edge.com ps4 1080p 30fps- xb1(900p sub 20fps. linky :-) o boy.dave

This topic is locked from further discussion.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#301 tormentos
Member since 2003 • 33793 Posts

 

 

 

7770 or 7790 is limited by two 64bit memory controllers and related L2 cache (64 byte per cycle per memory controller).

64 byte x 1050000000 Hz = 62 GB/s, hence 7770/7790 would have 124 GB/s i.e. 300 GB/s external memory would be pointless.


----

 

TFLOPS doesn't operate in isolation i.e. FLOP is supported by SRAM storage e.g. L1 cache.

The prototype-7850 with ~1.32 TFLOPS > 7770's 1.34 TFLOPS i.e. more L1 cache, wavefront buffer and LDS.

 

-----------------------

 

1.18 TFLOPS vs 1.76 TFLOPS = 55 percent more power is inconflict with

 

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 percent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

ronvalencia

 

Oh please dude STFU can you be a man an admit that 1.18TF is sh** not matter how  much cache or bandwidth you have 1.18TF is sh** and lower than 1.84 TF,that is the fact is the reason why Ryse is 900p and not 1080p,is the reason Killer Instint is 720p :lol: Is the reason why Titanfall is not 1080p,Or dead rising 3,basically the only first party game i know is 1080p is Forza which is a damn racing game with everythig baked and faked,oh and you know racing games demand less to make them run at 60FPS 1080p.

 

The 7870 on Dirt on 2560x1600 ultra will get 74  FPS,while those same setting on Metro will get you 20 FPS at best.

The whole Tflop don't work in isolation is sh** your theories are sh** and your arguments are been kill on the spot by xbox one games,let the game do the talking they are already talking 720p on a fighting game :lol:

Fine me a PS4 first party game that is 720p.

Avatar image for delta3074
delta3074

20003

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#302 delta3074
Member since 2007 • 20003 Posts

[QUOTE="delta3074"][QUOTE="tormentos"]the game is not 1080p so yeah Crytek trade resolution for frames.tormentos

Proof? did crytek say that themselves or are you justr assuming again, they could have cut the resolution to improve the game in many other areas or even, shock horror, as a design choice. Stop just assuming stuff mate and acting as if it's a definite when ,clearly, it's not

Dude the game had frames problems,in fact in neogaf were posted shots of the first video show and actual shots now,and the actual shots even lack certain effect show on the first video show,so apparently for graphics it wasn't.

Besides the point, stop posting what is essentially an assumption as a fact, you don't know for definite why they dropped the resolution, nobody except Crytek knows that.
Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#303 tormentos
Member since 2003 • 33793 Posts

I hardly know anything about hardware but enough to see that the extent of el tormos hardware knowledge is > his posts generally involve when something has bigger numbers than something else then victory is achieved. *marks down that no ps3 game will be in 720p*WilliamRLBaker

 

We know you don't know anything about anything..

 

First MS lie about bandwidth numbers of the xbox 360,they claimed that it had 278GB/s that bandwidth was add up,like you adding oranges yeah like that.

They join the 256GB/s of the EDRAM with GPU to memory bandwidth for a total of 278GB/s the GPU one to memory was 22GB/s,the problem is that the 256GB/s is a joke because it happend inside the EDRAM the connection from the ERDRAM to other components is 30 something GB/s,so it doesn't matter if the EDRAM could process inside 256GB/s,it can only comunicate with the hardware at 30 something GB/s which is a far cry from what they claim.

 

Basically MS lied to everything big time,oh and it was Major Nelson ESRAM number 1 defender.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#304 tormentos
Member since 2003 • 33793 Posts

Besides the point, stop posting what is essentially an assumption as a fact, you don't know for definite why they dropped the resolution, nobody except Crytek knows that.delta3074

 

No i can't 100% say,but i do know that when you have frame problems you give something up,and Ryse did had frames porblems it wasn't solid 30FPS lock.

So yeah 900p one way or the other resolution was give up for quality or frames you pick one.

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#305 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

[QUOTE="Kaz_Son"]Dat damage control. Ryse is 900p bro, it pretty much confirms the article.Kaz_Son

No it doesn't. Just like you have no proof that Crytek is "Struggling to get Ryse to run at 30FPS".

You are a cow fanboy who believes FUD about XB1, and every piece of hype for PS4.

 :lol: So much irony coming from a flaming lemming like yourself.

When have I made a post about the PS4? I just defend the XB1.

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#306 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

You don't know why they went to 900P.

Seriously, you guys are so fricken obsessed with the XBoxOne! Are you all really that insecure about your choice to pre-order the PS4?

tormentos

Lets see we get an article on how the PS4 is 50% stronger than the xbox one,the article claim that 1080p games running at 30FPS on PS4 are running at 900p at 20 something on xbox one (a scenario i have tell you about) and all of a sudden MS admit after claiming that Ryse was 1080p that it is 900p the same resolution the so call article quote,the game was running with problems on E3 is wasn't solid 30 and we all know it,so yeah lower resolution to fix frames problems.

The sad part about this is that this is Crytek you basically can't go better than this graphically with other lesser developers and Phill has stated that Ryse is in deed the best looking xbox one game.

All you do is go on this forum and bash the XB1. You can't keep your pie-hole shut about it for more that one post.

Ifyou are that worried about your PS4, you have issues.

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#307 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="cfisher2833"]

Tormentos: showing once again that he cares way too much about graphics and yet still clings to consoles.

casharmy

It's not really about graphics it's about trolling lemmings and making them rage and cry...and he does a very good job I must say.

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#308 tormentos
Member since 2003 • 33793 Posts

 

All you do is go on this forum and bash the XB1. You can't keep your pie-hole shut about it for more that one post.

Ifyou are that worried about your PS4, you have issues.

StormyJoe

 

Lets see all you do is defend the xbox one and bash the PS how are you any different.?

 

This post alone prove how imotinally invested you are on this..:lol:

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#309 tormentos
Member since 2003 • 33793 Posts

 

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

StormyJoe

 

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

 

So your stupidity is much greater.

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#310 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

All you do is go on this forum and bash the XB1. You can't keep your pie-hole shut about it for more that one post.

Ifyou are that worried about your PS4, you have issues.

tormentos

Lets see all you do is defend the xbox one and bash the PS how are you any different.?

This post alone prove how imotinally invested you are on this..:lol:

I don't "bash" the PS4. I defend the XB1.

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#311 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

tormentos

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

So your stupidity is much greater.

Neither do I. But, I am smart enough to know that there is more to a platform that it's raw GPU numbers.

Avatar image for Heil68
Heil68

60809

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#312 Heil68
Member since 2004 • 60809 Posts

[QUOTE="tormentos"]

[QUOTE="StormyJoe"]

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

StormyJoe

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

So your stupidity is much greater.

Neither do I. But, I am smart enough to know that there is more to a platform that it's raw GPU numbers.

Like Sonys proven track record of world class, GOTY winning exclusives? Awwww yeahhhhhh
Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#313 tormentos
Member since 2003 • 33793 Posts

 

Neither do I. But, I am smart enough to know that there is more to a platform that it's raw GPU numbers.

StormyJoe

 

Yeah like been a cble box..:lol:

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#314 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

[QUOTE="tormentos"]

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

So your stupidity is much greater.

Heil68

Neither do I. But, I am smart enough to know that there is more to a platform that it's raw GPU numbers.

Like Sonys proven track record of world class, GOTY winning exclusives? Awwww yeahhhhhh

Why is it that cows have to gang up on the ones who challenge them?

Anyway...

My point is this: go get Unreal Tournament for Windows and run it. Then, take an identical PC, install Linux on it, and run a copy of Unreal Tournament on it. Same hardware, but... HOLY F***ING SH!T!!! THEY DON'T RUN THE SAME!!!

It's amazing, I know...:roll:

If you cows educated yourselves as much as you moo for Sony, you'd be better off.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#315 tormentos
Member since 2003 • 33793 Posts

 

I don't "bash" the PS4. I defend the XB1.

StormyJoe

 

Yeah you don't... :lol:

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#316 tormentos
Member since 2003 • 33793 Posts

 

Why is it that cows have to gang up on the ones who challenge them?

Anyway...

My point is this: go get Unreal Tournament for Windows and run it. Then, take an identical PC, install Linux on it, and run a copy of Unreal Tournament on it. Same hardware, but... HOLY F***ING SH!T!!! THEY DON'T RUN THE SAME!!!

It's amazing, I know...:roll:

If you cows educated yourselves as much as you moo for Sony, you'd be better off.

StormyJoe

 

When L4D2 got to the point where it would actually load and run on Linux, the game managed a rather poor 6fps. Valve has since been working to introduce optimizations and get the game working better using OpenGL. The end result? L4D2 now runs at 315fps under Linux (Ubuntu 12.04 32-bit).

Not only is that a massive improvement, Valve has admitted it is faster than they have ever achieved for the game running under Windows using DirectX. In fact, running the game using the same hardware under Windows only managed 270fps.

Valve has since improved on that by further optimizing the Windows version of the game (running using OpenGL) based on what they learned from the Linux port. But they have still only managed 303.4fps12fps slower than the Linux version.

Valves explanation for the difference is that OpenGL just runs cleaner than DirectX. It has less overhead than DirectX for performing the same tasks, suggesting no amount of optimization from Valve engineers will ever bring the two inline. The DirectX APIs could improve though, to peg the gap.

 

http://www.geek.com/games/valve-gets-l4d2-running-faster-on-linux-than-windows-1506985/

 

And there goes your argument flying out of the window..:lol:

When optimized Linux actually run  games faster than window,thanks to OpenGL been cleaner and having less overhead than DIRECTX.

You know what DIRECTX is.? yeah is the API use on xbox one..

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#317 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

Why is it that cows have to gang up on the ones who challenge them?

Anyway...

My point is this: go get Unreal Tournament for Windows and run it. Then, take an identical PC, install Linux on it, and run a copy of Unreal Tournament on it. Same hardware, but... HOLY F***ING SH!T!!! THEY DON'T RUN THE SAME!!!

It's amazing, I know...:roll:

If you cows educated yourselves as much as you moo for Sony, you'd be better off.

tormentos

When L4D2 got to the point where it would actually load and run on Linux, the game managed a rather poor 6fps. Valve has since been working to introduce optimizations and get the game working better using OpenGL. The end result? L4D2 now runs at 315fps under Linux (Ubuntu 12.04 32-bit).

Not only is that a massive improvement, Valve has admitted it is faster than they have ever achieved for the game running under Windows using DirectX. In fact, running the game using the same hardware under Windows only managed 270fps.

Valve has since improved on that by further optimizing the Windows version of the game (running using OpenGL) based on what they learned from the Linux port. But they have still only managed 303.4fps12fps slower than the Linux version.

Valves explanation for the difference is that OpenGL just runs cleaner than DirectX. It has less overhead than DirectX for performing the same tasks, suggesting no amount of optimization from Valve engineers will ever bring the two inline. The DirectX APIs could improve though, to peg the gap.

http://www.geek.com/games/valve-gets-l4d2-running-faster-on-linux-than-windows-1506985/

And there goes your argument flying out of the window..:lol:

When optimized Linux actually run games faster than window,thanks to OpenGL been cleaner and having less overhead than DIRECTX.

You know what DIRECTX is.? yeah is the API use on xbox one..

Again, this is why I say you are a dumb*ss. You just proved my point that there is more to performance that raw hardware numbers.

Thanks, Tormentos.

OH, and BTW, the XB1 is running a custom DirectX API set specificially for the XB1's chipset.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#318 tormentos
Member since 2003 • 33793 Posts

 

Again, this is why I say you are a dumb*ss. You just proved my point that there is more to performance that raw hardware numbers.

Thanks, Tormentos.

OH, and BTW, the XB1 is running a custom DirectX API set specificially for the XB1's chipset.

StormyJoe

 

No i did not.

 

You have been trying other times to claim that thanks to MS API the difference will be smaller,i devated that with you that OpenGL was faster you refuse like always to admit it.

Oh please dude so does the PS4 is runs customs API as well,i already told you that fact is OpenGL >> DirectCX this is a fact,by the way advantage came with same hardware so Linux and OpenGl deliver 12 more frames per second at the end of the argument using the same hardware.

Imagine how much more linux and opengl can get with stronger hardware,which is the case here since the PS4 doesn't have the same specs as the xbox one it is stronger.

Avatar image for StormyJoe
StormyJoe

7806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#319 StormyJoe
Member since 2011 • 7806 Posts

[QUOTE="StormyJoe"]

Again, this is why I say you are a dumb*ss. You just proved my point that there is more to performance that raw hardware numbers.

Thanks, Tormentos.

OH, and BTW, the XB1 is running a custom DirectX API set specificially for the XB1's chipset.

tormentos

No i did not.

You have been trying other times to claim that thanks to MS API the difference will be smaller,i devated that with you that OpenGL was faster you refuse like always to admit it.

Oh please dude so does the PS4 is runs customs API as well,i already told you that fact is OpenGL >> DirectCX this is a fact,by the way advantage came with same hardware so Linux and OpenGl deliver 12 more frames per second at the end of the argument using the same hardware.

Imagine how much more linux and opengl can get with stronger hardware,which is the case here since the PS4 doesn't have the same specs as the xbox one it is stronger.

It's like debating with a vegetable...

You have no idea which API set will work better. All you have is the standard OpenGL API vs the standard DirectX API. Both the PS4 & XB1 use custom APIs, and the XB1's custom APIs are for it's custom CPU and GPU.

So... since before E3, you haven't been able to stfu about hardware specs, now your go off saying that APIs (and indirectly the platform's OS) do matter, but with out any evidence you claim a PS4 victory on that, too.

Wow...

Avatar image for jsmoke03
jsmoke03

13719

Forum Posts

0

Wiki Points

0

Followers

Reviews: 10

User Lists: 0

#320 jsmoke03
Member since 2004 • 13719 Posts

xb1's box looks clean...ps4 is just yuck. whats up with that blue?

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#321 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="WilliamRLBaker"]I hardly know anything about hardware but enough to see that the extent of el tormos hardware knowledge is > his posts generally involve when something has bigger numbers than something else then victory is achieved. *marks down that no ps3 game will be in 720p*tormentos

We know you don't know anything about anything..

First MS lie about bandwidth numbers of the xbox 360,they claimed that it had 278GB/s that bandwidth was add up,like you adding oranges yeah like that.

They join the 256GB/s of the EDRAM with GPU to memory bandwidth for a total of 278GB/s the GPU one to memory was 22GB/s,the problem is that the 256GB/s is a joke because it happend inside the EDRAM the connection from the ERDRAM to other components is 30 something GB/s,so it doesn't matter if the EDRAM could process inside 256GB/s,it can only comunicate with the hardware at 30 something GB/s which is a far cry from what they claim.

Basically MS lied to everything big time,oh and it was Major Nelson ESRAM number 1 defender.

MS didn't lie with Xbox 360's memory bandwidth.

You don't know you are talking about. My ATI Xenos slide has 8 pixel/clk info i.e. 8 ROPS. It's easy to do the math for raw bandwidth requirements for 8 ROPS.

ATI Xenos: 8 x 500,000,000 x 5 byte (for 40 bit FP color) = ~18 GB/s.

ATI Xenos: 8 x 500,000,000 x 5 byte (for 64 bit FP color) = ~29 GB/s.

NVIDIA's RSX: 8 x 500,000,000 x 8 byte (for 64 bit FP color) = ~29 GB/s. The graphic memory bandwidth is 22.4 GB/s.

The outcome is basically similar.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#322 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="StormyJoe"]

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

tormentos

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

So your stupidity is much greater.

1.18 TFLOPS / 1.84 TFLOPS is still in conflict with

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

----

1.84/1.18 = 1.559 or it's 56 percent more power. Gaijin dev > you

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#323 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="ronvalencia"]

7770 or 7790 is limited by two 64bit memory controllers and related L2 cache (64 byte per cycle per memory controller).

64 byte x 1050000000 Hz = 62 GB/s, hence 7770/7790 would have 124 GB/s i.e. 300 GB/s external memory would be pointless.


----

TFLOPS doesn't operate in isolation i.e. FLOP is supported by SRAM storage e.g. L1 cache.

The prototype-7850 with ~1.32 TFLOPS > 7770's 1.34 TFLOPS i.e. more L1 cache, wavefront buffer and LDS.

-----------------------

1.18 TFLOPS vs 1.76 TFLOPS = 55 percent more power is inconflict with

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 percent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

tormentos

Oh please dude STFU can you be a man an admit that 1.18TF is sh** not matter how much cache or bandwidth you have 1.18TF is sh** and lower than 1.84 TF,that is the fact is the reason why Ryse is 900p and not 1080p,is the reason Killer Instint is 720p :lol: Is the reason why Titanfall is not 1080p,Or dead rising 3,basically the only first party game i know is 1080p is Forza which is a damn racing game with everythig baked and faked,oh and you know racing games demand less to make them run at 60FPS 1080p.

The 7870 on Dirt on 2560x1600 ultra will get 74 FPS,while those same setting on Metro will get you 20 FPS at best.

The whole Tflop don't work in isolation is sh** your theories are sh** and your arguments are been kill on the spot by xbox one games,let the game do the talking they are already talking 720p on a fighting game :lol:

Fine me a PS4 first party game that is 720p.

You STFU.

1.18 TFLOPS / 1.84 TFLOPS is still in conflict with

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

----

1.84/1.18 = 1.559 or it's 56 percent more power. Gaijin dev > you

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#324 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="StormyJoe"]

 

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

tormentos

 

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

 

So your stupidity is much greater.

Your stupidity is much greater since your 1.18 vs 1.84 = 56 precent more power which is in conflict with actual dev's "40 percent more power".
Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#325 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="ronvalencia"]IMG0034556_zps6675b28d.jpg

7770 or 7790 is limited by two 64bit memory controllers and related L2 cache (64 byte per cycle per memory controller).

64 byte x 1050000000 Hz = 62 GB/s, hence 7770/7790 would have 124 GB/s i.e. 300 GB/s external memory would be pointless.


----

TFLOPS doesn't operate in isolation i.e. FLOP is supported by SRAM storage e.g. L1 cache.

The prototype-7850 with ~1.32 TFLOPS > 7770's 1.34 TFLOPS i.e. more L1 cache, wavefront buffer and LDS.

-----------------------

1.18 TFLOPS vs 1.76 TFLOPS = 55 percent more power is inconflict with

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 percent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

btk2k2

First of all those slight differences between the 7770/7790 and the X1 GPU are very minor compared to vast differences between the X1 GPU and the PS4 GPU. It is missing the general picture just to focus on the minutia. Of course the difference between a 7770 and the 7850 is likely not exactly the same as the difference between the X1 and the PS4 for a lot of reasons but as a general rule of thumb it is going to be in the ballpark of those figures with relatively large error bars. Now that quote you keep posting, why not post more of it, tell you what I will.
So its about policy, not power. AY: No its not about power. KY: We are scalable. We are running on a toaster [laughs]. AY: Well, obviously PlayStation 4 is more powerful than Xbox One. How much more powerful? AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing. How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4? AY: Yep. KY: Probably yes. But again, thats not a very big deal. AY: Well, for an online game the difference between Xbox One and PS4 is not that big a deal. PS4 is more powerful, basically, but the main reason is not about power.Edge-Online
So in essence one particular thing is better on the X1, which is memory write performance, but in every other metric the PS4 is faster by a good 40-50%. Now if all games required was massive memory write performance then the X1 would be golden but in reality games require a lot more than that.

Again, ROPS doesn't operate in isolation e.g. it requires memory writes. Alpha blends requires memory read and memory write operations.


This anon source http://www.edge-online.com/news/power-struggle-the-real-differences-between-ps4-and-xbox-one-performance/

Xbox One does, however, boast superior performance to PS4 in other ways. Lets say you are using procedural generation or raytracing via parametric surfaces that is, using a lot of memory writes and not much texturing or ALU Xbox One will be likely be faster, said one developer.

links with http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 percent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

Raytracing via parametric surfaces's write memory bias can be googled from other sources.

Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#326 04dcarraher
Member since 2004 • 23857 Posts
[QUOTE="tormentos"]

[QUOTE="StormyJoe"]

 

He doesn't make people cry. Well, unless stupidity makes someone cry... then yes, he would make some people cry.

ronvalencia

 

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

 

So your stupidity is much greater.

Your stupidity is much greater since your 1.18 vs 1.84 = 56 precent more power which is in conflict with actual dev's "40 percent more power".

some cant do the math 1.18 is 64% of the processing power of 1.84, which means a 36% difference. Again we also have no idea what allocation will the PS4 will use from its gpu for its features like voice/face recognition and other unknown addons. If we were to compare 100% vs 100% from the gpu's its only a 29% difference. Praising GDDR5 over ESram +DDR3 is dumb since both have their pros and cons,and in the end will have similar results for graphic workloads.
Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#327 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="ronvalencia"][QUOTE="tormentos"]

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

So your stupidity is much greater.

04dcarraher

Your stupidity is much greater since your 1.18 vs 1.84 = 56 precent more power which is in conflict with actual dev's "40 percent more power".

some cant do the math 1.18 is 64% of the processing power of 1.84, which means a 36% difference. Again we also have no idea what allocation will the PS4 will use from its gpu for its features like voice/face recognition and other unknown addons. If we were to compare 100% vs 100% from the gpu's its only a 29% difference. Praising GDDR5 over ESram +DDR3 is dumb since both have their pros and cons,and in the end will have similar results for graphic workloads.

No, my math is sound i.e. there's two ways to show the power difference.

Method One.

1.84 / 1.18 = 1.559 which is 55.9 precent more powerful. Using 1.18 as 100 percent.

1.84 / 1.31 = 1.405 which is 40.5 precent more powerful which matches

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

----

Using 1.31 number as 100 percent. To have "40 percent more powerful", you would have 1.40 x 1.31 = ~1.83.

The second method is as shown in your post. I have posted the second method in the past.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#328 tormentos
Member since 2003 • 33793 Posts

 

MS didn't lie with Xbox 360's memory bandwidth.

 

You don't know you are talking about. My ATI Xenos slide has 8 pixel/clk info i.e. 8 ROPS. It's easy to do the math for raw bandwidth requirements for 8 ROPS.

ATI Xenos: 8 x 500,000,000 x 5 byte (for 40 bit FP color) = ~18 GB/s.

ATI Xenos: 8 x 500,000,000 x 5 byte (for 64 bit FP color) = ~29 GB/s.

 

NVIDIA's RSX: 8 x 500,000,000 x 8 byte (for 64 bit FP color) = ~29 GB/s. The graphic memory bandwidth is 22.4 GB/s.

 

The outcome is basically similar.

ronvalencia

 

WTF you blind biased FANBOY MS claimed 278GB/s bandwith they joined bandwidth of the xbox 360 how dense can you be..?

 

b6rbt0.jpg

 

This ^^^^^^^^^^^^^^^^ is what MS climed this chart was made by MS and posted by Major Nelson it self you walking mozarella stick,Major nelson claimed the xb ox 360 had 278GB/s and they got to that number by adding the EDRAM 256 Gb/s + the memory to GPU 22GB/s...

22+256 = 278 more clear it can't be i don't even know why to try to prove me wrong on this is well know that the lied.

 

2003cdh.gif

 

It was inside the EDRAM that the speed was 256GB/s the EDRAM could send data only at 32GB/s..

 

So how the fu** did MS came to the 278GB/s claim.?

 

The memory system bandwidth in Xbox 360 exceeds the PS3s by five times.

 

The Xbox 360s CPU has more general purpose processing power because it has three general purpose cores, and Cell has just one.

 

This one is funny apparently to Major Nelson SPE didn't exist or were good for anything.

 

Xbox 360 has 278.4 GB/s of memory system bandwidth. The PS3 has less than one-fifth of Xbox 360s (48 GB/s) of total memory system bandwidth.

 

http://majornelson.com/2005/05/20/xbox-360-vs-ps3-part-1-of-4/

 

MS lied about the xbox one memory speed and Major Nelson claimed 5 times the speed of the PS3 which is a total joke..

 

Now lemming try to spin this one..

 

 

 

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#329 tormentos
Member since 2003 • 33793 Posts

 You STFU.

1.18 TFLOPS / 1.84 TFLOPS is still in conflict with

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

 

1.84/1.18 = 1.559 or it's 56 percent more power. Gaijin dev > youronvalencia

 

In conflict where the fu** the xbox one has a 10% GPU reservation that MS hasn't deny,the xbox one has some TV features like snap that do require GPU resources hell isn't metro UI GPU acelerared in windows 8.?

 

http://arstechnica.com/information-technology/2012/07/windows-8-gpu-acceleration-good-news-for-metro/

Oh yes it does....

 

http://www.engadget.com/2013/05/21/xbox-one-runs-three-operating-systems/

what the xbox one runs..:lol:

 

You are to hypocrite to use that article which claim 40% more power period,point is what he think the gap is not a big deal,but the gap still is 40% which is a big deal to many.

Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#330 04dcarraher
Member since 2004 • 23857 Posts

[QUOTE="04dcarraher"][QUOTE="ronvalencia"] Your stupidity is much greater since your 1.18 vs 1.84 = 56 precent more power which is in conflict with actual dev's "40 percent more power".ronvalencia

some cant do the math 1.18 is 64% of the processing power of 1.84, which means a 36% difference. Again we also have no idea what allocation will the PS4 will use from its gpu for its features like voice/face recognition and other unknown addons. If we were to compare 100% vs 100% from the gpu's its only a 29% difference. Praising GDDR5 over ESram +DDR3 is dumb since both have their pros and cons,and in the end will have similar results for graphic workloads.

No, my math is sound i.e. there's two ways to show the power difference.

Method One.

1.84 / 1.18 = 1.559 which is 55.9 precent more powerful. Using 1.18 as 100 percent.

1.84 / 1.31 = 1.405 which is 40.5 precent more powerful which matches

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

----

Using 1.31 number as 100 percent. To have "40 percent more powerful", you would have 1.40 x 1.31 = ~1.83.

The second method is as shown in your post. I have posted the second method in the past.

Not talking about your math..... others math of the 50%+, 40% is in the ballpark
Avatar image for btk2k2
btk2k2

440

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#331 btk2k2
Member since 2003 • 440 Posts
Again, ROPS doesn't operate in isolation e.g. it requires memory writes. Alpha blends requires memory read and memory write operations.


This anon source http://www.edge-online.com/news/power-struggle-the-real-differences-between-ps4-and-xbox-one-performance/

Xbox One does, however, boast superior performance to PS4 in other ways. Lets say you are using procedural generation or raytracing via parametric surfaces that is, using a lot of memory writes and not much texturing or ALU Xbox One will be likely be faster, said one developer.

links with http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 percent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

Raytracing via parametric surfaces's write memory bias can be googled from other sources.

ronvalencia
Nothing is in isolation though is it? As a general rule of thumb the PS4 will have a similar advantage over the X1 as the 7850 has over the 7770 (or the 7870Ghz has over the 7790). That is not to say that the PS4 will have a 20/30 FPS advantage over the X1 it is to say that it will have a 40-50% advantage over the X1 at the same image quality. In some cases it will be less because the game uses the ESRAM effectively and leans heavily on the areas the X1 is actually good it but in other games it will be more because the ESRAM is not used effectively and it leans heavily on what the PS4 is good at. Again, that quote you keep parading around is missing most of the information. They stated, more than once, that the PS4 is faster than the X1. There are a couple of areas where the X1 has a slight advantage over the PS4 (triangle set up rate/memory writes) but other than that the PS4 has distinct advantages over the X1. Do you think the X1 and the PS4 are approximately equal in terms of maximum graphical output?
Avatar image for Krelian-co
Krelian-co

13274

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#332 Krelian-co
Member since 2006 • 13274 Posts

[QUOTE="ronvalencia"][QUOTE="tormentos"]

 

I am not the one thinking that 1.18 TF and 184 TF will yield the same performance..

 

So your stupidity is much greater.

04dcarraher

Your stupidity is much greater since your 1.18 vs 1.84 = 56 precent more power which is in conflict with actual dev's "40 percent more power".

some cant do the math 1.18 is 64% of the processing power of 1.84, which means a 36% difference. Again we also have no idea what allocation will the PS4 will use from its gpu for its features like voice/face recognition and other unknown addons. If we were to compare 100% vs 100% from the gpu's its only a 29% difference. Praising GDDR5 over ESram +DDR3 is dumb since both have their pros and cons,and in the end will have similar results for graphic workloads.

your math is ok, but is a significant value, "only a 29%" difference, i dont know about you but 29% is a pretty big number. Also people praise gddr5 over ddr3 and esram because its 8 gb of faster ram which is easier to program for than a tiny esram (i think is 32 mb lol) that accelerates a slower 8 gb ddr3 ram and is harder to program for.

Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#333 04dcarraher
Member since 2004 • 23857 Posts

[QUOTE="04dcarraher"][QUOTE="ronvalencia"] Your stupidity is much greater since your 1.18 vs 1.84 = 56 precent more power which is in conflict with actual dev's "40 percent more power".Krelian-co

some cant do the math 1.18 is 64% of the processing power of 1.84, which means a 36% difference. Again we also have no idea what allocation will the PS4 will use from its gpu for its features like voice/face recognition and other unknown addons. If we were to compare 100% vs 100% from the gpu's its only a 29% difference. Praising GDDR5 over ESram +DDR3 is dumb since both have their pros and cons,and in the end will have similar results for graphic workloads.

your math is ok, but is a significant value, "only a 29%" difference, i dont know about you but 29% is a pretty big number. Also people praise gddr5 over ddr3 and esram because its 8 gb of faster ram which is easier to program for than a tiny esram (i think is 32 mb lol) that accelerates a slower 8 gb ddr3 ram and is harder to program for.

29% is not a whole alot when your comparing gpu's that have the same base of features. Look at the 7850 vs 7870 or even 7770 vs 7790 both sets of those gpu's have similar or larger gap in processing power then these consoles and yet only see 8-12 fps difference at 1080 gaming. Where using one or two lower quality assets/settings can allow the X1 to be on par with the PS4 with multiplat games with performance and most people wouldn't be able to tell the difference. Do not pull the esram is harder to code for while GDDR5 is easier BS. Dev's learned to use Cell on the PS3 and learned how to use EDRAM on the 360. The esram is much like cache for cpu's but its for the APU to use for what ever it needs and also esram can read/write data 4x the speed of GDDR5, And 256bit DDR3 that moves 68 gb/s is enough for graphics and system work. Also dont forget the dual lanes 10gb/s in the PS4's APU to allow the gpu and cpu to communicate directly bypassing GDDR5.
Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#334 04dcarraher
Member since 2004 • 23857 Posts
[QUOTE="ronvalencia"]Again, ROPS doesn't operate in isolation e.g. it requires memory writes. Alpha blends requires memory read and memory write operations.


This anon source http://www.edge-online.com/news/power-struggle-the-real-differences-between-ps4-and-xbox-one-performance/

Xbox One does, however, boast superior performance to PS4 in other ways. Lets say you are using procedural generation or raytracing via parametric surfaces that is, using a lot of memory writes and not much texturing or ALU Xbox One will be likely be faster, said one developer.

links with http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 percent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

Raytracing via parametric surfaces's write memory bias can be googled from other sources.

btk2k2
Nothing is in isolation though is it? As a general rule of thumb the PS4 will have a similar advantage over the X1 as the 7850 has over the 7770 (or the 7870Ghz has over the 7790). That is not to say that the PS4 will have a 20/30 FPS advantage over the X1 it is to say that it will have a 40-50% advantage over the X1 at the same image quality. In some cases it will be less because the game uses the ESRAM effectively and leans heavily on the areas the X1 is actually good it but in other games it will be more because the ESRAM is not used effectively and it leans heavily on what the PS4 is good at. Again, that quote you keep parading around is missing most of the information. They stated, more than once, that the PS4 is faster than the X1. There are a couple of areas where the X1 has a slight advantage over the PS4 (triangle set up rate/memory writes) but other than that the PS4 has distinct advantages over the X1. Do you think the X1 and the PS4 are approximately equal in terms of maximum graphical output?

There is major flaws comparing 7770 vs 7850 and even 7870 vs 7790. One inherit flaw is ignoring that the 7700 series are limited by the 128bit bus. while the 7800's are on a 256bit bus. The 7790 TFLOP performance is actually higher then 7850's(1785 vs 1755). If you compare 7770 vs 7790 there is a 505 GFLOP difference and only see 8 fps difference with BF3 1200p max settings. Now 7850 vs 7870 there is a 800 GFLOP difference and yet same game same settings see 10 fps difference. The end results between the PS4 and X1 are going to be minor, X1 will have to lower a few settings/assets to make up the difference but those are not going to be deal breakers and most people will not see the differences. But since we know the PS4 is suppose to support voice and facial recognition and other unknown features we have no idea how many gpu cycles(allocation will be used to process it) So the difference between the two may be smaller yet.
Avatar image for btk2k2
btk2k2

440

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#335 btk2k2
Member since 2003 • 440 Posts
29% is not a whole alot when your comparing gpu's that have the same base of features. Look at the 7850 vs 7870 or even 7770 vs 7790 both sets of those gpu's have similar or larger gap in processing power then these consoles and yet only see 8-12 fps difference at 1080 gaming. Where using one or two lower quality assets/settings can allow the X1 to be on par with the PS4 with multiplat games with performance and most people wouldn't be able to tell the difference. Do not pull the esram is harder to code for while GDDR5 is easier BS. Dev's learned to use Cell on the PS3 and learned how to use EDRAM on the 360. The esram is much like cache for cpu's but its for the APU to use for what ever it needs and also esram can read/write data 4x the speed of GDDR5, And 256bit DDR3 that moves 68 gb/s is enough for graphics and system work. Also dont forget the dual lanes 10gb/s in the PS4's APU to allow the gpu and cpu to communicate directly bypassing GDDR5. 04dcarraher
30-40% is about how much faster the Radeon R9 290X GPU will be over the 7970Ghz. That is a new generation of GPU and if 30-40% is considered enough for a generational increase then a 30-40% difference between the PS4 and the X1 is actually quite a considerable margin, sure in reality it means a few extra graphics settings or 1080p vs 900p on the PS4 but that is an advantage and that advantage costs less. Unless the X1 has something you really want I do not see the point in buying it yet. Also, just because the devs learned how to make Cell work and got to grips with the EDRAM does not change the fact that the PS4 IS easier to code for. While the ESRAM is like a cache it is manually managed rather than hardware managed which means the developers have to actually make it work. Until MS come out with some good APIs that can make effective use of the ESRAM without much developer intervention it is going to be left under utilised.
Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#336 04dcarraher
Member since 2004 • 23857 Posts

[QUOTE="04dcarraher"]29% is not a whole alot when your comparing gpu's that have the same base of features. Look at the 7850 vs 7870 or even 7770 vs 7790 both sets of those gpu's have similar or larger gap in processing power then these consoles and yet only see 8-12 fps difference at 1080 gaming. Where using one or two lower quality assets/settings can allow the X1 to be on par with the PS4 with multiplat games with performance and most people wouldn't be able to tell the difference. Do not pull the esram is harder to code for while GDDR5 is easier BS. Dev's learned to use Cell on the PS3 and learned how to use EDRAM on the 360. The esram is much like cache for cpu's but its for the APU to use for what ever it needs and also esram can read/write data 4x the speed of GDDR5, And 256bit DDR3 that moves 68 gb/s is enough for graphics and system work. Also dont forget the dual lanes 10gb/s in the PS4's APU to allow the gpu and cpu to communicate directly bypassing GDDR5. btk2k2
30-40% is about how much faster the Radeon R9 290X GPU will be over the 7970Ghz. That is a new generation of GPU and if 30-40% is considered enough for a generational increase then a 30-40% difference between the PS4 and the X1 is actually quite a considerable margin, sure in reality it means a few extra graphics settings or 1080p vs 900p on the PS4 but that is an advantage and that advantage costs less. Unless the X1 has something you really want I do not see the point in buying it yet. Also, just because the devs learned how to make Cell work and got to grips with the EDRAM does not change the fact that the PS4 IS easier to code for. While the ESRAM is like a cache it is manually managed rather than hardware managed which means the developers have to actually make it work. Until MS come out with some good APIs that can make effective use of the ESRAM without much developer intervention it is going to be left under utilised.

Again comparing a new generation vs current gen gpu series and stating the 30-40% difference as a whole new generation increase is a joke, because there is a 30% difference between 7850 and 7870 is that a generation increase nope. Both gpu's have the same base features. Also dont forget the games on ps4 that are below 1080 too.... the ESram is more than less cache its not hard to use.....

Avatar image for Tighaman
Tighaman

1038

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#337 Tighaman
Member since 2006 • 1038 Posts
[QUOTE="04dcarraher"]29% is not a whole alot when your comparing gpu's that have the same base of features. Look at the 7850 vs 7870 or even 7770 vs 7790 both sets of those gpu's have similar or larger gap in processing power then these consoles and yet only see 8-12 fps difference at 1080 gaming. Where using one or two lower quality assets/settings can allow the X1 to be on par with the PS4 with multiplat games with performance and most people wouldn't be able to tell the difference. Do not pull the esram is harder to code for while GDDR5 is easier BS. Dev's learned to use Cell on the PS3 and learned how to use EDRAM on the 360. The esram is much like cache for cpu's but its for the APU to use for what ever it needs and also esram can read/write data 4x the speed of GDDR5, And 256bit DDR3 that moves 68 gb/s is enough for graphics and system work. Also dont forget the dual lanes 10gb/s in the PS4's APU to allow the gpu and cpu to communicate directly bypassing GDDR5. btk2k2
the problem on this s board is that everything is one way and the only way you all never talk about the advantages that the x1 has only looking at a down fall but high praise Sony when yall in know that the have power reserved for that camera but ignore it, video encoding/decoding ,audio processing, game chat, etc use power for the GPU and also suggesting using the GPU also as compute. You all spue this 1.18tfs but never try to find what ps4 real tfs because you know and I know its not 1.8 after considering all the other stuff that what comes with making a game. 30-40% is about how much faster the Radeon R9 290X GPU will be over the 7970Ghz. That is a new generation of GPU and if 30-40% is considered enough for a generational increase then a 30-40% difference between the PS4 and the X1 is actually quite a considerable margin, sure in reality it means a few extra graphics settings or 1080p vs 900p on the PS4 but that is an advantage and that advantage costs less. Unless the X1 has something you really want I do not see the point in buying it yet. Also, just because the devs learned how to make Cell work and got to grips with the EDRAM does not change the fact that the PS4 IS easier to code for. While the ESRAM is like a cache it is manually managed rather than hardware managed which means the developers have to actually make it work. Until MS come out with some good APIs that can make effective use of the ESRAM without much developer intervention it is going to be left under utilised.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#338 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="ronvalencia"]

You STFU.

1.18 TFLOPS / 1.84 TFLOPS is still in conflict with

http://www.edge-online.com/news/gaijin-games-on-why-war-thunder-isnt-coming-to-xbox-one/

How much more powerful?

AY: It depends what youre doing. GPU, like 40 per cent more powerful. DDR5 is basically 50 per cent more powerful than DDR3, but the memory write [performance] is bigger on Xbox One so it depends on what youre doing.

How is that going to translate to on-screen results for the kinds of games you want to make? So to optimise War Thunder on both consoles you could hypothetically make a better, prettier version on PS4?

AY: Yep.

KY: Probably yes. But again, thats not a very big deal.

1.84/1.18 = 1.559 or it's 56 percent more power. Gaijin dev > youtormentos

In conflict where the fu** the xbox one has a 10% GPU reservation that MS hasn't deny,the xbox one has some TV features like snap that do require GPU resources hell isn't metro UI GPU acelerared in windows 8.?

http://arstechnica.com/information-technology/2012/07/windows-8-gpu-acceleration-good-news-for-metro/

Oh yes it does....

http://www.engadget.com/2013/05/21/xbox-one-runs-three-operating-systems/

what the xbox one runs..:lol:

You are to hypocrite to use that article which claim 40% more power period,point is what he think the gap is not a big deal,but the gap still is 40% which is a big deal to many.

Windows 8 acceleration is done by on-demand basis i.e. sleeps during full screen game mode.

Since I'm running Windows 8 X64 with Media Center and IE 10, my desktop AMD Radeon HD driver (via AMD's Overdrive panel) reports that my GCNs workload are running near zero or zero percent. Only a non-AMD GCN user would make this mistake.

10 percent of 1.3 TFLOPS GCN is a joke since when my AMD C-50/Z-01 APU (with 80 stream processors @ 276Mhz VLIW5, Radeon HD 6250M) based tablet easily handles Windows 8 32bit with MCE i.e. smooth. PS; I have swaped out the tablet's slow SSD with a faster SSD.

Avatar image for Tessellation
Tessellation

9297

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#339 Tessellation
Member since 2009 • 9297 Posts
damn ronvalencia causing damage and butthurt to the basement dwellers camp :cool:.
Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#340 tormentos
Member since 2003 • 33793 Posts

29% is not a whole alot when your comparing gpu's that have the same base of features. Look at the 7850 vs 7870 or even 7770 vs 7790 both sets of those gpu's have similar or larger gap in processing power then these consoles and yet only see 8-12 fps difference at 1080 gaming. Where using one or two lower quality assets/settings can allow the X1 to be on par with the PS4 with multiplat games with performance and most people wouldn't be able to tell the difference. Do not pull the esram is harder to code for while GDDR5 is easier BS. Dev's learned to use Cell on the PS3 and learned how to use EDRAM on the 360. The esram is much like cache for cpu's but its for the APU to use for what ever it needs and also esram can read/write data 4x the speed of GDDR5, And 256bit DDR3 that moves 68 gb/s is enough for graphics and system work. Also dont forget the dual lanes 10gb/s in the PS4's APU to allow the gpu and cpu to communicate directly bypassing GDDR5. 04dcarraher

 

I did and even posted benchmakrs and even fallowing your rules which i should not since the xbox one doesn't have a PTCAIRN GPU it has a BONAIRE one.

 

The difference could be as big as 24 FPS.

But mostly hover between 10  and 18.

 

Come on a game targeting 30FPS consoles will  run 10 to 18 frames slower on xbox one.

This is witout taking into nothing the PS4 customization for compute.

Also do you have a link to ESRAM been 4 times faster than GDDR5.?

 

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#341 tormentos
Member since 2003 • 33793 Posts

 Windows 8 acceleration is done by on-demand basis i.e. sleeps during full screen game mode.ronvalencia

 

Windows 8 has  GPU acelerated UI..Nothing more to say,and the xbox one has a snap feature that requires GPU resources as well,so yeah there you have it 10% confirmed..

Look at it from the bright side,the xbox one UI shoulod be way faster and smooth than the PS4 one.

Well i think..

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#342 tormentos
Member since 2003 • 33793 Posts

There is major flaws comparing 7770 vs 7850 and even 7870 vs 7790.One inherit flaw is ignoring that the 7700 series are limited by the 128bit bus. while the 7800's are on a 256bit bus. The 7790 TFLOP performance is actually higher then 7850's(1785 vs 1755). If you compare 7770 vs 7790 there is a 505 GFLOP difference and only see 8 fps difference with BF3 1200p max settings. Now 7850 vs 7870 there is a 800 GFLOP difference and yet same game same settings see 10 fps difference. The end results between the PS4 and X1 are going to be minor, X1 will have to lower a few settings/assets to make up the difference but those are not going to be deal breakers and most people will not see the differences. But since we know the PS4 is suppose to support voice and facial recognition and other unknown features we have no idea how many gpu cycles(allocation will be used to process it) So the difference between the two may be smaller yet.04dcarraher

 

No is not limited by the 128 Bit bus where the fu** do you people keep pulling this sh** from,the 7770 is a damn weak GPU that doesn't need a 256 bit bus with 200GB/s bandwidth,AMD i am sure knows a hell of allot more than you about their GPU and i am sure that they will not cripple on purppose their GPU vs the competition ones by not giving them enough bandwidth,want proof that is not limited.?

 

Is the 7770 was limited by the 128 bus that mean that power over 1.28TF over that 128 bit bus will not show any kind of gain,let say if you over clock the GPU to actually be 1.35 TF it would not show any gains because the 128 bit bus is limiting the GPU.

How do you know that is not right.? Easy the 7790 also has a 128 bit bus and has 1.79 TF and performs better than the 7770 so yeah the bus is not a problem for the 7770,hell if you tell me that of the 7790 i could believe it because it has more flops than the 7850 and performs worse,while the xbox one has a bonaire GPU which is 7790 it has 2 less CU and lower clock speed the GPU is actually 30,000Gflops more than a 7770,1310 vs 1280 from the 7770,so as you can see the xbox one doesn't need 204Gb/s bandwidth it will serve it for nothing.

 

Is not just 10 frames i have tell you..

 

As big as 24 FPS in dirt 3.

10 on total war

As big as 12 FPS in hitman

Sleeping dogs as big as 19 FPS

Far cry 3 10 FPS

BF3 as big as 14 FPS

Civillization as big as 14 FPS.

 

http://www.anandtech.com/bench/product/857?vs=778

7 games from those 7 only 2 games show a 10FPS difference the rest show more than 10.

 

I don't know why people have such a damage control for this trying to imply that a 7770 performance will match some how 7850 OC performance is silly you can give the xbox one 176Gb'/s bandwidth of the PS4 and still it would change nothing.

 

 

 

Avatar image for Chutebox
Chutebox

51561

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#343 Chutebox  Online
Member since 2007 • 51561 Posts

damn ronvalencia causing damage and butthurt to the basement dwellers camp :cool:.Tessellation
You really need a new shtick.

Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#344 04dcarraher
Member since 2004 • 23857 Posts

[QUOTE="04dcarraher"] 29% is not a whole alot when your comparing gpu's that have the same base of features. Look at the 7850 vs 7870 or even 7770 vs 7790 both sets of those gpu's have similar or larger gap in processing power then these consoles and yet only see 8-12 fps difference at 1080 gaming. Where using one or two lower quality assets/settings can allow the X1 to be on par with the PS4 with multiplat games with performance and most people wouldn't be able to tell the difference. Do not pull the esram is harder to code for while GDDR5 is easier BS. Dev's learned to use Cell on the PS3 and learned how to use EDRAM on the 360. The esram is much like cache for cpu's but its for the APU to use for what ever it needs and also esram can read/write data 4x the speed of GDDR5, And 256bit DDR3 that moves 68 gb/s is enough for graphics and system work. Also dont forget the dual lanes 10gb/s in the PS4's APU to allow the gpu and cpu to communicate directly bypassing GDDR5. tormentos

 

I did and even posted benchmakrs and even fallowing your rules which i should not since the xbox one doesn't have a PTCAIRN GPU it has a BONAIRE one.

 

The difference could be as big as 24 FPS.

But mostly hover between 10  and 18.

 

Come on a game targeting 30FPS consoles will  run 10 to 18 frames slower on xbox one.

This is witout taking into nothing the PS4 customization for compute.

Also do you have a link to ESRAM been 4 times faster than GDDR5.?

 

8-15 will the average difference at best. Bonaire is not the same as the 7770, 7790 uses the Bonaire and is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture as Tahiti and Pitcairn top is Bonaire,Tahti and Pitcaim while bottom is the 7770 the 7770 core is the smallest, with a max of 640 stream processors, and a single primitive pipeline. [img] http://www.extremetech.com/wp-content/uploads/2013/03/GPU-Frontends.png[img]

There is a thing called optimization customizing the settings and assets to fit the hardware. Heck BF3 with a 7770 max settings at 1280x800 with 2x MSAA, gets 51 fps average, while 7790 gets 65 FPS average. expect the X1 to be better then the 7770 overall without the inherit 128 bus limits and 7770 flaws. We all know the PS4 gpu is stronger then the X1 but dont ignore the new facts about the PS4 is suppose to support voice and facial recognition and that will eat processing cycles as well. We have no real idea how much of the PS4's gpu will be allocated for other jobs. So the difference between the two may be smaller then some think.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#345 tormentos
Member since 2003 • 33793 Posts

 

8-15 will the average difference at best. Bonaire is not the same as the 7770, 7790 uses the Bonaire and is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture as Tahiti and Pitcairn top is Bonaire,Tahti and Pitcaim while bottom is the 7770 the 7770 core is the smallest, with a max of 640 stream processors, and a single primitive pipeline. [img] http://www.extremetech.com/wp-content/uploads/2013/03/GPU-Frontends.png[img]

There is a thing called optimization customizing the settings and assets to fit the hardware. Heck BF3 with a 7770 max settings at 1280x800 with 2x MSAA, gets 51 fps average, while 7790 gets 65 FPS average. expect the X1 to be like the 7790 with 7770 processing power overall without the inherit 128 bus limits and 7770 flaws. We all know the PS4 gpu is stronger then the X1 but dont ignore the new facts about the PS4 is suppose to support voice and facial recognition and that will eat processing cycles as well. We have no real idea how much of the PS4's gpu will be allocated for other jobs. So the difference between the two may be smaller then some think.

04dcarraher

 

Bonaire and PITCAIRN and not the samejust like cape verde and Bonaire aren't.

There is no 128 bit bus limitation it doesn't exist,should i post how much bullsh** that claim is.?

The 7950 has a higher bit bus than the 660Ti also almost 100GB/s more bandwidth 144 for the 660TI vs the 7950 240,and yet they both perform almost the same.

 

The PS eye is not mandatory there is nothing to eat dude,Kinect is build in into the software and is included,the ps eye is not and has not reservation what so ever,just like the PS3 didn't reserve anything for the PS eye either some of you are just graspping..

Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#346 04dcarraher
Member since 2004 • 23857 Posts

[QUOTE="04dcarraher"]

 

8-15 will the average difference at best. Bonaire is not the same as the 7770, 7790 uses the Bonaire and is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture as Tahiti and Pitcairn top is Bonaire,Tahti and Pitcaim while bottom is the 7770 the 7770 core is the smallest, with a max of 640 stream processors, and a single primitive pipeline. [img] http://www.extremetech.com/wp-content/uploads/2013/03/GPU-Frontends.png[img]

There is a thing called optimization customizing the settings and assets to fit the hardware. Heck BF3 with a 7770 max settings at 1280x800 with 2x MSAA, gets 51 fps average, while 7790 gets 65 FPS average. expect the X1 to be like the 7790 with 7770 processing power overall without the inherit 128 bus limits and 7770 flaws. We all know the PS4 gpu is stronger then the X1 but dont ignore the new facts about the PS4 is suppose to support voice and facial recognition and that will eat processing cycles as well. We have no real idea how much of the PS4's gpu will be allocated for other jobs. So the difference between the two may be smaller then some think.

tormentos

 

 

Bonaire and PITCAIRN and not the samejust like cape verde and Bonaire aren't.

There is no 128 bit bus limitation it doesn't exist,should i post how much bullsh** that claim is.?

The 7950 has a higher bit bus than the 660Ti also almost 100GB/s more bandwidth 144 for the 660TI vs the 7950 240,and yet they both perform almost the same.

 

The PS eye is not mandatory there is nothing to eat dude,Kinect is build in into the software and is included,the ps eye is not and has not reservation what so ever,just like the PS3 didn't reserve anything for the PS eye either some of you are just graspping..

:lol:

 Bonaire is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture.

 The Cape Verde core is the smallest, with a max of 640 stream processors, a 128-bit memory interface, and a single primitive pipeline (maximum process rate of one primitive per clock). This cut-down design didnt work very well

You can not directly compare a Nvidia gpu vs an AMD gpu, both beands handle processors and memory differently :roll:

Avatar image for 04dcarraher
04dcarraher

23857

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#347 04dcarraher
Member since 2004 • 23857 Posts

[QUOTE="04dcarraher"]

 

8-15 will the average difference at best. Bonaire is not the same as the 7770, 7790 uses the Bonaire and is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture as Tahiti and Pitcairn top is Bonaire,Tahti and Pitcaim while bottom is the 7770 the 7770 core is the smallest, with a max of 640 stream processors, and a single primitive pipeline. [img] http://www.extremetech.com/wp-content/uploads/2013/03/GPU-Frontends.png[img]

There is a thing called optimization customizing the settings and assets to fit the hardware. Heck BF3 with a 7770 max settings at 1280x800 with 2x MSAA, gets 51 fps average, while 7790 gets 65 FPS average. expect the X1 to be like the 7790 with 7770 processing power overall without the inherit 128 bus limits and 7770 flaws. We all know the PS4 gpu is stronger then the X1 but dont ignore the new facts about the PS4 is suppose to support voice and facial recognition and that will eat processing cycles as well. We have no real idea how much of the PS4's gpu will be allocated for other jobs. So the difference between the two may be smaller then some think.

tormentos

 

 

Bonaire and PITCAIRN and not the samejust like cape verde and Bonaire aren't.

There is no 128 bit bus limitation it doesn't exist,should i post how much bullsh** that claim is.?

The 7950 has a higher bit bus than the 660Ti also almost 100GB/s more bandwidth 144 for the 660TI vs the 7950 240,and yet they both perform almost the same.

 

The PS eye is not mandatory there is nothing to eat dude,Kinect is build in into the software and is included,the ps eye is not and has not reservation what so ever,just like the PS3 didn't reserve anything for the PS eye either some of you are just graspping..

:lol:

 Bonaire is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture.

all vs 7770

GPU-Frontends.png

 The Cape Verde core is the smallest, with a max of 640 stream processors, a 128-bit memory interface, and a single primitive pipeline (maximum process rate of one primitive per clock). This cut-down design didnt work very well

You can not directly compare a Nvidia gpu vs an AMD gpu, both beands handle processors and memory differently :roll:

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#348 ronvalencia
Member since 2008 • 29612 Posts

[QUOTE="ronvalencia"]

 Windows 8 acceleration is done by on-demand basis i.e. sleeps during full screen game mode.tormentos

 

Windows 8 has  GPU acelerated UI..Nothing more to say,and the xbox one has a snap feature that requires GPU resources as well,so yeah there you have it 10% confirmed..

Look at it from the bright side,the xbox one UI shoulod be way faster and smooth than the PS4 one.

Well i think..

Still zero precent GPU usage with Windows 8's snap feature. Radeon HD 6250M says Hi.
Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#349 tormentos
Member since 2003 • 33793 Posts

 

:lol:

 Bonaire is built on the same 28nm process technology as AMDs other GCN parts and it uses the same architecture.

 The Cape Verde core is the smallest, with a max of 640 stream processors, a 128-bit memory interface, and a single primitive pipeline (maximum process rate of one primitive per clock). This cut-down design didnt work very well

You can not directly compare a Nvidia gpu vs an AMD gpu, both beands handle processors and memory differently :roll:

04dcarraher

 

2zpp7bq.jpg

 

No...

 

Ultra close-up photography of AMD's Pitcairn graphics processor (left) and its less capable sibling, Bonaire (right). We strongly suspect that the GPU tech of the PS4 is based on the former, with Xbox One's graphics hardware remarkably similar to the latter.

 

http://www.eurogamer.net/articles/digitalfoundry-can-xbox-one-multi-platform-games-compete-with-ps4

 

Not the same.

Avatar image for StrongBlackVine
StrongBlackVine

13262

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#350 StrongBlackVine
Member since 2012 • 13262 Posts

Ryse is running 900p now. Coincidence....I think not.