DF: A Plague Tale: Requiem PC vs. Consoles Graphics Comparison

[Old Gamer]

Well that goes for several 3rd party companies - and yet - it's these very engines that do the most in terms of graphics features. Sony companies have had the PS4/PS5 for a bit now and the performance/graphics features on 1st party exclusive are the same as the 3rd party engines. We don't see any advantage when comparing their performance metrics to 3rd party company performance.

Of course it goes for any 3rd party company - what one can conclude in any given scenario is how well a particular engine implementation performed for given hardware + SDK configurations. If you want to benchmark console APUs, port and run PC benchmarks on the consoles.

One cannot take a frame time ratio in one game and say "of course, it's because that GPU is that much percent slower" - that kind of statement is naive at best and maliciously deceptive at worse.

I would also like to point out how often Sony first party games appear to underperform on PC when ported, even though PC hardware is potentially much more capable - why would that be? Could it be that porting a game engine originally designed for GNM to Direct3D can also result in inneficient code?

So, you see, what I said also goes for 1st party games. Any game.

Game Engines are not hardware benchmarks.

 

[Deleted member 13]

Spiderman Miles Morales running at 80fps with Ray Tracing at 1440p? I'd love to see a third party game pulling that off with decent visual fidelity.

RT reflections is very light on the GPU. That's why we normally see those in games as opposed to the more expensive RTGI, RT shadows, RT ambient occlusion, RT area lights, etc..

Not trying to be argumentative but Spiderman runs on a PS4. Are you saying that their graphics engine somehow extracts more GPU power than 3rd party companies like Epic with UE?

 

[Deleted member 13]

Of course it goes for any 3rd party company - what one can conclude in any given scenario is how well a particular engine implementation performed for given hardware + SDK configurations. If you want to benchmark console APUs, port and run PC benchmarks on the consoles.

I have no issues with that but surely we can't throw out the mass majority of games and comparisons just because they aren't specifically written for one particular hardware SKU.

One cannot take a frame time ratio in one game and say "of course, it's because that GPU is that much percent slower" - that kind of statement is naive at best and maliciously deceptive at worse.

Of course. DF was just pointing out the difference just happens to be an interesting correlation compared to the hardware specs of both systems.

I would also like to point out how often Sony first party games appear to underperform on PC when ported, even though PC hardware is potentially much more capable

Underperform meaning for similar hardware (i.e. 2080, 70, etc..) with the same graphics settings that the performance is below the PS5 version?

Could it be that porting a game engine originally designed for GNM to Direct3D can also result in inneficient code?

Sure that can be the case but I wouldn't call the code from 1st party graphics engines perfectly efficient code and the 3rd party developers to be incompetent in making the best the console can push out. iDSoftware is a perfect example of an incredibly efficient graphics engine across all platforms. I trust when they say they pushed the consoles to their limits, I don't doubt it.

Game Engines are not hardware benchmarks.

I don't think DF was saying that. I'm not saying that either. But you CAN benchmark a game and compare the results like DF. It has been this way for years with people not really arguing the validity of the benchmarks.

Are you getting skeptical at the article made by DF over a 3rd party game where the Xbox wins out by a small 25% because you assume that the PS5 should be more performant than the Xbox in any scenario?

 

[thelastword]

It's exactly where the XSX and PS5 hardware architecture reflects though. The XSX is indeed, 25% more powerful and the FPS reflect that.

I do agree though that they know the PC more due to FS2020 which took several years of development.

The XBOX is not more powerful, it has strengths, but PS has strengths too, perhaps a bit more than XBOX. So if a game really utilizes the PS5 it will load faster, have better textures, it will also have the advantage with lots of alpha onscreen and at high refresh rates (the latter due to faster clocks). To make this clear, the Series S is performing better than PS5 here, I guess the Series S is more powerful over PS5 too?

Well that goes for several 3rd party companies - and yet - it's these very engines that do the most in terms of graphics features. Sony companies have had the PS4/PS5 for a bit now and the performance/graphics features on 1st party exclusive are the same as the 3rd party engines. We don't see any advantage when comparing their performance metrics to 3rd party company performance.

Comparing third parties to first is so outlandish. There is a huge divide currently between first and third party and it will extend some more when the true next gen (built ground up for PS5 games launch). There have been a few already, but these were first stabs at the PS5. Demon Souls Remake and Ratchet and Clank are two games that have not be rivalled by any third party this generation. Also we have a quite a few cross gen PS games that are unrivalled relative to visuals

Horizon Forbidden West, God of War, Miles Morales, Gran Turismo 7 along with Demon Souls and Ratchet are at the top of visuals this generation so far. A Plague's tale is really not as impressive near Forbidden West or God of War or any of these games mentioned. The only third party game I'm willing to give some props is Calisto Protocol, but we will see how it looks and performs on consoles.

 

[Darth Vader]

RT reflections is very light on the GPU. That's why we normally see those in games as opposed to the more expensive RTGI, RT shadows, RT ambient occlusion, RT area lights, etc..

Not trying to be argumentative but Spiderman runs on a PS4. Are you saying that their graphics engine somehow extracts more GPU power than 3rd party companies like Epic with UE?

So light that plenty of "3rd party" games are not pulling them off. Even 1st party studios have not leveraged it that much if at all. On PS consoles, Insomniac seems to be the only one that has been able to get decent performance so far. What I'm saying is that the game runs with pristine image quality and ray tracing at 1440p and 80fps or higher.

Are you getting skeptical at the article made by DF over a 3rd party game where the Xbox wins out by a small 25% because you assume that the PS5 should be more performant than the Xbox in any scenario?

In some scenarios the Xbox is more performant, in others the PS5 is more performant. Funny that you feel the need to only point out one of them, even going as far as saying that it's expected because the Xbox is more powerful by 25% (which it isn't). But I'm sure you're just being unbiased.

The XBOX is not more powerful, it has strengths, but PS has strengths too, perhaps a bit more than XBOX. So if a game really utilizes the PS5 it will load faster, have better textures, it will also have the advantage with lots of alpha onscreen and at high refresh rates (the latter due to faster clocks). To make this clear, the Series S is performing better than PS5 here, I guess the Series S is more powerful over PS5 too?

I don't think we've seen 3rd party games with better textures on the PS5, but I'm willing to be proven wrong. We have sen better alpha, better performance overall (since PS games are usually vsynced and Xbox ones aren't), etc.

Horizon Forbidden West, God of War, Miles Morales, Gran Turismo 7 along with Demon Souls and Ratchet are at the top of visuals this generation so far. A Plague's tale is really not as impressive near Forbidden West or God of War or any of these games mentioned. The only third party game I'm willing to give some props is Calisto Protocol, but we will see how it looks and performs on consoles.

Plague Tale looks really good, but after watching plenty of 4k ultra videos, I don't see anything overly special that puts it head and shoulders above HFW (which is open world) or Demon's Souls (which is not). It is a poorly optimized game, and with team Asobo having a lot more experience with the Series X and S due to their work with FS it's only natural that that platform presents the best and more stable performance.

As for performance overall, as I've pointed at some point before, when you don't get a massive jump in fps between low settings and ultra settings, either your engine scales like shit or your game is not properly optimized.

Comparisons

Plague tale

Again: It's a gorgeous game that's been poorly optimized, even on PC.

PS: I want to see if DF will come back to console comparisons every single update to see if the performance gap shortens. Sure they will

 

[thelastword]

I don't think we've seen 3rd party games with better textures on the PS5, but I'm willing to be proven wrong. We have sen better alpha, better performance overall (since PS games are usually vsynced and Xbox ones aren't), etc.

I think Valhalla had better textures and some light shafts too. I think Dirt had better textures and more detail overall on PS5. I don't know if some of these were reverted because some games actually got worse performance and downgraded graphics on PS5 with updates, that is when the XBOX version was not up to par. I think there may be others but since I have not been on GAF with more active comparison threads like in it's hey day, I've not been keeping up with the comparisons as such.

 

[Deleted member 13]

The XBOX is not more powerful, it has strengths, but PS has strengths too, perhaps a bit more than XBOX.

I'm not wanting to get into a hardware debate about these consoles. We all know what they can do and can't do. When I say more powerful, I'm talking about the TFLOPS in raw rasterizing. Of course PS5 has strengths as well as you mentioned.

Comparing third parties to first is so outlandish. There is a huge divide currently between first and third party and it will extend some more when the true next gen (built ground up for PS5 games launch). There have been a few already, but these were first stabs at the PS5. Demon Souls Remake and Ratchet and Clank are two games that have not be rivalled by any third party this generation

Thats your opinion and not a fact. There is nothing in either of those games that will make Xbox perform less or subtract graphics features. They are both using the typical rendering features found in last gen. The SSD level-switching is the only thing I would say is in the PS5's favor. And I'm not knowledgable enough on memory cost/bandwidth of the level switching algorithm to say it can't be done on an Xbox.

Horizon Forbidden West, God of War, Miles Morales, Gran Turismo 7 along with Demon Souls and Ratchet are at the top of visuals this generation so far.

They are all "top of the visuals" but use the same rendering techniques every other 3rd party studio does. In other words, you just made a subjective comment. Some 3rd parties like Asobo go beyond what those games have done for rendering for example - parallax occlusion mapping (very expensive) + normal mapping + displacement mapping. Those 3 texture layers combined are extremely expensive and where most of the cost of that games' FPS went into. The amount of detail seen on the rocks and environments rival UE5 demo's Nanite.

A Plague's tale is really not as impressive near Forbidden West or God of War or any of these games mentioned. The only third party game I'm willing to give some props is Calisto Protocol, but we will see how it looks and performs on consoles.

Your opinion.

 

[thelastword]

I'm not wanting to get into a hardware debate about these consoles. We all know what they can do and can't do. When I say more powerful, I'm talking about the TFLOPS in raw rasterizing. Of course PS5 has strengths as well as you mentioned.


Thats your opinion and not a fact. There is nothing in either of those games that will make Xbox perform less or subtract graphics features. They are both using the typical rendering features found in last gen. The SSD level-switching is the only thing I would say is in the PS5's favor. And I'm not knowledgable enough on memory cost/bandwidth of the level switching algorithm to say it can't be done on an Xbox.


They are all "top of the visuals" but use the same rendering techniques every other 3rd party studio does. In other words, you just made a subjective comment. Some 3rd parties like Asobo go beyond what those games have done for rendering for example - parallax occlusion mapping (very expensive) + normal mapping + displacement mapping. Those 3 texture layers combined are extremely expensive and where most of the cost of that games' FPS went into. The amount of detail seen on the rocks and environments rival UE5 demo's Nanite.


Your opinion.

You brought up 25% more powerful, so you wanted to get that point across even though in on screen visuals that means nothing and PS has areas (in hardware) that are better than 25% better and has shown games with better resolution and framerate too, better effects/alpha, better 120fps modes, that's not opinion.

So of course using A Plague Tale Requiem as proof of power is disingenuous, because I surely don't see you doing that when PS5 has the better performing games.

So just answer me this, if this is about power and PS is being properly utilized, why is the S beating the PS5? Do not pass GO before you answer this question...

 

[Deleted member 13]

You brought up 25% more powerful, so you wanted to get that point across even though in on screen visuals that means nothing and PS has areas (in hardware) that are better than 25% better and has shown games with better resolution and framerate too, better effects/alpha, better 120fps modes, that's not opinion.

I mentioned 25% because that's what DF showed in this particular test for this particular thread. I have no financial stake in the consoles. It doesn't matter to me one bit which one outperforms the other. Your feelings shouldn't be hurt because of what they showed in the test and then take it out on me like I made the video. It's way too emtional for my liking.

So just answer me this, if this is about power and PS is being properly utilized, why is the S beating the PS5? Do not pass GO before you answer this question...

I have no idea why the S is outperforming the PS5 in this particular game. It could very well be that their code is catered to MS DX API and their porting skills aren't ideal. I won't defend them over that at all because I just don't know. But I *do* know what their game is doing as far as graphics features are concerned. I can argue that all day long.

 

[Darth Vader]

And as I had said... This game is poorly optimized. One patch later, runs much better across the board.

 

[Deleted member 13]

And as I had said... This game is poorly optimized. One patch later, runs much better across the board.

There was never any question of it needing some optimization but "poorly" optimized "for consoles" - and specifically the PS5 - was a very hyperbolic statement. Secondly, the engine needed optimization on all platforms and not just one therefore all of the platforms performed on equal footing (i.e. meaning if it ran slower on one system, it would run slower on the other system). Some people mentioned the developers weren't knowledgable enough on the PS SDK to make it perform better and lent more towards DX12. That was also a false statement. These developers don't need to bang their head up against a wall to extract out power from the PS. We aren't in the days of PS3 where the architecture was completely unorthodox.

 

[Darth Vader]

There was never any question of it needing some optimization but "poorly" optimized "for consoles" - and specifically the PS5 - was a very hyperbolic statement. Secondly, the engine needed optimization on all platforms and not just one therefore all of the platforms performed on equal footing (i.e. meaning if it ran slower on one system, it would run slower on the other system). Some people mentioned the developers weren't knowledgable enough on the PS SDK to make it perform better and lent more towards DX12. That was also a false statement. These developers don't need to bang their head up against a wall to extract out power from the PS. We aren't in the days of PS3 where the architecture was completely unorthodox.

It was poorly optimized for consoles and PC, as I had said, proven exactly by this video. And I love that you call others liars, yet the only platform being tested that is not reliant on DX12 is... the PS5. So your argument that "it's a false statement" is at most as valid as mine saying "it's a true statement".

But hey, let's tell this to developers working with apps and games for Intel Mac machines and Windows machines. The baseline is the same (X86), so the SDKs should be exactly the same, RIGHT?

 

[Old Gamer]

And as I had said... This game is poorly optimized. One patch later, runs much better across the board.

Surprising to say the least, as given the developer history I would say specialization of skill set had more of a hand on it than time constraints, but there it is... they just didn't have the time to optimize before launch.

As for what are lies and misinformation, I will get to that in a moment...

 

[Hezekiah]

Wait Series X is 25% more powerful now, not 18%? Damn they've been hiding it well this gen it has to be said.

 

[thelastword]

Wait Series X is 25% more powerful now, not 18%? Damn they've been hiding it well this gen it has to be said.

No doubt, I never really understood how the GTX 1080 based PS5, has kept up so well with the RDNA 2 beast....

 

[Deleted member 13]

It was poorly optimized for consoles and PC, as I had said, proven exactly by this video.

My issue with your comments is the word "poorly" optimized. It's status quo for games to get a little bit more juice out or a patch that already has the optimizations. We saw this with H:FW - a game strictly for the PS consoles. The game wasn't so dramatically unoptimized that there was tons of performance on hand. That's my beef. They extracted a few FPS to get it working constantly but that doesn't mean added rendering features and 60FPS was possible. The talk is that the consoles have 50% or more juice in them when they are literally equivalent to 2080s GPUs. It's you holding on to the belief of the old days - that consoles are some black box that needs 6yrs to master in order for you to believe that all the performance has been squeezed out of it. There literally is 0 evidence of this.

And I love that you call others liars,

I never call anyone a liar bro.. That's a really really harsh word to use. A statement being 'inaccurate'? Sure.

yet the only platform being tested that is not reliant on DX12 is... the PS5. So your argument that "it's a false statement" is at most as valid as mine saying "it's a true statement".

All I'm saying is that the PS5/XSX are pretty much a known by the developers of what they can peak out. Let's not make this about them being dumb and don't know how to code properly. Plague is doing some cool stuff that no other games have tried and the results show for themselves. They are the first company to come up with a brute force approach to getting highly detailed environments that rival Nanite. And it should be no suprise as FS2020 is still, imo, the best looking next-gen game/sim since this generation came out. That sim is doing some crazy stuff that I've only seen in making CG films (i.e. texture synthesis).

File:Imagequilting.gif - Wikipedia

en.wikipedia.org

 

[Darth Vader]

y issue with your comments is the word "poorly" optimized. It's status quo for games to get a little bit more juice out or a patch that already has the optimizations.

Fact: The game is poorly optimized, proven by bumps of over 20fps after the update. And there's still probably performance on the table in all platforms.

The game wasn't so dramatically unoptimized that there was tons of performance on hand. That's my beef. They extracted a few FPS to get it working constantly but that doesn't mean added rendering features and 60FPS was possible.

Why are you talking about Forbidden West? Is that your way to try and strawman this?

The talk is that the consoles have 50% or more juice in them when they are literally equivalent to 2080s GPUs.

Funny enough... Most people have said this. You're just trying to deflect to an already established argument. Good job.

hat consoles are some black box that needs 6yrs to master in order for you to believe that all the performance has been squeezed out of it. There literally is 0 evidence of this.

Nobody said this. But the longer you get into a gen the more performance is squeezed from a console. That's a known fact and has been so for generations. SDKs improve, firmware improves, developers find more and more clever ways to spare a few precious ms... This is not rocket science.

I never call anyone a liar bro.. That's a really really harsh word to use. A statement being 'inaccurate'? Sure.

Some people mentioned the developers weren't knowledgable enough on the PS SDK to make it perform better and lent more towards DX12. That was also a false statement.

Sure

All I'm saying is that the PS5/XSX are pretty much a known by the developers of what they can peak out. Let's not make this about them being dumb and don't know how to code properly.

Thats not what I said.

 

[Deleted member 13]

Fact: The game is poorly optimized, proven by bumps of over 20fps after the update.

The game did NOT receive bumps of 20FPS across the entire game under ALL situations. LOL! You took the worst case scenario to argue?

Why are you talking about Forbidden West? Is that your way to try and strawman this?

Trying to get you to admit that games come out not as optimized as it could be and a company which you highly regard as company (that knows the PS hardware more than 3rd party companies because of their experience with the platform) had to do the same thing.

Let's do this. Do you think the developers have even more performance to extract out to where it can run 1440p or even native 4k/60FPS like the high end GPUs? If not, then why keep bashing the game?

 

[Darth Vader]

The game did NOT receive bumps of 20FPS across the entire game under ALL situations. LOL! You took the worst case scenario to argue?

I did not say in all situations. Can you argue in good faith for once? Or are you going to continue to try and get some gotcha and fail miserably?

Trying to get you to admit that games come out not as optimized as it could be. Let's do this. Do you think the developers have even more performance to extract out to where it can run 1440p or even native 4k/60FPS like the high end GPUs? If not, then why keep bashing the game?

No. But I never made that argument, I think?

And I'm not bashing the game, I'm bashing the performance.

 

[Deleted member 13]

And I'm not bashing the game, I'm bashing the performance.

To me that's bashing the developer and what they put out. They clearly have pushed graphics to the next-level and that's with a custom proprietary engine. The guys should be highly regarded along with the 1st party developers that you guys respect.