• 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] Marvel's Guardians of the Galaxy PS5 vs Xbox Series X|S Frame Rate Comparison

Topher

Gold Member


Xbox code provided by Square-Enix. Marvel's Guardians of the Galaxy frame rate test comparing the framerate/fps on PS5 vs Xbox Series X vs Xbox Series S. Subscribe for more gaming tech analysis: http://bit.ly/VGTechSubscribe

The versions tested were 1.003.000 on PS5 and 2.5.2110.220 on the Xbox Series consoles.

Timestamps:
00:00 - PS5/Xbox Series X Quality Mode and Xbox Series S
08:37 - Performance Mode

A frame can be dropped when performing a dash on PS5 and Xbox Series consoles https://bit.ly/3w1my8a

PS5 in Quality Mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 2880x1620.

Xbox Series X in Quality Mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 3200x1800.

Pixel counts at 3840x2160 were common on PS5 and Xbox Series X in Quality Mode.

The beginning scene of the game seen at the start of the video is where PS5 in Quality Mode drops to 2880x1620 and the resolution on Xbox Series X in this scene only drops to approximately 3648x2052. Outside of the scene at the beginning of the game, pixel counts were similar on PS5 and Xbox Series X with the lowest resolution found on both being 3200x1800.

PS5 and Xbox Series X in Performance Mode render at a native resolution of 1920x1080. Xbox Series S doesn't have a Performance Mode and renders at a native resolution of 1920x1080.

Quality Mode has an improved LOD transition distance compared to Performance Mode on PS5 and Xbox Series X. Scene complexity is noticeably reduced on Xbox Series S compared to both modes on PS5 and Xbox Series X.

Stats: https://bit.ly/3GQ20ET
Frames Pixel Counted: https://bit.ly/31kA6jX
 

FrankWza

Member
Season 2 Fox GIF by The Four
 

Topher

Gold Member
Thinking Reaction GIF


"The quality mode runs at 3840x2160 on PS5 and Series X, capped at 30 frames per second. Temporal anti-aliasing is used on each, and it's the premier way to enjoy the games visuals. There's a chance of dynamic resolution scaling here too. However, every pixel count returns a native 4K on this mode."

 

DenchDeckard

Moderated wildly
This game is crippling these systems, i think it deffo needs some patches.

Theres no way the PS5 should be dropping into the 40s for no apparent reason at 11:10 imo.
 

ethomaz

Banned
Thinking Reaction GIF


"The quality mode runs at 3840x2160 on PS5 and Series X, capped at 30 frames per second. Temporal anti-aliasing is used on each, and it's the premier way to enjoy the games visuals. There's a chance of dynamic resolution scaling here too. However, every pixel count returns a native 4K on this mode."

I believe DF is trying to change the mind of dynamic resolution to be fair.
Seems like it is not like they doesn't know the drops but they don't want to talk about that in the articles... it is something I noticied lately.

I wonder what changed for them internally to take that decision.
 

Topher

Gold Member
I believe DF is trying to change the mind of dynamic resolution to be fair.
Seems like it is not like they doesn't know the drops but they don't want to talk about that in the articles... it is something I noticied lately.

I wonder what changed for them internally to take that decision.

But they do talk about it. They say all pixel counts were native 4K.
 

Darsxx82

Member
According to VGTech, the XSX version not only has a better average fps, it also has an advantage in resolution. Curiously in that intruduction scene where XSX was stable 60fps and PS5 ~ 53fps, the latter also worked at a lower resolution.

It must be a particularly demanding scene in some way when it is also where there is a significant graphic cut (specially vegetation) in XSS which, clearly, is a version that has not received much love 😅
 

Tripolygon

Banned
Stats Quality Mode

PlatformsPS5Xbox Series XXbox Series S
Frame Amounts
Game Frames154371543115433
Video Frames309763097530975
Frame Tearing Statistics
Total Torn Frames200
Lowest Torn Line125--
Frame Height216021601080
Frame Time Statistics
Mean Frame Time33.44ms33.46ms33.45ms
Median Frame Time33.33ms33.33ms33.33ms
Maximum Frame Time166.67ms233.33ms233.33ms
Minimum Frame Time32.37ms33.33ms33.33ms
95th Percentile Frame Time33.33ms33.33ms33.33ms
99th Percentile Frame Time33.33ms33.33ms33.33ms
Frame Rate Statistics
Mean Frame Rate29.9fps29.89fps29.89fps
Median Frame Rate30fps30fps30fps
Maximum Frame Rate30fps30fps30fps
Minimum Frame Rate25fps24fps24fps
5th Percentile Frame Rate29fps29fps29fps
1st Percentile Frame Rate28fps28fps28fps
Frame Time Counts
16.67ms-33.33ms1 (0.01%)0 (0%)0 (0%)
33.33ms15386 (99.67%)15382 (99.68%)15387 (99.7%)
33.33ms-50ms3 (0.02%)0 (0%)0 (0%)
50ms2 (0.01%)2 (0.01%)0 (0%)
66.67ms42 (0.27%)44 (0.29%)43 (0.28%)
83.33ms1 (0.01%)0 (0%)0 (0%)
100ms1 (0.01%)0 (0%)0 (0%)
116.67ms0 (0%)1 (0.01%)1 (0.01%)
133.33ms0 (0%)1 (0.01%)1 (0.01%)
166.67ms1 (0.01%)0 (0%)0 (0%)
233.33ms0 (0%)1 (0.01%)1 (0.01%)
Other
Dropped Frames000
Runt Frames000
Runt Frame Thresholds20 rows20 rows20 rows

Performance Mode

PlatformsPS5Xbox Series X
Frame Amounts
Game Frames3062630627
Video Frames3097530975
Frame Tearing Statistics
Total Torn Frames652123
Lowest Torn Line13071
Frame Height21602160
Frame Time Statistics
Mean Frame Time16.86ms16.86ms
Median Frame Time16.67ms16.67ms
Maximum Frame Time183.33ms266.67ms
Minimum Frame Time15.69ms16.12ms
95th Percentile Frame Time16.67ms16.67ms
99th Percentile Frame Time32.35ms32.82ms
Frame Rate Statistics
Mean Frame Rate59.32fps59.32fps
Median Frame Rate60fps60fps
Maximum Frame Rate60fps60fps
Minimum Frame Rate39fps45fps
5th Percentile Frame Rate56fps55fps
1st Percentile Frame Rate44fps53fps
Frame Time Counts
0ms-16.67ms91 (0.3%)9 (0.03%)
16.67ms29758 (97.17%)30189 (98.57%)
16.67ms-33.33ms700 (2.29%)206 (0.67%)
33.33ms65 (0.21%)213 (0.7%)
50ms5 (0.02%)4 (0.01%)
50ms-66.67ms1 (0%)0 (0%)
66.67ms5 (0.02%)4 (0.01%)
166.67ms0 (0%)1 (0%)
183.33ms1 (0%)0 (0%)
266.67ms0 (0%)1 (0%)
Other
Dropped Frames00
Runt Frames00
Runt Frame Thresholds20 rows20 rows

PS5 and XSX in terms of performance are not bad, if the framerate cap was removed it would be well over 60fps. A few more updates and they could be perfect.
 
Last edited:

Thief1987

Member
According to VGTech, the XSX version not only has a better average fps, it also has an advantage in resolution. Curiously in that intruduction scene where XSX was stable 60fps and PS5 ~ 53fps, the latter also worked at a lower resolution.

It must be a particularly demanding scene in some way when it is also where there is a significant graphic cut (specially vegetation) in XSS which, clearly, is a version that has not received much love 😅
What stats are you looking at? They basically the same - 29.9 vs 29.89 and 59.32 vs 59.32
 

Mr Moose

Member
According to VGTech, the XSX version not only has a better average fps
Huh? That's not what it says.
29.9fps v 29.89fps, 59.32fps v 59.32fps.
it also has an advantage in resolution.
It also doesn't say this.
Outside of the scene at the beginning of the game, pixel counts were similar on PS5 and Xbox Series X with the lowest resolution found on both being 3200x1800.
 

Darsxx82

Member
What stats are you looking at? They basically the same - 29.9 vs 29.89 and 59.32 vs 59.32
Those that show you that XSX has higher minimums and is more stable in more demanding moments, as can also be seen in the DF analysis. You have to understand that the VGtech average is affected by being the average of all the clips. Obviously, if it only counted the demanding scenes, the difference would be greater. The game is stable 60fps until any confrontation starts. The XSX version performs better in performance mode. And better average resolution in 4K mode. From here on, they are very very similar versions.
 

Riky

$MSFT
According to VGTech, the XSX version not only has a better average fps, it also has an advantage in resolution. Curiously in that intruduction scene where XSX was stable 60fps and PS5 ~ 53fps, the latter also worked at a lower resolution.

It must be a particularly demanding scene in some way when it is also where there is a significant graphic cut (specially vegetation) in XSS which, clearly, is a version that has not received much love 😅
Looks that way both a higher resolution floor and a higher framerate floor, plus with the framerate VRR makes it the best way to play on console.
 

Darsxx82

Member
It also doesn't say this.
???

"PS5 in Quality Mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 2880x1620.

Xbox Series X in Quality Mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 3200x1800."
Which clearly means that XSX at the most demanding moments maintains the resolution better. Significant that it's also in the same startup scene where in performance mode XSX was holding 60fps while PS5 was ~ 53fps.
 

DenchDeckard

Moderated wildly
Yeah xbox is at a higher res in example screenshots and has a higher lowest franerate. Less torn frames and less tearing in general.

This engine seems to be nuking these consoles for some reason.

The ps5 is struggling with the alpha effect grass in the intro and the xbox is rock solid.

There's just some weird stuff going on in this game.
 

Mr Moose

Member
???

"PS5 in Quality Mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 2880x1620.

Xbox Series X in Quality Mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 3200x1800."
Which clearly means that XSX at the most demanding moments maintains the resolution better. Significant that it's also in the same startup scene where in performance mode XSX was holding 60fps while PS5 was ~ 53fps.
Pixel counts at 3840x2160 were common on PS5 and Xbox Series X in Quality Mode.

The beginning scene of the game seen at the start of the video is where PS5 in Quality Mode drops to 2880x1620 and the resolution on Xbox Series X in this scene only drops to approximately 3648x2052. Outside of the scene at the beginning of the game, pixel counts were similar on PS5 and Xbox Series X with the lowest resolution found on both being 3200x1800.
 
So according to VGTech there is no DRS in the performance mode. Odd.
Yeah xbox is at a higher res in example screenshots and has a higher lowest franerate. Less torn frames and less tearing in general.

This engine seems to be nuking these consoles for some reason.

The ps5 is struggling with the alpha effect grass in the intro and the xbox is rock solid.

There's just some weird stuff going on in this game.
I don't think the opening scene framerate (and resolution) drops on PS5 are caused by alphas. It's probably engine related. In others demanding scenes at 60fps the PS5 has actually the advantage at the same 1080p resolution when plenty of alphas during gameplay.

RpCPunR.png

At 60fps the average framerate is actually identical between both machines. I think the game clearly lacks optimization on both machines (but particularly on PS5).
 
Last edited:

Thief1987

Member
Those that show you that XSX has higher minimums and is more stable in more demanding moments, as can also be seen in the DF analysis. You have to understand that the VGtech average is affected by being the average of all the clips. Obviously, if it only counted the demanding scenes, the difference would be greater. The game is stable 60fps until any confrontation starts. The XSX version performs better in performance mode. And better average resolution in 4K mode. From here on, they are very very similar versions.
PS5 has lower dips but if they have the same mean framerate that means that PS5 is more stable in general (otherwise they wouldn't have the same average), and drops more frames in some odd specific spots, like beginning scene, or scene on 10.55. And those scenes doesn't look "demanding", like what demanding about 10.55, where ps5 drops to 40fps, and 30 sec later it's locked 60 in a fight with a bunch of enemies? It's just looks like optimization of this game is a mess.
 
Last edited:

Elios83

Member
PS5 has lower dips but if they have the same mean framerate that means that PS5 is more stable in general (otherwise they wouldn't have the same average), and drops more frames in some odd specific spots, like beginning scene, or scene on 10.55. And those scenes doesn't look "demanding", like what demanding about 10.55, where ps5 drops to 40fps, and 30 sec later it's locked 60 in a fight with a bunch of enemies? It's just looks like optimization of this game is a mess.

It's the usual irrelevant differences.
Average frame rate and resolution is the same in most circumstances as clearly explained by Vgtech.
Some people are unhappy with that, they take the worst cases even if they happen once in the whole game and pretend those represent the norm even when it's identical 99% of the time.
The usual for these threads.
 
Last edited:

Kuranghi

Member
I know console games have sometimes had tearing issues in the past, but it seems much more common these days, really sucks. V-sync stutter is crap but I think I still prefer it over tearing.

I never had a 360 at the time but I don't remember PS3 having much tearing across the biggest titles, I'm guessing it due to striving for lower input lag in general.
 
Thinking Reaction GIF


"The quality mode runs at 3840x2160 on PS5 and Series X, capped at 30 frames per second. Temporal anti-aliasing is used on each, and it's the premier way to enjoy the games visuals. There's a chance of dynamic resolution scaling here too. However, every pixel count returns a native 4K on this mode."

Friendship ended with VG Tech, now DF is my best friend.
 

Darsxx82

Member
PS5 has lower dips but if they have the same mean framerate that means that PS5 is more stable in general (otherwise they wouldn't have the same average), and drops more frames in some odd specific spots, like beginning scene, or scene on 10.55. And those scenes doesn't look "demanding", like what demanding about 10.55, where ps5 drops to 40fps, and 30 sec later it's locked 60 in a fight with a bunch of enemies? It's just looks like optimization of this game is a mess.What?? It is clear that you do not know how to interpret the Vgtech data. The average fps is not a game average but....
What?? It is clear that you do not know how to interpret the Vgtech data. The average fps is not a game average but the average that results from the clips. If they only took into account the clips where fps dip occur, which is where the framerate is affected, then you would have a clearer average in favor of XSX.
It is also that you can see it in the DF analysis and they agree and draw the same conclusion when pointing out that XSX better maintains 60fps at those times.

You also have that opening scene that is very significant. In performance mode clearly XSX runs at higher fps. But that advantage is transformed into higher resolution in 4K mode in the same scene.
 

Topher

Gold Member
What?? It is clear that you do not know how to interpret the Vgtech data. The average fps is not a game average but the average that results from the clips. If they only took into account the clips where fps dip occur, which is where the framerate is affected, then you would have a clearer average in favor of XSX.

If the average frame rate is the same while PS5 has a lower floor frame rate then that means XSX drops more frequently, just not as low.
 
Last edited:

skit_data

Member
What?? It is clear that you do not know how to interpret the Vgtech data. The average fps is not a game average but the average that results from the clips. If they only took into account the clips where fps dip occur, which is where the framerate is affected, then you would have a clearer average in favor of XSX.
So you’re saying PS5 has higher avg. fps during gameplay?
 
Last edited:

Topher

Gold Member
That's not true at all. It means a bunch of things but just those numbers don't mean much of anything.

How is it not true? If the top end is 60fps then explain how the average is the same for PS5 and XSX while PS5 has a lower minimum frame rate?
 
How is it not true? If the top end is 60fps then explain how the average is the same for PS5 and XSX while PS5 has a lower minimum frame rate?
Cause that's not how average works over this many frames. Could be down to more drops, or longer sustained drops. With the framerate being good for both it could be the difference of 1 second of drops vs 1.2 seconds of drops
 

Topher

Gold Member
Cause that's not how average works over this many frames. Could be down to more drops, or longer sustained drops. With the framerate being good for both it could be the difference of 1 second of drops vs 1.2 seconds of drops

Longer sustained drops? This is an average of frame rate PER SECOND. It doesn't matter if that second of frame rate drop is one of many consecutively or more sporadicly spread out. There are still more drops for XSX if the average is the same and PS5 has lower floor.
 
Last edited:

avin

Member
Looking at the data [edit: performance mode], the PS5 is also delivering many more frames before it should. I think that would actually increase the calculated average frame rate, but it doesn't seem like a good thing.

Overall, I would say we don't have good metrics for what is trying to be measured.

avin
 
Last edited:

Tripolygon

Banned
I would say we don't have good metrics for what is trying to be measured.

Look at the data. The PS5 is also delivering many more frames before it should. I think that would actually increase the calculated average frame rate, but it isn't seem like a good thing.

avin
Yea shows there is some headroom in terms of performance. If they unlocked the framerate, it would show a much higher average. But the downside would be a very bad gaming experience without VRR. XSX would benefit greatly from an unlocked frame rate option with VRR enabled TVs.
Longer sustained drops? This is an average of frame rate PER SECOND. It doesn't matter if that second of frame rate drop is one of many consecutively or more sporadicly spread out. There are still more drops for XSX if the average is the same and PS5 has lower floor.
Evidently, people don't know how to read a frame time graph.
 

ethomaz

Banned
That's not true at all. It means a bunch of things but just those numbers don't mean much of anything.
It means exactly what he said.

Edit - What is weird in the sheet is that there is a lower frametime on Series X but that are not counted as framerate? That is the only thing I found confuse in the data.

Edit 2 - Not weird anymore... it is totally possible a single frame take more time in Series X but the others frames togheter in the second be higher... after all the framerate per second account how many frame it was generated in a second no matter if one of them took longer to be generated.
 
Last edited:
Longer sustained drops? This is an average of frame rate PER SECOND. It doesn't matter if that second of frame rate drop is one of many consecutively or more sporadicly spread out. There are still more drops for XSX if the average is the same and PS5 has lower floor.
No this is an argument of semantics. A drop can last 5 seconds or 1. Now to make it more drops you would have to talk frame time, and missing it's target. Now both framerates are quite good so to say anything other than they are pretty much equal is kinda silly anyway.
 

Topher

Gold Member
No this is an argument of semantics. A drop can last 5 seconds or 1. Now to make it more drops you would have to talk frame time, and missing it's target. Now both framerates are quite good so to say anything other than they are pretty much equal is kinda silly anyway.

When measuring frames per second, a drop that last 5 seconds is the equivalent to 5 drops (1 per second). This is what is bringing the XSX average frame rate down to the same level as PS5. The only semantics is recognizing what "per second" means here.

But I agree both frame rates coming to the same average is what is most important.
 
Last edited:

avin

Member
Yea shows there is some headroom in terms of performance. If they unlocked the framerate, it would show a much higher average. But the downside would be a very bad gaming experience without VRR. XSX would benefit greatly from an unlocked frame rate option with VRR enabled TVs.

I guess my point is that the early frames also pull the average frame rate up, but I don't know if that's a good thing, given what we're trying to measure. Do early frames feel as bad as dropped frames, or does one not notice these? As always, grateful for any insight.

avin
 
Last edited:

Warnen

Don't pass gaas, it is your Destiny!
Personally, I can't stand it. I don't understand those who say quality mode is preferred. To each their own, but it isn't for me.
Took a few mins to get used to but it does look a lot better. Think I’ll leave it in that mode from now on. Better get myself used t0 30fps again, gonna be a long gen.
 

CrustyBritches

Gold Member
In performance mode PS5 has the lower Min, 1% Min, and 5 times the amount of torn frames. Otherwise they seem the same. 60fps is much better than 30fps in this game, imo.
 
Last edited:
When measuring frames per second, a drop that last 5 seconds is the equivalent to 5 drops (1 per second). This is what is bringing the XSX average frame rate down to the same level as PS5. The only semantics is recognizing what "per second" means here.

But I agree both frame rates coming to the same average is what is most important.
Well the issue with that is I see a lot of talk about a big drop or small drop in fps and no one says 6 big drops in a row for a 6 second stretch. They just call it a big drop. Fps is also bad for this since fps is just the avg of frame time for the second.
Again just arguing semantics for the fun of it. Both are performing fine aside from the tearing.
 

ethomaz

Banned
Took a few mins to get used to but it does look a lot better. Think I’ll leave it in that mode from now on. Better get myself used t0 30fps again, gonna be a long gen.
Like I said in the others thread if you keep sifting between the de diference is really big... so if you started playing in 60fps and right away change to 30fps to see if it good your eyes will have to adapt..

Now if you started already in 30fps or did not play any higher framerate game in the last few hours you will barely notice difference.

Depend from person to person but for me it take around 30 minutes to lose that sensation of slowdown when shifting from 60 to 30fps... that doesn't happen if I wake up for example and start to play directly in 30fps. The opposite happens too... shifting from 30 to 60fps gives me a fast forward sensation but it is gone in around 30 minutes and if I started playing in 60fps direct it feels normal.

Funny enough that happened with me playing Destiny (30fps) and Bloodborne (30fps with bad frametime)... I was playing Bloodborn and so friend called for Raid in Destiny and when I shifted the Destiny give me the sestation for several minutes that it was playing in fast foaward mode lol... after that the sensation gone.

Without shifting I have no issue with either 30 or 60fps.
 
Last edited:
Top Bottom