Support NeoGAF

[Mr Moose]

we shall see, just remember my posts.
Im glad control is already showing what i had posted about last year.

I'll remember your posts.

It is time for all gamers to accept MS designed a better and more powerful console for next gen. Sony has been out-thought.
'Sustained' is going to hit home hard, just wait.

What is left is the price war.

"Imo MS locking at sustained speeds is a mistake to be corrected"

I think SX designs has more to offer. But by locking at only 1.8ghz, seems like MS is selling it too short
Look at RX6800, the closest to SX.

RX6800 runs metro 1440p ultra at around 2.25ghz, takes only 230W. It generates 85fps, beats 2080Ti Fe.
From reviews, a 7% overclock 6800, gives about 3% fps improvements
Working backwards, 1.8ghz to 2.25ghz is a 25% increase. So if we capped RX6800 to 1.8ghz, it will lose 10~12% in perf, so Metro 1440p may lose 10fps in doing so.

Let see if any reviewers review capped 6800/6800XT clocks. Please share if you come across them.

 

[Clear]

Using photo mode is far from ideal as it skews so hard towards the rasterization side of the engine/GPU's duties.

The scene is basically "frozen" so all dynamic processes (physics etc.) that would normally be run on there will be suspended, basically its just throwing triangles at the screen and performing shader operations.

 

[FatKingBallman]

Hammered. Standing in the same spot DF used I was at 29fps compared to PS5's 32fps and XSX's 33fps. It's a very heavy scene GPU-wise. I can't offer a reasonable explanation for the close perf in this scene other than XSX has been pushed to bottleneck with a combo of RT Reflections, Transparent Reflections, and volumetrics.

The gameplay is capped at 30fps. If it was uncapped this would be the result. ^Already answered.

What was the resolution when you tested this scene on your PC?

 

[J_Gamer.exe]

Saw a comment on YouTube that ps5 variable clocks were in play here, I mean they always are, but in that maybe its not needed to be higher, I know it was mentioned in like a map screen for eg it may lower clocks when not needed, would photomode do that also? Given the repetitive nature of the processing, would that lead to whats deemed unnecessary heat therefor its lowered? I didn't think it would tbh am just curious of others thoughts. But maybe the console doesn't see the need to be full throttle in photomode? Don't think this is the case though although could be wrong.

Another totally separate point is about the sx when rendering dense foliage. Is that one of the sx major weaknesses and ps5 strengths?

 

[Armorian]

Dude is full of shit.

What he does with photo mode is benchmark the GPU. Its very much a GPU benchmark. Is it a valid environment of how a game runs no. But that's not the point. They check what the GPU's are running and how they perform and its clear the xbox GPU is faster then what the PS5 has which we all know that it was the case.

In the most heavy place (that is not CPU limited) - "corridor of doom", both perform almost the same, this is probably the most "real world" example in all of this video. In other places it's most likely just pure RT advantage thanks to more GPU cores (but there are some small differences in GFX quality that were posted in this thread already). This should be reflected in game like WDL for example that is heavy in RT compared to other console games but it drops res exactly the same on both - Ubisoft Parity?

 

[CrustyBritches]

What was the resolution when you tested this scene on your PC?

1440p base resolution, 4K output resolution. Same settings as provided by Remedy to DF.

 

[FatKingBallman]

1440p base resolution, 4K output resolution. Same settings as provided by Remedy to DF.

With RT?

 

[CrustyBritches]

With RT?

Yes. I tested with RT Medium(reflections + transparency reflections), RT reflections alone, transparency reflections alone, and no RT. I was getting 29fps in that spot with RT, 63fps without.

 

[Allandor]

It is not exactly the same spot. Might be the RT distance, or camera angle.
Or just a bug

 

[FireFly]

In the most heavy place (that is not CPU limited) - "corridor of doom", both perform almost the same, this is probably the most "real world" example in all of this video. In other places it's most likely just pure RT advantage thanks to more GPU cores (but there are some small differences in GFX quality that were posted in this thread already). This should be reflected in game like WDL for example that is heavy in RT compared to other console games but it drops res exactly the same on both - Ubisoft Parity?

The difference in the texture, compute and intersection test rates should be the same, since they are all calculated per CU.

 

[FatKingBallman]

Yes. I tested with RT Medium(reflections + transparency reflections), RT reflections alone, transparency reflections alone, and no RT. I was getting 29fps in that spot with RT, 63fps without.

Lol. Your CPU is i think weaker than CPUs in PS5/XSX. And miniscule CPU usage. Up to 30% of CPU usage is for less demanding games. And Control surely isn't a less demanding game.

 

[Fredrik]

It is not exactly the same spot. Might be the RT distance, or camera angle.
Or just a bug

I tried experimenting on PC, the texture never disappear no matter what settings you use or if you disable RT or how you twist the camera. Like I said earlier I won’t buy this version just to confirm that it’s a bug, but it looks like the texture is simply missing on that object and since there is settings parity that says bug to me. People always like to make a hen from a feather but the game is out so anybody with a Xbox and the game can confirm this within like a minute. Just fast travel there and go check, take a photo and post. If it’s there then it’s there. If it’s not there then someone who think it’s important can contact Remedy and they’ll add it. Then it’s there.

 

[Three]

its actually the best benchmark yet because the game and setting are exactly the same..... photo mode just unlocks the frame cap.

Photo mode doesn't just remove the framecap without reason. The framecap gets removed because you no longer have to deal with any physics, AI or anything else gameplay related so you can afford to uncap it. During gameplay you can't.

 

[Three]

Reeeera made a thread as well. Going as predictable as this one. Somehow worse there. The sony blood runs deep in both these forums. People gotta twist the facts so sony comes out at least on par. Even with the numbers smacking them in the face.

Think back to all the idoits who complain that replay mode or photo mode looks much better in GT Sport and say that is not a fair comparison. There is no "sony blood" in these forums.

 

[Elios83]

As many others have pointed out this video has not much sense.
They used photomode to try to create a specific GPU limited benchmark, they got some results which are in line with the on-paper GPU specs difference but they ended up with tons of unanswered questions.
The three obvious conclusions being that:

1) GPU related differences are not enough to translate into different presets for the game even just for ray tracing settings, actually as some peole have found out it seems like there are spots where things are reflected with higher quality in the PS5 version. The usual bug that always hits only the XSX version?

2)Once you shift to a real world situation XSX loses all its GPU related advantages and it actually ends up perfoming worse overall than PS5. What is this suggesting? That this game is CPU limited and PS5 fares better in scenarios where CPU is a bottleneck so much that the GPU avantages are pretty much lost?

3)If they could reproduce the on paper GPU difference (16%) on the GPU side, what about the tools excuse? The tools are just needed for the CPU?

 

[MasterCornholio]

It isn't, but when your feelings are hurt people lash out.

People lash out in different ways when their feelings are hurt. Sometimes they insult others other times they keep making the same type of posts to prove a point. One of my least favorite things is the constant bombardment of cherry picked screenshots.

 

[MightySquirrel]

There is no stuttering on PC. I played the game for hours a few days ago.

Fact. Its smooth as butter.

[as long you set the resolution and settings appropriate to your PC specs]

It is not exactly the same spot. Might be the RT distance, or camera angle.
Or just a bug

Yea the distance from the player and the thing hanging on the wall is slightly greater on SX and that could be the exact cutoff point where rt render distance drops on consoles. A smudge closer and you're likely to see same reflection on SX . [dev after all confirmed that both cunsoles use identical settings]

 

[MightySquirrel]

I tried experimenting on PC, the texture never disappear no matter what settings you use or if you disable RT or how you twist the camera. Like I said earlier I won’t buy this version just to confirm that it’s a bug, but it looks like the texture is simply missing on that object and since there is settings parity that says bug to me. People always like to make a hen from a feather but the game is out so anybody with a Xbox and the game can confirm this within like a minute. Just fast travel there and go check, take a photo and post. If it’s there then it’s there. If it’s not there then someone who think it’s important can contact Remedy and they’ll add it. Then it’s there.

PC has double RT reflections resolution over console versions and could possibly have greater distance for rays to report back, but you're right anyone on SX can check this place within a minute and dispute or confirm that screenshot.

 

[DForce]

It is not exactly the same spot. Might be the RT distance, or camera angle.
Or just a bug

I just tested this scene out on the PS5. It appears at any angle and you can see the reflection as soon as you enter the door.

 

[MightySquirrel]

I just tested this scene out on the PS5. It appears at any angle and you can see the reflection as soon as you enter the door.

Well there it is*. Now all we need an xbox owner to check that spot.

 

[MasterCornholio]

ps5 quoted performance is 'up to' with 'continuous variable boosts'
while SX is sustained performance, not just for apu but even its ssd.

So where's your evidence that the PS5s SSD has variable performance? I read that both are sustained but your claiming that there's throttling with the PS5s SSD.

 

[FireFly]

2)Once you shift to a real world situation XSX loses all its GPU related advantages and it actually ends up perfoming worse overall than PS5. What is this suggesting? That this game is CPU limited and PS5 fares better in scenarios where CPU is a bottleneck so much that the GPU avantages are pretty much lost?

We don't know that all GPU advantages are lost, since the frame rate is capped. It could be that XSX performs better overall, but has drops in fillrate or IO limited situations that the PS5 doesn't have.

 

[Radical_3d]

the delta could be 20-30%, lets just remember this.

ps5 quoted performance is 'up to' with 'continuous variable boosts'
while SX is sustained performance, not just for apu but even its ssd.

now look at the size, noise, cooling and power requirements of rx6800.

in this control benchmark with uncapped fps, the cpu usage is much less, which means more power is channeled to the gpu.
And ps5 still lags 16% behind SX.
in future games which keeps cpu and gpu and ssd always busy, it will be fun to watch the variable vs sustained specs design.

This will be fun to quote the next game where the PS5 rams over the XSX.

 

[Night.Ninja]

I reckon we are going to get the size difference video soon. The difference between the two game sizes is crazy.

25GB is amazing compared to 42GB.

 

[Shmunter]

So XsX is running lower fidelity RT than PS5? A bug? Could the bug be perpetual increasing performance?

 

[DForce]

PS5

I tried walking far away until the door automatically closed.

I'm not here to show that XsX has lower quality RT reflections, I'm just providing this photo.

 

[Elios83]

We don't know that all GPU advantages are lost, since the frame rate is capped. It could be that XSX performs better overall, but has drops in fillrate or IO limited situations that the PS5 doesn't have.

Isn't that the same of stating that PS5 might have other advantages at play that end up balancing the whole thing?
If the GPU related avantages were consequential the most obvious use would have been to enable higher quality ray tracing settings with the same frame rate cap on the XSX. Remedy couldn't do that.
And removing the cap with the current settings has no sense since the system is nowhere in a position of being able to render close to 60fps so you'd just end up with a hugely fluctuating frame rate all the time.
So yeah, the GPU difference that DF has discovered using a really specific and academic benchmark situation simply couldn't be used in any factual way by Remedy in a real world situation and in real world siutation where CPU is involved (with phisics, AI and more) the two systems overall end up performing the same, actually PS5 has no stutters.

 

[Cleared_Hot]

this gen more than any really demands to wait at least a year, but more like 2-3.

great games take way longer than ever before to develop; furthermore, game developers need more time to exploit the hardware. plus, historically the best games will be out starting at the midway point of the console life (3-6yrs)

 

[phil_t98]

So XsX is running lower fidelity RT than PS5? A bug? Could the bug be perpetual increasing performance?

on PS5 there is a scene where objects are not appearing on table where they are on the X so seems like both having differences

 

[Shmunter]

PS5

I tried walking far away until the door automatically closed.

I'm not here to show that XsX has lower quality RT reflections, I'm just providing this photo.

Need a budding analysts to post some XsX shots

 

[Shmunter]

on PS5 there is a scene where objects are not appearing on table where they are on the X so seems like both having differences

In the DF video? Time stamp?

 

[paulyboy81]

Well there it is*. Now all we need an xbox owner to check that spot.

Posted this last night, but it's been lost in the swamp of posts.

The reflection is completely missing on Series X, during RT mode or Photo Mode, regardless of how close you stand.

BUT, walk around to the opposite side of that pillar and the exact same poster/map is there, being reflected just fine.

Video here!

It's a bug, basically, just like I'd wager the rest of these tiny differences that people keep finding are.

 

[Shmunter]

Posted this last night, but it's been lost in the swamp of posts.

The reflection is completely missing on Series X, during RT mode or Photo Mode, regardless of how close you stand.

BUT, walk around to the opposite side of that pillar and the exact same poster/map is there, being reflected just fine.

Video here!

It's a bug, basically, just like I'd wager the rest of these tiny differences that people keep finding are.

Think that settles it. Bug.

Wonder what the FPS is on the other side.

 

[MightySquirrel]

Posted this last night, but it's been lost in the swamp of posts.

The reflection is completely missing on Series X, during RT mode or Photo Mode, regardless of how close you stand.

BUT, walk around to the opposite side of that pillar and the exact same poster/map is there, being reflected just fine.

Video here!

It's a bug, basically, just like I'd wager the rest of these tiny differences that people keep finding are.

Very strange. I mean rays travel from players perspective, they hit something and travel back to show you what they saw regardless of what it was. They shouldn't care what they hit.

Rays on SX must be truly evolutionary next gen and are capable of discriminating against certain objects.

 

[Leyasu]

Thi
Think that settles it. Bug.

Wonder what the FPS is on the other side.

If the RT settings were less on the XsX like what some are hoping for, then wouldn't it be apparent everywhere? And easily noticeable in every screenshot?

 

[Shmunter]

If the RT settings were less on the XsX like what some are hoping for, then wouldn't it be apparent everywhere? And easily noticeable in every screenshot?

Alex Buttuglier only did a limited selection apparently

 

[Leyasu]

Alex Buttuglier only did a limited selection apparently

Just looking at the screenshots, apart from that missing reflection, do there appear to be any differences?

I can't see any. Can you? or anyone else?

 

[martino]

I tried experimenting on PC, the texture never disappear no matter what settings you use or if you disable RT or how you twist the camera. Like I said earlier I won’t buy this version just to confirm that it’s a bug, but it looks like the texture is simply missing on that object and since there is settings parity that says bug to me. People always like to make a hen from a feather but the game is out so anybody with a Xbox and the game can confirm this within like a minute. Just fast travel there and go check, take a photo and post. If it’s there then it’s there. If it’s not there then someone who think it’s important can contact Remedy and they’ll add it. Then it’s there.

same for me. it's always there on pc coming from the long stair.
i can manage to make the one of the other side disappear with multiple camera angles but not the one on this side.

i found corridor of doom in the process and it's indeed doom

 

[Fredrik]

Posted this last night, but it's been lost in the swamp of posts.

The reflection is completely missing on Series X, during RT mode or Photo Mode, regardless of how close you stand.

BUT, walk around to the opposite side of that pillar and the exact same poster/map is there, being reflected just fine.

Video here!

It's a bug, basically, just like I'd wager the rest of these tiny differences that people keep finding are.

It’s there in photo mode as well?
I found a potato cam photo from a XSX playthrough showing the texture but I held on to it because this isn’t really my fight and I assumed that Xbox gamers would’ve shown some interest to dispute that bit. It’s a great game after all, people should play it and check for differences instead of zooming in on a video. I’m playing it on Geforce Now, plays great, looks great, 60fps, highest settings.

 

[Thirty7ven]

Gloating about 16% more performance just seems so pedantic when it’s all said and done.

My problem with the video is really how a static image isn’t a good representation of performance during gameplay. And it should have been a video theorizing about the reasons why, instead it’s being taken at face value.

But the amount of energy being spent on a 16% difference is astonishing.

 

[RGB'D]

And here comes the PS brigade. You don't have to win every time kiddos... it will be all right.

 

[FatKingBallman]

It is not exactly the same spot. Might be the RT distance, or camera angle.
Or just a bug

That room is so spacious that ot doesn't matter if you are few steps behind or infront. Reflections should have to be seen.

 

[martino]

Gloating about 16% more performance just seems so pedantic when it’s all said and done.

My problem with the video is really how a static image isn’t a good representation of performance during gameplay. And it should have been a video theorizing about the reasons why, instead it’s being taken at face value.

But the amount of energy being spent on a 16% difference is astonishing.

you should read the thread again without bias you sadly cannot remove because there is more energy spent at being angry of it being shown than playing with the result or claiming a win....

 

[Topher]

And here comes the PS brigade. You don't have to win every time kiddos... it will be all right.

Obviously you didn't listen to Arnold's talk on what is "best in life"....

 

[paulyboy81]

It’s there in photo mode as well?

In all the instances I've tried, photo mode is identical to RT gameplay.

That poster/map is missing in both, just as it's present on the opposite side of that pillar in both.

 

[MightySquirrel]

PS5

I tried walking far away until the door automatically closed.

I'm not here to show that XsX has lower quality RT reflections, I'm just providing this photo.

This is on the opposite side to the DF one.

 

[MasterCornholio]

And here comes the PS brigade. You don't have to win every time kiddos... it will be all right.

Win what exactly? I know this is just photomode but with the gameplay they are practically the same. The only thing that I can see being a big delta is the install size but that really doesn't matter of you have plenty of space on the drive.

 

[Thirty7ven]

you should read the thread again without bias you sadly cannot remove because there is more energy spent at being angry of it being shown than even playing with the result or claiming a win....

The merits of this benchmark should be discussed, in what it’s good for and why it’s a good indicator of A but not a good indicator of B.

It could be used as a learning experience on how game engines work.

If this type of benchmark is used as an indicator of overall performance, then it’s a lie.

 

[Elog]

There are some posts in this thread that are a bit sad.

Before the launch of the XSX and PS5 the very same people claimed that the XSX had a significant advantage over the PS5 (we all saw the posts). After launch the two consoles are neck to neck in performance in cross-generation/patched titles and we have yet to see any title taking full advantage over any of the two systems. Across all these titles - where the two systems are very close - the PS5 overrall edges out to be slightly ahead (no huge difference but it is there) but there are also titles where the XSX comes out ahead.

Every time there is some sort of XSX benchmark that is ahead of the PS5 the story is 'now all those TFLOPs are coming' - instead of seeing the big picture of two consoles very close to each other and if anything a very slight advantage to the PS5 (so far).

The performance difference between these two should not be a purchase argument for anyone buying these systems. You should just for your self say how much you like gamepass, Sony 1st party games and the two controllers. That is it.

 

[phil_t98]

In the DF video? Time stamp?