• 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.

[Brazil Pixel] Ghostwire Tokyo: Xbox Series X vs PS5 vs PS5 Pro - Performance and Resolution Test



Note from creator:
"PS: There was a small error when measuring the resolution of the PS5 PRO in HFR Quality V-Sync mode (When comparing the 3 consoles on the same screen). The resolution is fixed at 1080p but for some frames it was measured at 968p. Consider constant 1080p."
 

Skifi28

Member
On PS5 there were framerate drops and on PC constant micro-stuttering, the game was just horrible everywhere. I might just give it replay now on the pro.
 

lh032

I cry about Xbox and hate PlayStation.
Nah it's alright.

The game suffered more from the map-vomit checklist thing.

Very fun game otherwise.
i remember the input lag/delay is really bad, i have to tweak the control setting myself but the game control just dont feel tight at all.
 

adamsapple

Or is it just one of Phil's balls in my throat?
i remember the input lag/delay is really bad, i have to tweak the control setting myself but the game control just dont feel tight at all.

Yes that was definitely an issue, the dead zone defaults were whack. Needed a lot of tweaking. Played both the PS5 and Xbox versions and it was the same on both.
 

Lysandros

Member
Pretty unknown channel, how do we know their measurement tools are accurate? Especially when Ghostwire Tokyo results beetween PS5/XSX are completely at odds with other outlets like DF which showed to the game having a significant advantage in favor of PS5 in both the performance and resolution/RT over XSX. I don't think this comparison quite earns this thread in this context.
 

Skifi28

Member
Pretty unknown channel, how do we know their measurement tools are accurate? Especially when Ghostwire Tokyo results beetween PS5/XSX are completely at odds with other outlets like DF which showed to the game having a significant advantage in favor of PS5 in both the performance and resolution/RT over XSX. I don't think this comparison quite earns this thread in this context.
DF had a section on the channel and there were pretty impressed with what they've done.
 

JaksGhost

Member
Pretty unknown channel, how do we know their measurement tools are accurate? Especially when Ghostwire Tokyo results beetween PS5/XSX are completely at odds with other outlets like DF which showed to the game having a significant advantage in favor of PS5 in both the performance and resolution/RT over XSX. I don't think this comparison quite earns this thread in this context.
Funny because DF said they were pretty credible with their tools in a recent video. They even sourced them and gave them a shoutout. Also this was probably done after the DLC update so tweaks were made to the overall game.
 
Last edited:

SkylineRKR

Member
Its okay. The map isn't too big and once you can glide far its quite easy to traverse. Action is just an FPS, there is a melee weapon, a shotgun type, pistol type etc. Those talismans were grenades with different properties. And some forced stealth sections because your powers are robbed... yawn. I mostly prefered to avoid direct combat but rather picked enemies off from a building.

I kind of liked the indoor designs of the apartments in this game.

Art direction is very good, game looks colorful and it uses RT in a meaningful way. The game is just okay... nothing groundbreaking.
 
  • Like
Reactions: N0S

Lysandros

Member
Funny because DF said they were pretty credible with their tools in a recent video. They even sourced them and gave them a shoutout. Also this was probably done after the DLC update so tweaks were made to the overall game.
I don't know, seems pretty peculiar. I don't think the game's performance profile made such a drastic 180 degree turn and suddenly bacame at odds with both the IGN/NXgamer's and DF's analysis in a game using an engine which pretty consistently favored PS5. Smells a bit fishy to me.
 

Mr Moose

Member
Funny because DF said they were pretty credible with their tools in a recent video. They even sourced them and gave them a shoutout. Also this was probably done after the DLC update so tweaks were made to the overall game.
The DLC was out when DF did their vs, it released with the Xbox version.



Ghostwire-Xbox-Analysis-Revised-H264.mp4_snapshot_08.40_%5B2023.04.18_09.30.00%5D.jpg
I don't know, seems pretty peculiar. I don't think the game's performance profile made such a drastic 180 degree turn and suddenly bacame at odds with both the IGN/NXgamer's and DF's analysis in a game using an engine which pretty consistently favored PS5. Smells a bit fishy to me.
The quality HFR mode has better RT on PS5.
 
Last edited:

JaksGhost

Member
Its okay. The map isn't too big and once you can glide far its quite easy to traverse. Action is just an FPS, there is a melee weapon, a shotgun type, pistol type etc. Those talismans were grenades with different properties. And some forced stealth sections because your powers are robbed... yawn. I mostly prefered to avoid direct combat but rather picked enemies off from a building.

I kind of liked the indoor designs of the apartments in this game.

Art direction is very good, game looks colorful and it uses RT in a meaningful way. The game is just okay... nothing groundbreaking.
Once I got the ability to summon Tengus on the rooftop, it became a cakewalk. For the most part, I used the same strategy of taking out most of the powerful enemies from the rooftops and using the stun talisman on groups. There was a good game in there, but they started to bloat it out quite a lot, and once I finished the story, I was done with the game for good and didn't even try the Spider Threads DLC.
 

Toots

Gold Member
This game had wonderful ambiance and art direction.
Needed a few more spells and enemy types and it would have been stellar.
 
  • Like
Reactions: RNG

Darsxx82

Member
Pretty unknown channel, how do we know their measurement tools are accurate? Especially when Ghostwire Tokyo results beetween PS5/XSX are completely at odds with other outlets like DF which showed to the game having a significant advantage in favor of PS5 in both the performance and resolution/RT over XSX. I don't think this comparison quite earns this thread in this context.
The fact is that DF herself has praised the tools and fidelity of this same website in recent videos.... Not to mention that the game has received successive patches and the situation may not be the same as the launch.

But hey!, we're going to ask the mods to ban another source because what it says doesn't match what we want...:messenger_smirking:
The DLC was out when DF did their vs, it released with the Xbox version.



Ghostwire-Xbox-Analysis-Revised-H264.mp4_snapshot_08.40_%5B2023.04.18_09.30.00%5D.jpg
The quality HFR mode has better RT on PS5.

XSX version had 2 specific performance updates after that DF video...Any analysis after those updates/patches??
 

Mr Moose

Member
XSX version had 2 specific performance updates after that DF video...Any analysis after those updates/patches??
Not from DF I don't think, but the one in the OP is.
Performance Enhancements
We’ve added various optimizations to improve overall game performance, particularly on Xbox consoles—including an HFR Performance Graphics Mode for Xbox Series S, which should result in a smoother experience when enabled. You can find this new setting under Options > Graphics.
Doesn't seem like it did anything though, it's still 10-11fps difference at the same spot.
Might've just been improvements for the S.
 

Darsxx82

Member
Not from DF I don't think, but the one in the OP is.
Doesn't seem like it did anything though, it's still 10-11fps difference at the same spot.
Might've just been improvements for the S.
But only in 120fps performance mode. In the rest, including quality mode with unlocked fps vsync the result is a even or slightly higher in XSX. Not to mention that, assuming the tools are correct, XSX would also have a very slight advantage in DRS also.

As far as I remember, at launch the XSX version lacked in framerate and resolution in all modes vs PS5. Presumably, those patches did mean some improvements for the XSX version vs launch version.
 

Mr Moose

Member
But only in 120fps performance mode. In the rest, including quality mode with unlocked fps vsync the result is a even or slightly higher in XSX. Not to mention that, assuming the tools are correct, XSX would also have a very slight advantage in DRS also.

As far as I remember, at launch the XSX version lacked in framerate and resolution in all modes vs PS5. Presumably, those patches did mean some improvements for the XSX version vs launch version.
It has worse RT in quality mode. I don't know if DF did a side by side with that mode (quality HFR). Wouldn't have an advantage in res because it's 1080p.
 
i remember the input lag/delay is really bad, i have to tweak the control setting myself but the game control just dont feel tight at all.

Started it on the PS5, finished it on the Pro : no such problem exists anymore Ih0, game plays really well now, granted, it's no Halo/COD when it comes to the free look settings but, everything feels snappy compared to when it came out.

It's also a pretty chill and atmospheric game with some really nice moments (school DLC etc), its shortcomings being A)the weak/cliche story and B) the Ubisoft-like game loop and OCD inducing, map filled with 10073913 icons/activities/shitty collectibles.

If you don't try to do everything and burn yourself out 'cause of it, I'd say that it's a pretty solid 6.5/10 game
 

Darsxx82

Member
It has worse RT in quality mode. I don't know if DF did a side by side with that mode (quality HFR). Wouldn't have an advantage in res because it's 1080p.
I'm referring to the quality modes where the resolution varies between 1300p-4k. And in quality mode, there on gameplay same spot, resolution and performance are on par or even slightly above in XSX. At launch. DF compared each mode and pointed out an advantage in resolution (DRS) and framerate on all quality modes where there is DRS .

That's why I say that the patches seem to have improved the release version situation of XSX although it has not done so to the point of matching it to the PS5 version in all aspects and modes.
 
Last edited:

Mr Moose

Member
I'm referring to the quality modes where the resolution varies between 1300p-4k. And in quality mode, there on gameplay same spot, resolution and performance are on par or even slightly above in XSX. At launch. DF compared each mode and pointed out an advantage in resolution (DRS) and framerate on all modes quality modes .

That's why I say that the patches seem to have improved the release version situation of XSX although it has not done so to the point of matching it to the PS5 version in all aspects and modes.
The 30fps mode? Hard to say. They aren't exact matched gameplay (though, close enough) and it says Xbox has a lower lowest. Not sure how the one in this video works either but we know it isn't completely accurate as it's showing 968p at times for PS5 Pro and we know that's not correct. We'd need someone to stand in the exact same spot with a good edge to be sure.

You could be right though.
 

Darsxx82

Member
The 30fps mode? Hard to say. They aren't exact matched gameplay (though, close enough) and it says Xbox has a lower lowest. Not sure how the one in this video works either but we know it isn't completely accurate as it's showing 968p at times for PS5 Pro and we know that's not correct. We'd need someone to stand in the exact same spot with a good edge to be sure.

You could be right though.
According to this video, in all modes the resolution and/or performance are on par or slightly higher in XSX. It is in the 120fps performance mode where the deviation in favor of PS5 continues.

In DF's analysis at launch on XSX it was worse in all those sections in all modes. If this analysis on this website is correct, perhaps it is true that the XSX version improved with those patches. As you say, difficult to say without further analysis to corroborate.
 
According to this video, in all modes the resolution and/or performance are on par or slightly higher in XSX. It is in the 120fps performance mode where the deviation in favor of PS5 continues.

In DF's analysis at launch on XSX it was worse in all those sections in all modes. If this analysis on this website is correct, perhaps it is true that the XSX version improved with those patches. As you say, difficult to say without further analysis to corroborate.

I quick math'd about a 1.6-2.2 million pixel delta in favor of the PS5 PRO in all modes that utilize DRS, in addition, higher FPS averages.
 

BigLee74

Member
From what little I played (awful game) it ran like shit on the XSX - but this was a while ago and likely before any patching.
 
Last edited:

Skifi28

Member
I enjoyed it.
fat-cat.jpg
Same, but it took me a while to warm up to it. Almost dropped it several times in the first 3-4 hours. Ended up really liking it and finishing it on both console and PC. It was very weird since I have a pretty good grasp of my likes and dislikes after all these years and usually I can instantly tell if I'll like something or not. I can't remember the last time when this happened.
 
Last edited:

Fafalada

Fafracer forever
Pretty unknown channel, how do we know their measurement tools are accurate? Especially when Ghostwire Tokyo results beetween PS5/XSX are completely at odds with other outlets like DF which showed to the game having a significant advantage in favor of PS5 in both the performance and resolution/RT over XSX. I don't think this comparison quite earns this thread in this context.
Well the performance modes are still considerably in favour of PS5.
Resolution metrics are impossible to verify - DF (or any other manual pixel counter) approach isn't giving you resolution over time metrics - so it's always comparing individual frames (no matter which platform shows advantage), making their analysis at best anecdotal, not representative of performance overall (even though people love to cherry pick said frames as evidence of X% whatever).
This channel provides actual timeline of resolution - so 'if' reliable - it would superseed anything DF does with pixel counting. However - methodology is near impossible to prove without access to ground truth (meaning actual internal metrics from the console version). And I've also heard from DF that the setup is - numerically not very stable/robust, which for me says - take it with a grain of salt until proven otherwise.
 

StereoVsn

Gold Member
I liked the game. To be sure it could have used more work, but overall it was quite good, IMO. Too bad now we won’t have a sequel :(.
 

SKYF@ll

Member
According to this video, in all modes the resolution and/or performance are on par or slightly higher in XSX. It is in the 120fps performance mode where the deviation in favor of PS5 continues.

In DF's analysis at launch on XSX it was worse in all those sections in all modes. If this analysis on this website is correct, perhaps it is true that the XSX version improved with those patches. As you say, difficult to say without further analysis to corroborate.
I've played all versions. (PS5,PS5 Pro,XSX)
I think the XSX version has better performance than at launch.
It works similarly to the PS5 version, except for the poor quality of the RT reflections.
AKIdReq.jpg
 

mckmas8808

Mckmaster uses MasterCard to buy Slave drives
I've played all versions. (PS5,PS5 Pro,XSX)
I think the XSX version has better performance than at launch.
It works similarly to the PS5 version, except for the poor quality of the RT reflections.
AKIdReq.jpg

NICE!
 

Lysandros

Member
I've played all versions. (PS5,PS5 Pro,XSX)
I think the XSX version has better performance than at launch.
It works similarly to the PS5 version, except for the poor quality of the RT reflections.
AKIdReq.jpg
What's the source for those metrics? Did you test it yourself?
 
Last edited:

Xyphie

Member
Theoretically, the accuracy and methodology of his automated pixel counting method should be fairly easy to validate on PC. Figure out how a game stores the internal resolution and just read it out every frame and compare that value against the prediction and then do that in a set of different games with dynamic resolution options. In a few hours of gameplay you could validate millions of frames.
 
Top Bottom