Again, if developers optimize for it poorly, that's not the hardware's fault. XSX has more bandwidth available than PS5.
Digital Foundry Article Technical Discussion [2023]
- Thread starter BRiT
- Start date
- Status
There was an interview with a crytek developer in Farsi. I tried searching for it using the wayback machine but, no dice. Thankfully some neogaf dude captured images of the article. This guy did touch on api's a little bit.
Rough Translation of relevant piece: "As a programmer and developer, which do you consider the best console for working and coding? PlayStation 5 or Xbox X series?
Definitely PlayStation 5.
As a programmer, I would say that the PlayStation 5 is much better, and I don't think you can find a programmer who chooses XBX over PS5. For the Xbox, they have to put DirectX and Windows on the console, which is many years old, but for each new console that Sony builds, it also rebuilds the software and APIs in any way it wants. It is in their interest and in our interest. Because there is only one way to do everything, and theirs is the best way possible."
Original Link: https://vigiato.net/p/85342
DF talking about API's ps4 & xb1:
"Digital FoundryDirectX 11 vs GNMX vs GNM - what's your take on the strengths and weakness of the APIs available to developers with Xbox One and PlayStation 4? Closer to launch there were some complaints about XO driver performance and CPU overhead on GNMX.
Oles Shishkovstov
Let's put it that way - we have seen scenarios where a single CPU core was fully loaded just by issuing draw-calls on Xbox One (and that's surely on the 'mono' driver with several fast-path calls utilised). Then, the same scenario on PS4, it was actually difficult to find those draw-calls in the profile graphs, because they are using almost no time and are barely visible as a result.
In general - I don't really get why they choose DX11 as a starting point for the console. It's a console! Why care about some legacy stuff at all? On PS4, most GPU commands are just a few DWORDs written into the command buffer, let's say just a few CPU clock cycles. On Xbox One it easily could be one million times slower because of all the bookkeeping the API does.
But Microsoft is not sleeping, really. Each XDK that has been released both before and after the Xbox One launch has brought faster and faster draw-calls to the table. They added tons of features just to work around limitations of the DX11 API model. They even made a DX12/GNM style do-it-yourself API available - although we didn't ship with it on Redux due to time constraints."
This is just a snippet from the article but, there are more comments here.
Metro Redux: what it's really like to develop for PS4 and Xbox One
As tech interviews go, this one's a corker. Readers of our previous Metro 2033 and Metro Last Light tech Q&As will know…
www.eurogamer.net
These are the ones I remember off the top of my head. If you trawl through some older developer technical presentations, you'll find nuggets in there that hint at it. Since the PS4 days, its been a generally accepted belief that the Playstation Api's were superior and much faster to that of Microsoft's.
Last edited:
Retrospectively can we at least agree how incredible their open world engines are? I am currently having fun with AC Odyssey and the performance is virtually perfect in an incredibly big and seamless open world with high resolution assets everywhere, very few pop in (Horizon!!!), very good AF, a very dense world (NPCs, wildlife) etc.. For me this is the best open world engine currently.
Again it depends of the game and its needs. We have developers (of The Touryst) saying the specific architecture on XSX (hardware, not APIs) prevented them to reach resolution parity against PS5. With games becoming bigger and bigger (notably with RT assets), we might encounter others cases where the memory architecture (and only 10GB of fastest bandwidth) could become an issue with more optimization needed, and we know what that means.
But what those big open world games are showing (notably on PC) is that I/O and assets management is clearly becoming the biggest bottleneck (ignoring shader compilation that is another maybe easier problem on PC). They say it's "CPU bottleneck" but the CPU is usually not that busy. It's I/O (which can mean plenty of different things like API, hardware legacies) bottleneck. And in that area the PS5 is unquestionably much superior than the XSX (and PC). Some games already show us that (Spider-man vs PC), but it was actually already seen in UE5 Matrix demo, PS5 versus XSX generally and specifically in crossroads when there is more assets management.
I couldn't disagree more. Their engine is average at best by AAA standards. Performance is weak for the visuals on offer and scales very poorly with more powerful GPUs. Pop-in is horrendous and all terrain about 30 meters out looks almost as bad as Oblivion from 2006. Grass rendering at the highest settings is about 15 feet. AF doesn't work properly on much of the terrain. Most of the assets are also very far from high resolution. The game utilizes many CPU cores but performance hardly increases. While not necessarily engine related it must be pointed out how awful the animation is.
Games being developed exclusively for one singular console architecture by first party development studios doesn't really speak anything about superiority of one architecture over the other... it speaks more to the fact that the game was designed in a specific way, and that there isn't proper time or resources to ensure the game is as efficient as can be on a different architecture.
Last edited:
I think this has always been the way. Sony's tools since PS3 have just been faster to iterate on.
Edit: In particular, build times. Going off memory, no link, but build times on XB360 could be hours versus minutes (a lot of them though
) on PS3. Only validation I can offer is from experience of XNA development for XBOne in Unity which had epically long build times versus PC and Android, although Android has gotten a lot slower too in recent years. iOS build times were also super long, so I prioritised the platforms that would build fast and create content most efficiently, PC and Android.
Last edited:
davis.anthony
Veteran
I was not aware of this, did they do an interview somewhere? Got a link?
davis.anthony
Veteran
I think this has always been the way. Sony's tools since PS3 have just been faster to iterate on.
Edit: In particular, build times. Going off memory, no link, but build times on XB360 could be hours versus minutes (a lot of them though
Well Mark Cerny is all about that 'time to triangle'
There was an interview with a crytek developer in Farsi. I tried searching for it using the wayback machine but, no dice. Thankfully some neogaf dude captured images of the article. This guy did touch on api's a little bit.
Rough Translation of relevant piece: "As a programmer and developer, which do you consider the best console for working and coding? PlayStation 5 or Xbox X series?
Definitely PlayStation 5.
As a programmer, I would say that the PlayStation 5 is much better, and I don't think you can find a programmer who chooses XBX over PS5. For the Xbox, they have to put DirectX and Windows on the console, which is many years old, but for each new console that Sony builds, it also rebuilds the software and APIs in any way it wants. It is in their interest and in our interest. Because there is only one way to do everything, and theirs is the best way possible."
Original Link: https://vigiato.net/p/85342
DF talking about API's ps4 & xb1:
"Digital FoundryDirectX 11 vs GNMX vs GNM - what's your take on the strengths and weakness of the APIs available to developers with Xbox One and PlayStation 4? Closer to launch there were some complaints about XO driver performance and CPU overhead on GNMX.
Oles Shishkovstov
Let's put it that way - we have seen scenarios where a single CPU core was fully loaded just by issuing draw-calls on Xbox One (and that's surely on the 'mono' driver with several fast-path calls utilised). Then, the same scenario on PS4, it was actually difficult to find those draw-calls in the profile graphs, because they are using almost no time and are barely visible as a result.
In general - I don't really get why they choose DX11 as a starting point for the console. It's a console! Why care about some legacy stuff at all? On PS4, most GPU commands are just a few DWORDs written into the command buffer, let's say just a few CPU clock cycles. On Xbox One it easily could be one million times slower because of all the bookkeeping the API does.
But Microsoft is not sleeping, really. Each XDK that has been released both before and after the Xbox One launch has brought faster and faster draw-calls to the table. They added tons of features just to work around limitations of the DX11 API model. They even made a DX12/GNM style do-it-yourself API available - although we didn't ship with it on Redux due to time constraints."
This is just a snippet from the article but, there are more comments here.
Metro Redux: what it's really like to develop for PS4 and Xbox One
As tech interviews go, this one's a corker. Readers of our previous Metro 2033 and Metro Last Light tech Q&As will know…
These are the ones I remember off the top of my head. If you trawl through some older developer technical presentations, you'll find nuggets in there that hint at it. Since the PS4 days, its been a generally accepted belief that the Playstation Api's were superior and much faster to that of Microsoft's.
I'm not sure if discussions of how the XBO's DX11 based API compare to the PS4's are really that relevant to the Series consoles. Everything is DX12 these days and even the article above references that being much better and suggests it's more in line with GNM.
And yet Forsaken loads faster on PC with a similar SSD. CPU bottlenecks are absolutely a factor here. Just because the overall CPU doesn't show high utilisation does not mean a single thread isn't bottlenecking performance. We have very well documented sources explaining just how much IO is limited by the CPU on both consoles and PC. And on PC, there is often simply more to do at load time than there is on the console, hence why load times can be slower, In Spidermans case for example the developers themselves specifically said shader complication and (I think) BHV building were both an additional CPU burden on the PC side vs the PS5.
Yea that history is there for sure. They’ve come a long way from mono driver. DX12There was an interview with a crytek developer in Farsi. I tried searching for it using the wayback machine but, no dice. Thankfully some neogaf dude captured images of the article. This guy did touch on api's a little bit.
Rough Translation of relevant piece: "As a programmer and developer, which do you consider the best console for working and coding? PlayStation 5 or Xbox X series?
Definitely PlayStation 5.
As a programmer, I would say that the PlayStation 5 is much better, and I don't think you can find a programmer who chooses XBX over PS5. For the Xbox, they have to put DirectX and Windows on the console, which is many years old, but for each new console that Sony builds, it also rebuilds the software and APIs in any way it wants. It is in their interest and in our interest. Because there is only one way to do everything, and theirs is the best way possible."
Original Link: https://vigiato.net/p/85342
DF talking about API's ps4 & xb1:
"Digital FoundryDirectX 11 vs GNMX vs GNM - what's your take on the strengths and weakness of the APIs available to developers with Xbox One and PlayStation 4? Closer to launch there were some complaints about XO driver performance and CPU overhead on GNMX.
Oles Shishkovstov
Let's put it that way - we have seen scenarios where a single CPU core was fully loaded just by issuing draw-calls on Xbox One (and that's surely on the 'mono' driver with several fast-path calls utilised). Then, the same scenario on PS4, it was actually difficult to find those draw-calls in the profile graphs, because they are using almost no time and are barely visible as a result.
In general - I don't really get why they choose DX11 as a starting point for the console. It's a console! Why care about some legacy stuff at all? On PS4, most GPU commands are just a few DWORDs written into the command buffer, let's say just a few CPU clock cycles. On Xbox One it easily could be one million times slower because of all the bookkeeping the API does.
But Microsoft is not sleeping, really. Each XDK that has been released both before and after the Xbox One launch has brought faster and faster draw-calls to the table. They added tons of features just to work around limitations of the DX11 API model. They even made a DX12/GNM style do-it-yourself API available - although we didn't ship with it on Redux due to time constraints."
This is just a snippet from the article but, there are more comments here.
Metro Redux: what it's really like to develop for PS4 and Xbox One
As tech interviews go, this one's a corker. Readers of our previous Metro 2033 and Metro Last Light tech Q&As will know…
These are the ones I remember off the top of my head. If you trawl through some older developer technical presentations, you'll find nuggets in there that hint at it. Since the PS4 days, its been a generally accepted belief that the Playstation Api's were superior and much faster to that of Microsoft's.
Has been a thing on Xbox approaching now a decade and mono driver deprecated. The first comment from the crysis developer is still fairly early days, Xbox was arguably a year behind ps5 in terms of tool maturity.
I was hoping for some more recent news about the situation. The api situation is always changing, I’m general biased to not use information that old to represent the games of today. Unless there is something architecturally wrong with dx12.
GNM will always have the advantages of getting a little lower, but I don’t know how that makes stutter go away as a result. As I understand it, GNM or Vulcan is prone to this issue, so I still firmly believe it’s just the available time to optimize.
D
Deleted member 11852
Guest
In itself, this boggles my mind. Microsoft are the biggest software company in the world and also make the most popular IDE environment - which Sony also use in PlayStation SDK. Just.. how...?
Inuhanyou
Veteran
Because Sony is a console dev. They only care about the hw they are using and nothing more. It's a pro and a con.In itself, this boggles my mind. Microsoft are the biggest software company in the world and also make the most popular IDE environment - which Sony also use in PlayStation SDK. Just.. how...?
Xbox from the start has been an extension of PC environment, literally direct x box. Due to this the abstraction layer allows compatibility with each machine.
Sony focuses on each individual piece of hw and maximizes it's potency. So we get to the metal coding but it's extremely difficult for them to carry things forward seamlessly.
I don't need to tell anyone here that obvious information, but it really matters because it shows MS coast on the work they do for PC a lot of the time and will get it working only as much as it needs to for console.
That's not a dunk on their software programmers either for PC or console. It's just that both companies are coming from completely different places when it comes to how they design content for their gaming ecosystem even if the result is supposed to be the same
Last edited:
Sony was trying to release a little earlier and MS had to wait for later because RDNA2 contained the feature set that MS was pushing on PC for DX12U.In itself, this boggles my mind. Microsoft are the biggest software company in the world and also make the most popular IDE environment - which Sony also use in PlayStation SDK. Just.. how...?
Also probably doesn’t help to have the memory architecture the way it is and a second console as well, all wrapped up into a single piece of devkit.
It’s efficient from some perspectives, from another it’s just added weight they probably didn’t need.
Sony uses SN Systems. This was a company created in 1990 to develop video game build tools specifically, headed by video=game developers who know what they want from build tools and only have ot create those. They have 30 years experience. Versus MS who specialises in broad-spectrum OS development followed by trying to get that OS to do everything on muddling hardwares.In itself, this boggles my mind. Microsoft are the biggest software company in the world and also make the most popular IDE environment - which Sony also use in PlayStation SDK. Just.. how...?
D
Deleted member 11852
Guest
Series X has weird memory and PS5 has weird clocks. Unless Microsoft are making different tools for Series S and Series X then two consoles shouldn't make any difference to the tools.
It's my understanding that SN Systems produce PlayStation-specific tools and hardware profilers, whilst, the PlayStation SDK IDE is VisualStudio. Whenever Insomniac or Guerrilla Games show snippets of their dev environment, it's VS. It's also my understanding that Sony include a lot of third party tools in their toolchain, including Yebis, Xaitcontrol, and Enlighten, along with third party middleware toolkits which are industry standard.
davis.anthony
Veteran
According to a few people in the industry on Twitter that was a last minute change.
VS is the IDE but the build tools are SN. You write the code in VS, it's built on SN's platform (compiler, linker, distributed build system, etc).
D
Deleted member 11852
Guest
The PlayStation linker/compiler has been the open source LLVM/Clang combo since PS4 and still is. Let's not oversubscribe what SN Tools do and what they don't. PS5's devkit has a bunch of third party tools and middleware.
I'm not trying to overstate. I'm trying to explain why PS development can be faster than XB development even using the same IDE.
- Status
Similar threads
