Support NeoGAF

[Fresh Face]

You make a good point. Arguments and criticism are of course always welcome, but there is a line that is sometimes crossed and that only serves to diminish the credibility of Gaf

To be fair though, claiming biased towards Sony every time anything related to Xbox is criticized is also wrong and has been happening more and more lately.

It's like that one law of thermodynamics: every action has a reaction. The crappier comments on Sony's said, the crappier on the other. Neither are really justified, because both are outside the realms of good posting, but it helps to see the cause. Read my above post for more on this (if you have the time, lol.)

 

[Ricerocket]

No

Cerny never debunked anything. He said about the rumor, but no numbers were given about PS4 GPU by Sony, except the gflops.

And also that subjects could change without previous information.

All those +98893743897432% specs were given by VGLeaks.

So are you going to tell us what 14+4 CU is? You sound like you know what it's all about to keep bringing it up.

 

[Cthulhu_Steev]

No

oh

okay! thanks anyway, I believe you!

 

[Finalizer]

No

Seriously.

Digital Foundry: Going back to GPU compute for a moment, I wouldn't call it a rumour - it was more than that. There was a recommendation - a suggestion? - for 14 cores [GPU compute units] allocated to visuals and four to GPU compute...

Mark Cerny: That comes from a leak and is not any form of formal evangelisation.

Stop peddling this nonsense already.

 

[Muppet of a Man]

Surely those rules apply to both systems equally? Why is it I can point to previous generations where the hardware was known to be superior and it was quite clearly evident in early launch tiles?

Examples I can give; (graphical highlights of launch games)

Heavenly Sword and Motorstorm on PS3 to Condemned and PGR 3 on Xbox 360.

Halo and Project Gotham Racing on Xbox vs. Timesplitters and Ridge Racer V on PS2 vs. Star Wars: Rogue Leader and Luigi's Mansion on Gamecube.

These games were a good indicator to the power of each machine and the difference we would see throughout the generation.

Further back there's Pilotwings 64 on Nintendo 64, WipEout on PS1 and Panzer Dragoon on Saturn. All pretty representative for their time in regards to each machine's technical capabilities in direct comparison to the competition. Why is it different now? Especially since the ambiguity of custom architecture is now a null point.

You're argument is boardering on being moot at this point. Bish IS a third party multiplatform developer. Just because one system is specced significantly lower than another doesn't preclude it from having nice looking games. Remember, this is a relative comparison between 2 consoles both capable of producing nice looking games. However, the threshold for one of them, the ps4, will just be significantly higher.

 

[e-gamer]

Seriously.



Stop peddling this nonsense already.

So the +928347983274982374% specs from PS4 GPU. That also came from a leak.

But that you believe...

Who is being nonsense?

 

[Fresh Face]

Thanks God the most of people appreciate those. It's the best part of neogaf, don't try to censored those masterpieces for your lack of irony, please.

Oh, don't get me wrong, I love them as well...at first. It's why I love GAF. To be honest, I've take part in the making fun as well. I believe I've posted some of those gifs myself.

But seriously, nothing is funny after being showed 300+ times. Nothing. There is a breaking point somewhere. There is a reason some gifs are banned.

I don't mind antoniobanderas gif. I love all the new spins people put on it. But there is going to come a point where it'll get tiring.

 

[2dshmuplover]

I believe this is wrong and a myth. If not, why would skyrim, FF13 and bayonetta happen?

Multiplatform devs will push whatever they can in the allotted time given or else they'll be far surpassed and not taken as seriously as console exclusives. IMO I could be wrong because I'm not a dev.

I believe they are an exception to the rule. Bayonetta history is well documented, it was designed to take advantage of the 360 hardware, which is why Platinum gave it to Sega (who they felt were more experienced with the PS3 than themselves) in the hope they could pull out a better port than they could. FFXIII was also developed with PS3 hardware in mind for much of it's dev cycle.

Skyrim on PS3 is ok now though it's it? It suffered many problems at launch on all formats but they got it to within decent levels of parity.

Does it not stand to reason a first party exclusive title will likely push the host hardware more than a multiplatform game? Have we not seen that made quite evident throughout all of gaming history?

On a side note, if judging exclusives in order to determine the technical differences between machines is off the table then I take it we cannot deduce that PS3 featured superior tech to the 360 due to God of War III, Uncharted 3, The Last of Us etc

"ffs you cant compare exclusives!" they say..

Yet people had no problem dong that this gen did they?

 

[Piggus]

Yes, since the data will be changing twice as often per clock period, power draw will be higher for GDDR5 vs DDR3. Active power is dissipated only when data changes. So if data is going to change twice as often, you're definitely going to see a higher power draw.

It's higher, but it's not "omg YLOD 2.0" higher. GDDR5 even on high end GPUs doesn't get very hot and is usually cooled passively. It's the MOSFET and main chip (in this case the APU) that need the most cooling attention.

 

[omonimo]

So the +928347983274982374% specs from PS4 GPU. That also came from a leak.

But that you agreed...

Who is being nonsense?

"Kiss in the front"

 

[ShapeGSX]

Skeff already posted earlier how DDR3 and GDDR5 voltages are essentially the same. Add the power consumption of the ESRAM then roles are reversed and DDR3+ESRAM combo is actually consuming more power than the GDDR5.

You really can't believe everything you read on the internet. Voltage is only one part of the power equation.

The fact that GDDR5 switches twice as often per clock cycle is important. GDDR5 definitely burns more power than DDR3, yes, even at the same voltage.

 

[Sword Of Doom]

So are you going to tell us what 14+4 CU is? You sound like you know what it's all about to keep bringing it up.

Seriously what does that mean?

 

[WordsintheWater]

Huh? Is this true? I knew that Xbone had been using high-end PCs to show their games, but the last I had only heard that PS4 dev kits were being used to show PS4 games.

It's not true, they were PS4 dev kits to my knowledge at least for Sony's own games. I am not 100% sure about third party titles.

 

[malboroking]

So the +928347983274982374% specs from PS4 GPU. That also came from a leak.

But that you agreed...

Who is being nonsense?

What tbe hell are you talking about? You should really think long and hard bdfore you continue this line of argument.

 

[Finalizer]

So the +928347983274982374% specs from PS4 GPU. That also came from a leak.

But that you agreed...

I didn't agree to any claim of yours.

Cerny explicitly states that the numbers came from a leak, and that they're not a formal evangelisation - it's just an example of a split developers can go with. They can use any number of CUs for either graphics or GPGPU as they so need.

 

[Metfanant]

So the +928347983274982374% specs from PS4 GPU. That also came from a leak.

But that you agreed...

Who is being nonsense?

it doesnt MATTER...lol

lets try again...

1. Mark Cerny stated to DF that the 14+4 stuff is simply a suggestion...they intentionally included some "extra" ALU as an incentive for developers to use it for GPGPU...HOWEVER...nothing is set in stone, they can use the CU's however fits their game best

2. The hardware specs have proven to be just about 100% spot on

3. i once again must please BEG you to explain to me how i was wrong about all 18CU's being used for games...

 

[omonimo]

I believe they are an exception to the rule. Bayonetta history is well documented, it was designed to take advantage of the 360 hardware, which is why Platinum gave it to Sega (who they felt were more experienced with the PS3 than themselves) in the hope they could pull out a better port than they could. FFXIII was also developed with PS3 hardware in mind for much of it's dev cycle.

Skyrim on PS3 is ok now though it's it? It suffered many problems at launch on all formats but they got it to within decent levels of parity.

Does it not stand to reason a first party exclusive title will likely push the host hardware more than a multiplatform game? Have we not seen that made quite evident throughout all of gaming history?

On a side note, if judging exclusives in order to determine the technical differences between machines is off the table then I take it we cannot deduce that PS3 featured superior tech to the 360 due to God of War III, Uncharted 3, The Last of Us etc

"ffs you cant compare exclusives!" they say..

Yet people had no problem dong that this gen did they?

I think the problem is ps360 are quite comparable, ps4/xbone are not. A single is weaker, in whatever way you want to see the thing, first or late the gap should appear.

 

[ShapeGSX]

It's higher, but it's not "omg YLOD 2.0" higher. GDDR5 even on high end GPUs doesn't get very hot and is usually cooled passively. It's the MOSFET and main chip (in this case the APU) that need the most cooling attention.

I didn't say it was an issue. I was correcting an incorrect statement. There was an earlier statement that even equated voltage and power. This is hilarious misinformation.

 

[SwiftDeath]

Agree that the voltage is more important for speed. Just wondering if there is any difference in power draw/amperage, since that's what the comment from MS seems to be alluding to. Unless I completely misinterpreted it.

Yes, since the data will be changing twice as often per clock period, power draw will be higher for GDDR5 vs DDR3. Active power is dissipated only when data changes. So if data is going to change twice as often, you're definitely going to see a higher power draw.

Anyone more technical want to take a look at these docs on GDDR5 memory?

Pages 19 & 20 of this pdf

Samsung Slide deck

Not sure how much I trust the Samsung slide deck but the ELPIDA pdf seems legitimate

 

[2dshmuplover]

You're argument is boardering on being moot at this point. Bish IS a third party multiplatform developer. Just because one system is specced significantly lower than another doesn't preclude it from having nice looking games. Remember, this is a relative comparison between 2 consoles both capable of producing nice looking games. However, the threshold for one of them, the ps4, will just be significantly higher.

My point is, I've never known a point in history where it's not been made quite clear which console is more technically capable, right from the very onset. Why is it different now?

The suggestion is that developers aren't pushing the machines, (which applies for both) but surely Sony want the graphics to impress? They must be aware that graphics sell machines to the average gamer?

There's developing titles and there's developing titles, are we talking AAA first party/exclusive titles here which are effectively selling the next-generation to us? because that is really what we are discussing.

 

[Cthulhu_Steev]

What tbe hell are you talking about? You should really think long and hard bdfore you continue this line of argument.

That's what happens when you give TheKayle your login details.

I didn't agree to any claim of yours.

Cerny explicitly states that the numbers came from a leak, and that they're not a formal evangelisation - it's just an example of a split developers can go with. They can use any number of CUs for either graphics or GPGPU as they so need.

I hope that one developer goes GPGPU physics wild and uses the CUs for some experimental game, they could call it 4+14.

 

[killatopak]

You really can't believe everything you read on the internet. Voltage is only one part of the power equation.

The fact that GDDR5 switches twice as often per clock cycle is important. GDDR5 definitely burns more power than DDR3, yes, even at the same voltage.

I see. I was not aware of those. Would the ESRAM add any power consumption if any at all?

Thanks for the info. This is why I love tech threads. I'm studying Computer Engineering and could use as much info as I can get.

 

[e-gamer]

What tbe hell are you talking about? You should really think long and hard bdfore you continue this line of argument.

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

 

[Piggus]

So the +928347983274982374% specs from PS4 GPU. That also came from a leak.

But that you believe...

Who is being nonsense?

Dude, just stop. It's 18 CUs and all can be used for compute or more traditional tasks. You're not gonna win this argument.

 

[WordsintheWater]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

The reason this leak you speak of is voided in our minds is because we have confirmation it isn't true.

 

[malboroking]

That's what happens when you give TheKayle your login details.

OMG, I've got soda up my nose. Thanks a lot!

 

[2dshmuplover]

I think the problem is ps360 are quite comparable, ps4/xbone are not. A single is weaker, in whatever way you want to see the thing, first or late the gap should appear.

Really? when both machines last gen were architecturally very different, now they are effectively very similar indeed.

 

[no angel]

I don't understand any of this hardware talk. Can someone tell me (theoretically) what sort of differences we will notice between xbone and ps4 multiplats? How noticeable do yu think they'll be? I know this is probably a difficult question to answer but I'm really torn on which preorder to keep ( impending baby means I can only justify one!), my head says ps4 but I have friends who live a few hundred miles away going for the xbox and I'll miss not being able to play with them. On the flip side I'll feel gypped that I'll be playing a gimped version of the game. Decisions decisions...

 

[malboroking]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

Well, I'm afraid I'll have to take the word of the playstations lead architect over the word of some random guy on the internet. They can use all 18 of the CUs however they want.

 

[omonimo]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

 

[Metfanant]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

once again...let me just break it down for you

1. Can ALL the PS4's CU's be used for graphics rendering?

2. if the developer CHOOSES to split the CU's between rendering/GPGPU, how many of the PS4's CU's are used for GAMES?

3. is there a set in stone split of the PS4 CU's for rendering/GPGPU?

 

[viveks86]

Yes, since the data will be changing twice as often per clock period, power draw will be higher for GDDR5 vs DDR3. Active power is dissipated only when data changes. So if data is going to change twice as often, you're definitely going to see a higher power draw.

I see. Thanks for the reply! Could this also lead to heating differences with respect to each other?

 

[dragonelite]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

It doesn't really matter if the 14 + 4 was real or not.
TFlops will be used as the developers want them to be used.
They will just adjust the code there are more then enough pixel shaders you can convert to compute shaders and vice versa.

 

[Piggus]

I didn't say it was an issue. I was correcting an incorrect statement. There was an earlier statement that even equated voltage and power. This is hilarious misinformation.

Sorry, I wasn't trying to correct something you didn't say haha. I was just adding to what you said. A lot of people automatically think more power draw = significantly higher risk of component failure.

 

[omonimo]

Really? when both machines last gen were architecturally very different, now they are effectively very similar indeed.

Let me rephrase: ps360 are quite similar in performance, xbone is clearly inferior to ps4. THat's all.

 

[vcc]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

You do realize there is also official information right? Presentation slides from guerrilla games etc... The leaks people 'accept' have been ones that are cross verified by official information. The ones that they don't accept like a undisclosed hidden second titan GPU on the XB1 are not accepted because it clashes with multiple official data dumps. Like their system diagrams. The 14+4 also was officially put down as a rule of thumb some programmers were using rather than physical difference or system limitation. Put down by both statements from cerny and all the system diagrams they have put out.

 

[Finalizer]

There is a 25 pages thread where sony fanboys discuss leakes against oficial information.

But the leakes they want to.

The line of argument never took place here, sorry.

You're twisting the argument. We already know what 14+4 is - a suggestion, not a hardware division. You and countless others are just trying to push the notion otherwise to vainly attempt to close an obvious power disparity.

 

[ethomaz]

once again...let me just break it down for you

1. Can ALL the PS4's CU's be used for graphics rendering?

2. if the developer CHOOSES to split the CU's between rendering/GPGPU, how many of the PS4's CU's are used for GAMES?

3. is there a set in stone split of the PS4 CU's for rendering/GPGPU?

1. For what Cerny said... they can all be used for graphics rendering... it is a dev choice.

2. How many the dev want or need.

3. No.

 

[2dshmuplover]

Clearly I missed this. Thread link please.

Link

 

[ShapeGSX]

I see. I was not aware of those. Would the ESRAM add any power consumption if any at all?

Thanks for the info. This is why I love tech threads. I'm studying Computer Engineering and could use as much info as I can get.

Yes, of course the ESRAM would add to the power consumption. There will be both static and active power dissipation. I have no idea what sort of value to attach to that, however. The PS4's extra hardware in the APU would also increase power dissipation.

 

[viveks86]

Skeff already posted earlier how DDR3 and GDDR5 voltages are essentially the same. Add the power consumption of the ESRAM then roles are reversed and DDR3+ESRAM combo is actually consuming more power than the GDDR5.

I had an issue with Skeff's post and brought it up. Didn't get a response back then. ShapeGSX seems to have clarified it

well of course P=IV, Voltage is only half of it.

However, the difference between the power draw is not important in this scenario, it is completely blown up my MS to create a FUD campaign and here we are.

GDDR5 can be cooled passively just like DDR3. GDDR5 during development actually had a target of having a lower power consumption than GDDR3, which it achieved comfortably.

Considering the cooling system's Sony generally uses (blowing air through the entire system in one direction, as oppose to a standard radial system, this is likely in the PS4 looking at the design.) the GDDR5 will be cooled "comfortably"

Well there you are! Was wondering why I got no reply!

Interesting! And considering the fact that the temperature ratings and overall power consumption are quite below ps3 specs (as per FCC), I think we can conclude that either MS couldn't come up with the same design Sony came up with (either due to constraints we aren't aware of or due to competency) or is deliberately spreading FUD.

 

[Skeff]

You really can't believe everything you read on the internet. Voltage is only one part of the power equation.

The fact that GDDR5 switches twice as often per clock cycle is important. GDDR5 definitely burns more power than DDR3, yes, even at the same voltage.

well of course P=IV, Voltage is only half of it.

However, the difference between the power draw is not important in this scenario, it is completely blown up my MS to create a FUD campaign and here we are.

GDDR5 can be cooled passively just like DDR3. GDDR5 during development actually had a target of having a lower power consumption than GDDR3, which it achieved comfortably.

Considering the cooling system's Sony generally uses (blowing air through the entire system in one direction, as oppose to a standard radial system, this is likely in the PS4 looking at the design.) the GDDR5 will be cooled "comfortably"

EDIT: Power draw changes on each and every chip, it's likely the GDDR5 will use insignificantly more power than DDR3 in the xb1.

I had an issue with Skeff's post and brought it up. Didn't get a response back then. ShapeGSX seems to have clarified it




Well there you are! Was wondering why I got no reply!

Sorry, was away from the thread and when I came back I must have missed it.

EDIT2: I'm attempting to get some data regarding the overclocking of GPU memory to see a tangible effect on the temperature increase, But it's difficult to find to be honest.

EDIT3:

I didn't say it was an issue. I was correcting an incorrect statement. There was an earlier statement that even equated voltage and power. This is hilarious misinformation.

I didn't equate it, I noted it was a very important factor, and Physics backs me up on this with P=IV.

Hilarious Misinformation my arse.

 

[Metfanant]

1. For what Cerny said... they can all be used for graphics rendering... it is a dev choice.

2. How many the dev want or need.

3. No.

appreciate your response...i was however asking him solely because he has tried to dispute my assertion that all 18CU's are available for games...

 

[Fresh Face]

Link

0_0

Um...what? I never heard about this. There's a thread about this? The guy asked for a link to a thread, btw.

 

[benny_a]

0_0

Um...what? I never heard about this. There's a thread about this? The guy asked for a link to a thread, btw.

http://www.neogaf.com/forum/showthread.php?t=663625

 

[e-gamer]

once again...let me just break it down for you

1. Can ALL the PS4's CU's be used for graphics rendering?

2. if the developer CHOOSES to split the CU's between rendering/GPGPU, how many of the PS4's CU's are used for GAMES?

3. is there a set in stone split of the PS4 CU's for rendering/GPGPU?

That's not my point.

I'm not interested about how many CUs have been around.

I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.

 

[Metfanant]

0_0

Um...what? I never heard about this. There's a thread about this? The guy asked for a link to a thread, btw.

that is an image posted on the misterX blog attempting to blow smoke so that we dont focus on how his house of cards regarding the Xbone hardware falls apart around him

 

[Cidd]

Lmao at the 14+4 CU claims although it's already been disputed a million times.

What I find funny is even if the 14+4 CU split was true that's still 18 CU vs 12 CU

14+4 CU vs 12CU

That some how make things balance?

WTF am I reading?

 

[phosphor112]

No, its not.

And the reasons were explained long time ago.

Those "confirmed" facts came from VGLeaks.

Also the "14+4 balanced".

Keep digging your grave.

 

[artist]

Link

Did he just link us to misterxmedia? Holy shit!