnalistaDeBits as a reliable source and giving them clicks?
[/URL][/URL]

wz0sisO.jpeg

uSOQZuU.jpeg


Let's play a game of spot the difference.
Do you think that this error has been intentional or simply an error in the fps counting tools? I bet on the second.
When he is accurate (which it is on many occasions).... What should we blame him for or reproach him for?


As I said, everyone is free to take each different analysis as they create. If the point is put in a supposed preference to one side... then putting NXG as an example of impartiality I don't think it's the smartest either.

It is the reason why I personally draw my conclusions based on the set of analyzes that are published and not just based on a specific one.

PS. I don't know if it was your intention, but it's funny that you attack him for fraud and then choose those screenshots that are not very consistent with the reality of both versions either....😉
 
Last edited:

Stuart360

Member
Not going to lie, its nice to see these 'vs' threads dont have the traction on here like they used to have.
I guess many people have realized the odd dropped frame here, a few pixels extra there, really doesnt matter at the end of the day.
 
Or we could stop using that fraud ElAnalistaDeBits as a reliable source and giving them clicks?

wz0sisO.jpeg

uSOQZuU.jpeg


Let's play a game of spot the difference.
Here is the fun part for ps5 vs xsx:
Top screenshots middle image (xsx) is very similar to the bottom screenshots left most image (PS5).

Xsx: 98fps
Ps5: 101fps
 

Mr Moose

Member
Do you think that this error has been intentional or simply an error in the fps counting tools? I bet on the second.
When he is accurate (which it is on many occasions).... What should we blame him for or reproach him for?


As I said, everyone is free to take each different analysis as they create. If the point is put in a supposed preference to one side... then putting NXG as an example of impartiality I don't think it's the smartest either.

It is the reason why I personally draw my conclusions based on the set of analyzes that are published and not just based on a specific one.

PS. I don't know if it was your intention, but it's funny that you attack him for fraud and then choose those screenshots that are not very consistent with the reality of both versions either....😉
One of those pics is Digital Foundry, the other is that fraud, the reality is he is full of shit.
You want to keep playing spot the difference, huh?
jFw97lm.png


bRBOHHk.png


There's a reason he's a banned source on here.
 

Darsxx82

Member
One of those pics is Digital Foundry, the other is that fraud, the reality is he is full of shit.
You want to keep playing spot the difference, huh?
jFw97lm.png


bRBOHHk.png


There's a reason he's a banned source on here.


You still don't answer... Do you think these errors are intentional or simply an error of the FPS counting tools? I think it's the second, I doubt very much that he wants to offer such a blatant image of an inaccurate source.

"Fraud" is a serious qualifier and accusation that he does it intentionally and I highly doubt that those mistakes you point out are intentional.
You have that in the vast majority of cases their results are consistent with the rest of the analysts... In those cases, what can you blame them for? What is the reason for not taking them into account? The prejudice??
 
Last edited:

Mr Moose

Member
You still don't answer... Do you think these errors are intentional or simply an error of the FPS counting tools? I think it's the second, I doubt very much that he wants to offer an image of an inaccurate source.

"Fraud" is a serious qualifier and accusation that he does it intentionally and I highly doubt that those mistakes you point out are intentional. You have that in the vast majority of cases their results are consistent with the rest of the analysts... In those cases, what can you blame them for? What is the reason for not taking them into account? The prejudice??
I thought it was obvious, if you keep making the same "errors" with your tools, then you stop using those tools. If you keep using them and giving fake results then you are a fraud.
 
Top Bottom