Support NeoGAF

[DirgeExtinction]

I'm not sure if it was just a rumor or not:
I've heard many times during this generation that MS wouldn't let multiplat games come to their console if the quality/graphics weren't on par with the competitor's console version. Is that true?

Pretty much everything since E3 has been shown on Xbox One kits. There was even youtube footage of a kid playing Ryse in his home. Microsoft has had a few in-store demos using dev kits. They're not hiding anything.

Except Titanfall, I believe. That game has only been shown running on PCs so far.

 

[IAmRandom31]

Hey guys! Newb here, be gentle.

Anyway, being in the tech world for the past 25 years and being a part of the gaming world for just as long, I have been taught a few times.

A.) What you see is what you get

B.) Console developers are wizards

C.) PC developers are generally lazy ( or extremely held back due to the nature of their engines having to talk to 100 different architectures )

In my opinion, I think the PS4 will have the noticeable edge in multiplatform titles. Mainly due to how easy it will be for developers to port the original code over from the main development platform on the PC and slap it into the devkit and then test. Will be very, very easy for them to do that. Due to that ease, the devs will probably have an additional 2-3 months or more of pure optimization time with the code on the devkit environments. This will help even third party devs toy around with the PS4 more. Who knows, maybe the third party devs will even begin to play with GPCPU on the PS4 by the 2nd generation, and not the 4th like Cerny thinks. Extra development time provided by ease of development is extremely positive.

On the X1 side of multiplatform titles, it will be a bit trickier. Especially if MS does not build VERY good API's that will allow developers to tap into the functionality of the eSRAM with alot of ease and simplicity. If MS builds fantastic API's, then you will begin seeing developers put the eSRAM to much more use. But if they have to build the structure themselves, well .. at first, you won't have much use from the eSRAM.

Even then though, the PS4 just has far more raw GPU power. 32 rendering processors compared to 16 is a major real world difference. Being able to feed 32 processors to render your output means well .. pretty simply that the PS4 will be able to render a scene twice as fast. But added in with the memory bandwidth from the GDDR5, you have the PS4 being fed more information faster and processing that information twice as fast to the television.

X1 exclusives that are built from the ground up utilizing the eSRAM's capabilities will look much better in comparison to multiplatform titles. Especially if the group at the helm is highly proficient like Crytek.

Thats my opinion on the matter. And as a newb, please rip into me if I am completely off base.

 

[Chobel]

You are both putting a lot of words in my mouth, please don't.

I think(know) the PS4 is graphically more powerful but I think that this will result in minimal differences largely unnoticeable to the untrained eye. So basically what developers have said.

40% could mean one game would run 1080p in PS4 and 900p in Xbox One.
You need a trained eye in order to not notice that difference.

 

[turk3y]

How much of the imagery for the XBO out there is from actual XBO hardware and not the pc dev kits? Can anyone point me to images or video that they know for sure comes from XBO hardware and not the PC kits we saw at E3?

We saw pc kits for a single digital download game that would be classed as arcade should that still exist. The only other iffy title was BF4 that was on console spec pc which probably means a beast of a pc running a cut down graphics setting pc build to show off the target similar to watchdogs at it's reveal.

Almost everything else has been Xbox or clearly marked as pc's to the console spec.

Pretty sure xbox's 12 gpu cores are more than enough to push 1080 along happily, the extra cores on the ps4 are blatantly for compute which is where the 14 4 rumour comes in, probably about what later titles will aim for after all this is what Sony have predicted and customised their hardware for.

 

[TechnicPuppet]

40% could mean one game would run 1080p in PS4 and 900p in Xbox One.
You need a trained eye in order to not notice that difference.

I'd be fine with that I think. I would need to test it though to be sure.

 

[Jonm1010]

You are both putting a lot of words in my mouth, please don't.

I think(know) the PS4 is graphically more powerful but I think that this will result in minimal differences largely unnoticeable to the untrained eye. So basically what developers have said.

I think the one thing we have pretty well established is that developers AREN'T in lockstep agreement about that.

However what you are saying is pretty definitive for relying on assumption. Kind of silly if you ask me. Kind of irrational too.

All anyone can say for certain is what the specs say and infer based on the developer feedback. The really only logical conclusions to draw from it all is that a power difference exists and its either going to be slight or negligible depending on the title up to noticeable and telling.

And that everything should be parsed out in a year or so when developers start really targeting these systems and understanding their strengths. However it is undeniable that the ps4 is the more powerful console.

 

[Chobel]

We saw pc kits for a single digital download game that would be classed as arcade should that still exist. The only other iffy title was BF4 that was on console spec pc which probably means a beast of a pc running a cut down graphics setting pc build to show off the target similar to watchdogs at it's reveal.

Almost everything else has been Xbox or clearly marked as pc's to the console spec.

Pretty sure xbox's 12 gpu cores are more than enough to push 1080 along happily, the extra cores on the ps4 are blatantly for compute which is where the 14 4 rumour comes in, probably about what later titles will aim for after all this is what Sony have predicted and customised their hardware for.

Seriously this needs to stop. 14+4 split was debunked.

 

[Finalizer]

I'm not sure if it was just a rumor or not:
I've heard many times during this generation that MS wouldn't let multiplat games come to their console if the quality/graphics weren't on par with the competitor's console version. Is that true?

FF13 is a well known case of the opposite, but I don't know of any others.

The more likely scenario is that a majority of developers simply used the 360 as the lead platform thanks to it being easier to develop for and, up until recently, having much more units out in the wild.

 

[Septimius]

So we should only believe anonymous developers or 2nd party Sony developers who more than likely don't even have an XB1 devkit. Got it.

Being skeptical is good. Being oblivious as a way to overlook something that seems to be obvious, however, isn't.

 

[Wishmaster92]

This is the visual difference i expect to see

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.

 

[helvetica bold]

Well, seeing as how the discussion originated with a statement comparing the graphics of the games in question, it actually is all about the graphics. It's true that they've made improvements to other areas of the game (the AI being generated via cloud processing using data from actual players has a lot of potential for this and future games), but the presentation still relies on old tricks. Using DX 11 is all fine and good, but it doesn't mean much when you're not taking advantage of all that it has to offer. Is there any part of the lighting system that's actually dynamic?

This article plains some of the new Forza 5 graphic tech, a lot has be done with materials.
http://www.nowgamer.com/features/19...dware_building_tracks_drivatar_explained.html

Forza 4 used HDR Image Based lighting so drive club using that it's nothing new.

My favorite racing game NFS Hot Pursuit (2010) had image based lighting.
http://www.eurogamer.net/articles/digitalfoundry-needforspeed-tech-interview?page=2

"Digital Foundry: What are the principles? How is it applied in Need for Speed?

Alex Fry: Image-based lighting basically means that the cars are lit by the image of the environment around them. So if you put that car anywhere in the world, it will be lit correctly. You haven't got to fake it, you haven't got to bake it. The cars are totally real-time image-based lit. That's one of the reasons they look good.

When you drive into a tunnel we haven't got to place any probes to capture some light here, capture some light there, it's totally, totally dynamic. So you're driving into a tunnel and you start to see the car get dark at the front while remaining bright at the back. You drive into the tunnel and the light will roll over, and then you go underneath some spotlights in the tunnel and you'll see these beautiful specular soft lights roll over the car.

You'll get to the open side and one side of the car will start to roll into the light and you'll see all these different frequencies like the really sharp reflection of the lacquer over the top to this sort of blurred diffuse underlying paint beneath it and you'll see all of these different layers building themselves up.

It's all totally real-time and it's one of the things we're most pleased about with this game: the cars had to look amazing at all times of the day, anywhere in that world. Image-based lighting then... it means that the cars light themselves based on the image around them."

Lastly IGN mentions subsurface scattering and HDR lighting
http://www.ign.com/articles/2013/06/12/e3-2013-forza-motorsport-5-hands-on

 

[mrklaw]

You are both putting a lot of words in my mouth, please don't.

I think(know) the PS4 is graphically more powerful but I think that this will result in minimal differences largely unnoticeable to the untrained eye. So basically what developers have said.

The fundamental problem is: although there is clearly a technical advantage to PS4, we do not have enough information to say how much. Therefore both of these statements are equally valid at this time:

"I think differences will be so small that they will be unnoticeable to most people"
"I think the differences will be immediate and obvious to anyone with an interest in games"

I think that is the crux of the argument - without any evidence to the actual quantifiable gap, we'll just go round and round in circles and nobody will be satisfied.

Basically I don't see us getting anywhere until November 22nd and the first time we see Multiplatform games on both machines (and even then it isn't clear because of vagaries of cross gen, inefficient initial games etc)

 

[TechnicPuppet]

I think the one thing we have pretty well established is that developers AREN'T in lockstep agreement about that.

However what you are saying is pretty definitive for relying on assumption. Kind of silly if you ask me. Kind of irrational too.

All anyone can say for certain is what the specs say and infer based on the developer feedback. The really only logical conclusions to draw from it all is that a power difference exists and its either going to be slight or negligible depending on the title up to noticeable and telling.

And that everything should be parsed out in a year or so when developers start really targeting these systems and understanding their strengths. However it is undeniable that the ps4 is the more powerful console.

There are 70 pages here so it really isn't that simple is it. In two months a picture will start to emerge anyway and this will all be a distant memory.

 

[cyberheater]

Even then though, the PS4 just has far more raw GPU power. 32 rendering processors compared to 16 is a major real world difference. Being able to feed 32 processors to render your output means well .. pretty simply that the PS4 will be able to render a scene twice as fast.

Xbox One GPU is 1.3 Tflops. PS4 is 1.79Tflops.
The PS4 GPU isn't twice as fast as the Xbox One GPU.

 

[Chobel]

This is the visual difference i expect to see

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.

That's two different games, each one using different techniques and created by different teams. Have the same team produce the same game maximising the power of each console, then let's talk if we can see significant difference or not.

 

[Wishmaster92]

That's two different games, each one using different techniques and created by different teams. Have the same team produce the same game maximising the power of each console, then let's talk if we can see significant difference or not.

Yes multiplats, let's wait for those. Just comparing since both are exclusives.

 

[nib95]

Xbox One GPU is 1.3 Tflops. PS4 is 1.79Tflops.
The PS4 GPU isn't twice as fast as the Xbox One GPU.

PS4 GPU is 1.84 Tflops. This has been known for the longest time. Where on Earth did you get the 1.79 figure from?

Here's how it stacks up.

PS4 | GPU: 1.84 Tflops and CPU: 100 Glops, Total 1.94 Tflops

Xbox One | GPU: 1.31 Tflops and CPU: 109 Gflops, Total 1.41 Tflops

PS4 based on raw performance is 38% more powerful than the Xbox One.


GPU's directly compared.

PS4: 1.84TF GPU ( 18 CUs)
PS4: 1152 Shaders
PS4: 72 Texture units
PS4: 32 ROPS
PS4: 8 ACE/64 queues
8gb GDDR5 @ 176gb/s

Verses

Xbone: 1.31 TF GPU (12 CUs)
Xbone: 768 Shaders
Xbone: 48 Texture units
Xbone: 16 ROPS
Xbone: 2 ACE/ 16 queues
8gb DDR3 @ 69gb/s+ 32MB ESRAM @109gb/s

 

[Jonm1010]

The fundamental problem is: although there is clearly a technical advantage to PS4, we do not have enou information to say how much. Therefore both of these statements are equally valid at this time:

"I think differences will be so small that they will be unnoticeable to most people"
"I think the differences will be immediate and obvious to anyone with an interest in games"

I think that is the crux of the argument - without any evidence to the actual quantifiable gap, we'll just go round and round in circles and nobody will be satisfied.

Basically I don't see us getting anywhere until November 22nd and the first time we see Multiplatform games on both machines (and even then it isn't clear because of vagaries of cross gen, inefficient initial games etc)

I'm not even sure November is gonna answer the question. A lot of the games releasing are games first designed for last gen hardware and/or are games that have not had a lot of time spent on development on final hardware and thus likely aren't going to be taking full advantage of the hardware.

I think the real differences, if they exist to the level speculated, are more likely to start rolling out on the next round of games when ground up multiplatform games with a lot of development time on the new hardware begin to trickle out.

 

[cyberheater]

This is the visual difference i expect to see

http://i4.minus.com/ibmcPEgLpPfAax.png
http://i5.minus.com/ib0Sp0omcDMu1C.jpg

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.

I think Ryse looks better. Both are fantastic looking games.

 

[mrklaw]

I'm not even sure November is gonna answer the question. A lot of the games releasing are games first designed for last gen hardware and/or are games that have not had a lot of time spent on development on final hardware and thus likely aren't going to be taking full advantage of the hardware.

I think the real differences, if they exist to the level speculated, are more likely to start rolling out on the next round of games when ground up multiplatform games with a lot of development time on the new hardware begin to trickle out.

Oh sure. For real console warriors there will be enough to play with to argue your corner. But I think at least it'll give the majority of us a decent view. Full understanding will take a couple of year probably though

 

[Gemüsepizza]

This is the visual difference i expect to see

http://i4.minus.com/ibmcPEgLpPfAax.png[IMG]
[IMG]http://i5.minus.com/ib0Sp0omcDMu1C.jpg[IMG]

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.[/QUOTE]

Those aren't real screenshots from the game. What's the point in comparing 3D models from a dev enviroment?

[quote="cyberheater, post: 81968537"]I think Ryse looks better. Both are fantastic looking games.[/QUOTE]

How can you make such a statement while looking at bullshots?

[quote="cyberheater, post: 81967957"]Xbox One GPU is 1.3 Tflops. PS4 is 1.79Tflops.
The PS4 GPU isn't twice as fast as the Xbox One GPU.[/QUOTE]

X1: 1.31 TFLOPS
PS4: 1.84 TFLOPS

I don't know where you have got your 1.79 TFLOPS number from. And while it is true that PS4 is probably not 100% faster than the X1, it's also true that it has a 100% advantage regarding ROPS. How this will translate to game performance remains to be seen.

 

[cyberheater]

PS4 GPU is 1.84 Tflops. This has been known for the longest time. Where on Earth did you get the 1.79 figure from?

From here:-

http://www.redgamingtech.com/ps4-xbox-720-gpu-vs-amd-7970m-and-7970/

 

[Can Crusher]

I think Ryse looks better. Both are fantastic looking games.

The real time video of The Order 1866 looks significantly better than what we've seen from Ryse. We have to wait and see how final build looks, but right now there's a clear gap.

 

[IAmRandom31]

Xbox One GPU is 1.3 Tflops. PS4 is 1.79Tflops.
The PS4 GPU isn't twice as fast as the Xbox One GPU.

I meant twice as fast at rendering a scene via the ROP's. X1 has 16 rendering processors to finalize the scene whereas the PS4 has 32 rendering processors. It is a simple number, but when it comes to the " end game " so to speak, that is what it is all about. You want as many of those as possible. Same goes with stream processors and CU's and the like. The more the merrier.

TFLOP is just the floating point calculation potential of a machine. Overall it does not mean anything ( sorry, does not mean much ... when it comes to rendering )

The PS3's RSX GPU had a floating point calculating ability of 1.8TFLOP's. I'm sure we know the RSX is not equal to the PS4 GPU or X1 GPU heh

 

[Krilekk]

Do you think there will actually be no difference? Even MS has skirted around the fact that there will be a discrepancy. What exactly is outlandish about a claim that the more powerful hardware yields the better unoptimized code? Seriously, I'm trying to understand people who want to talk down developer testimony.

Maybe they talk them down because the Sony side of things talked down Carmacks claims that both systems are awfully close in terms of performance. Or talked down Microsofts claims that they would never give Sony a 30 % or more advantage and that the games will show there is no such difference. Or the claims of EA devs that X1 is much more powerful than they first expected it to be based on specs. Or the claims that dedicated servers lead to a 10 % increase in overall power on X1 because you can free the ressources you need for if the player becomes the host. Or the claims that you could offload computation to the cloud, leading to another increase of X1 performance. So you see, every side talks down the other because neither can't handle the truth. Which is ridiculous because in the end these companies just want our money. They aren't our friends and they sure don't need us to defend them.

 

[Prince Vultan]

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.

What's with all the aliasing in the Ryse picture? I was just reading in the Ryse thread about how that game didn't have any.

Can't really see... well, any aliasing in the Order shot though.

 

[nib95]

From here:-

http://www.redgamingtech.com/ps4-xbox-720-gpu-vs-amd-7970m-and-7970/

LOL. Why would you get your figure from some random no name site when the official numbers are out there.


EDIT: In the article you posted, it states the PS4's GPU is 1.76 Tflops, not what you posted which was 1.79 Tflops. Both figures are wrong lol. You kind of flopped it on this one Cyberheater. Certainly doesn't help with all your other arguments on this sort of tech stuff...

 

[cyberheater]

How do you calculate the tflops of a GPU?

LOL. Why would you get your figure from some random no name site when the official numbers are out there.

It was the first one that came up in my search engine.

 

[Chobel]

PS4 GPU is 1.84 Tflops. This has been known for the longest time. Where on Earth did you get the 1.79 figure from?

I think 1.79 from ATI 7850.

 

[nib95]

It was the first one that came up in my search engine.

Hmm...you must be using Bing.


(jk)

 

[Finalizer]

The fundamental problem is: although there is clearly a technical advantage to PS4, we do not have enough information to say how much. Therefore both of these statements are equally valid at this time:

"I think differences will be so small that they will be unnoticeable to most people"
"I think the differences will be immediate and obvious to anyone with an interest in games"

I think that is the crux of the argument - without any evidence to the actual quantifiable gap, we'll just go round and round in circles and nobody will be satisfied.

On top of that, you've got the sticky situation of subjectivity - If we had something like 1080p vs 900p consistently across the board, then I would say, I'd certainly notice the difference, and I think most folks on this site, and other gaming enthusiast sites, would notice the difference. But how many would actually care a great deal? Some might call it a dealbreaker, some couldn't care less.

It's kinda the problem I've been noticing in these debates. Some folks talk about moving goalposts come November, but I don't think the goalposts were ever set to begin with.

 

[SlickShoesRUCrazy]

How much of the imagery for the XBO out there is from actual XBO hardware and not the pc dev kits? Can anyone point me to images or video that they know for sure comes from XBO hardware and not the PC kits we saw at E3?

this would pertain to the ps4 also

 

[Gemüsepizza]

And if both consoles reserve 123 GFLOPS for the OS (it was 10% of the 1.23 TFLOPS for the X1), than it's even more than a 38% advantage for PS4.

It was the first one that came up in my search engine.

bing?

Spoiler

scnr

Edit: too late^^

 

[cyberheater]

The real time video of The Order 1866 looks significantly better than what we've seen from Ryse. We have to wait and see how final build looks, but right now there's a clear gap.

I was referring to the screen shots that the other guy posted.

 

[IT Slave]

This is the visual difference i expect to see

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.

Why are you comparing a launch game to a 2014/2015 title?

Here's one for Quantum Break which is a fairer comparison:

 

[AngryBird]

This is the visual difference i expect to see

Do i think the order looks better? Yeah. By how much? Well that's debatable. Most people outside of gaf wouldn't be able to tell you which is running on the more powerful console tbh.

Biased comparisons are fun!

 

[turk3y]

Seriously this needs to stop. 14+4 split was debunked.

Hence saying it was a rumour.

however it is what later gen games will effectively be doing, Mark Cerny even says the GPU is unbalanced and GPGPU will really kick off in a couple of years which means splitting gpu time between rendering and compute. It's not a fixed ratio but I would guess it's probably quite accurate.

folks need to stop worrying about numbers and just enjoy the games, it's not like you can see them compute units on screen as you play.

Even if totally hung up on stats 14 is still mote than 12 you can relax

 

[nib95]

Why are you comparing a launch game to a 2014/2015 title?

Here's one for Quantum Break which is a fairer comparison:

Fairer still.

Anyway...back to tech talk.

 

[velociraptor]

Very good. But was does that have to do with Xbox One and PS4.

7770: 1.28TF
7850: 1.76TF

XBO: 1.31TF
PS4: 1.84TF

It gives a representation of what to expect.

 

[Gemüsepizza]

Why are you comparing a launch game to a 2014/2015 title?

Here's one for Quantum Break which is a fairer comparison:

http://s.glbimg.com/po/tt/f/620x388/2013/06/10/screen_shot_2013-06-10_at_2.19.29_pm.png[IMG][/QUOTE]

We have no idea what part of the Quantum Break footage was real-time ingame footage running on the console and what not, so a comparison doesn't make much sense right now.

 

[Prince Vultan]

Why are you comparing a launch game to a 2014/2015 title?

Here's one for Quantum Break which is a fairer comparison:

Isn't Quantum Break just target footage in that shot? Certainly isn't in-game whatever the case,so it's a bit pointless comparing any of these games graphically until we're sure we're seeing actual in-game imagery from each of them.

 

[dcx4610]

The only true difference you are going to see is going to be frame rate. There is a chance that resolution could come into play but I seriously doubt it.

When it comes down to it, the PS4 might be able to get 5-10 more frames per second on the screen in a hectic sequence over the Xbox One. Is that a big deal? I guess it's really going to be depend on the game and if you are sensitive to frame rates.

In the end, I expect intense games to be locked at 30fps despite being able to achieve higher frame rates if they pushed it. This way both the PS4 and XB1 will be identical and there literally will be no difference. Lesser intense games running at 60fps, the same deal.

Battlefield 4 is probably going to be the most interesting game out of the gate since unless something changed, 60fps is targeted on both systems. If the frame rate is locked at 60 (V-Sync) and both system can maintain it without any dips, XB1 is going to be fine. If it can't handle it, get ready for a shit storm.

 

[ElTorro]

Even if totally hung up on stats 14 is still mote than 12 you can relax

The funny thing about the 14CU vs 12CU argument (apart from being based on the wrong 14+4 premise) is that it silently assumes that 4 CUs worth of GPU time dedicated to general compute tasks would not lead to any difference in the game's presentation.

 

[Chobel]

Hence saying it was a rumour.

however it is what later gen games will effectively be doing, Mark Cerny even says the GPU is unbalanced and GPGPU will really kick off in a couple of years which means splitting gpu time between rendering and compute. It's not a fixed ratio but I would guess it's probably quite accurate.

folks need to stop worrying about numbers and just enjoy the games, it's not like you can see them compute units on screen as you play.

Even if totally hung up on stats 14 is still mote than 12 you can relax

It's not a rumor any more if it's debunked.
When did Cerny say PS4 GPU is unbalanced?

 

[Codeblew]

Hmm...you must be using Bing.

Haha, that was a low blow man. I like it.

 

[vcc]

40% could mean one game would run 1080p in PS4 and 900p in Xbox One.
You need a trained eye in order to not notice that difference.

Frame rates might also be affected because more performance means more room to weather busy scenes. That won't be hard to notice and it's a common difference ins PS3 vs 360 games. Most of the performance edges the 360 has over the PS3 the PS4 has on the XB1 in addition to the theoretical performance numbers.

So it might be 1080p vs 900p and a lot more scenes where either the frame rate drops or the screen tears. Things that most people might not be able articulate but translates into perceptibly worse experience.

If all their friends are there, people would ignore that. But if the big 3:1 or 4:1 launch ratio are true it might be more of a factor for people on the fence.

 

[mrklaw]

And if both consoles reserve 123 GFLOPS for the OS (it was 10% of the 1.23 TFLOPS for the X1), than it's even more than a 38% advantage for PS4.



bing?

Spoiler

scnr

Edit: too late^^

I'd expect PS4 to be less than that, as there is no 'snap' equivalent, the only OS processes will be in the background during gameplay, and when in the foreground they can have full access to all resources

 

[SlickShoesRUCrazy]

I think 1.79 from ATI 7850.


That 38% is only when the game is using 50% of time in CPU and 50 % in GPU. And usually in console games time CPU usage is less %50, a lot less. So this means more than 38% difference.

does anybody ever take into account that 4 of the cus of the 18 can be used for gpgpu and in fact, Cerny wants developers to do that? Doesn't that leave only 14 cus for graphics processing depending on the developer?

I really don't know, I am a lay person of the subject, but that's what I have understood.

edit: looks like already brought up, nm.

 

[Chobel]

does anybody ever take into account that 4 of the cus of the 18 can be used for gpgpu and in fact, Cerny wants developers to do that? Doesn't that leave only 14 cus for graphics processing depending on the developer?

I really don't know, I am a lay person of the subject, but that's what I have understood.

edit: looks like already brought up, nm.

All the 18 CU's can be used in graphics and GPGPU stuff.

 

[ElTorro]

I'd expect PS4 to be less than that, as there is no 'snap' equivalent, the only OS processes will be in the background during gameplay, and when in the foreground they can have full access to all resources

Snap is particularly important because it is quite obviously the reason why the XBO reserves 10% of GPU time, leading to 1,18TF for games. Snapped applications are Metro-based, and Metro uses hardware-accelerated UIs. I don't want to to be overly nit-picky, but that thought always comes to my mind when these discussions go into the finest details like audio hardware.