NGGP: NextGen Garbage Pile (aka: No one reads the topics or stays on topic) *spawn*

[XpiderMX]

No, but 16 vs 32 rops, and 12 vs 18 CUs can.

There is no %&"#& way making the 4 CUs special will make them less good than just not mucking with them. If it would, they wouldn't do it. For multiplat titles, the only thing that is easy to scale is the graphics -- if Durango isn't running physics or whatever on CUs, Orbis won't be doing that either. So when comparing multiplats 12 vs 18 is the correct number.

Also, for more pixels on the screen, rops are maybe even more important than CUs. Twice the pixels always means twice the load on the rops, but the part of CU workload that is used on geometry doesn't get any harder when res gets higher.

4CUs are the "special sauce" for PS4?

 

[bagofsuck]

please don't kill me for this question

If durango has 12CUs and orbis has 14CUs + 4 reserved for X reason.. lets say these 4 CUs aren't for rendering/graphics etc... it's going to make that much different? and i know we also have the factor of 16 rops vs 32 rops thanks

 

[xJumPeRJumPzZ]

it's already stated sooooo many times, that 4 CU in orbis are not for graphic, because they can improve only slightly, this is why they are destinated to other tasks

so are 12 vs 14 CU,

the
same
thing

1080P is 2.073.600 pixels
720P is 921.600 pixels

14 CU over 12 CU = +16%
1080P over 720P = + 125%

so can finally end there the false statement "orbis is 1080p where durango is 720p" or we have to explain why a total workload of 225% can't be done with only more 16% CU units PLEASE?

and don't talk about bandwidth, because unltil contrary trial, they have both 170 GB/s, one from gddr and the other with combined/tiled ram-esram



NO, their ROPs don't work in the same way, if I remember well orbis can't resolve in eSRAM and can't do hardware tiling on this

and durango runs physics on CPU.

Sadly not so many people will see that post or accept this. If its really only 16% diffrence between the two, then who cares, nobody wins. So it would be basicly the same situation as current gen. Ps3 here and there a little bit better and 360 on other things. I can live with that
But i'm going to wait until final judgement when they announce the real specs.

 

[Averagejoe]

1080P (Orbis) vs 720P (Durango)? Are you sure?

Can 12CUs vs 14CUs make this kind of differences?

Aren't you ignoring 4 extra CU con orbis?

Anything done on that has to be done usinh CPU resources and GPU resources as well so those 12 vs 14 could more like 9 vs 14.

 

[XpiderMX]

Aren't you ignoring 4 extra CU con orbis?

Anything done on that has to be done usinh CPU resources and GPU resources as well so those 12 vs 14 could more like 9 vs 14.

Two HD cameras image processing (PS4)?

better CPU (custom Jaguar) for Durango?

Who know? maybe best physics on first party PS4 games, and same third party games on both (same resolution, same physics).

You are ignoring that we don't know all details from Durango. Durango has special hardware for Kinect processing too.

 

[Xenio]

Aren't you ignoring 4 extra CU con orbis?

Anything done on that has to be done usinh CPU resources and GPU resources as well so those 12 vs 14 could more like 9 vs 14.

again?
doesn't matter how many times a thing will be explained.

try to explain what do you don't understand of "the 4 CU's will be not used for graphics because they can help only slightly."

and for a 1000th time, what they do can be done more simple in cpu (rumored customized version of the orbis one, how much custom nobody knows yet)

this
don't
change
the
target
resolution

you can see 1280x1080 in durango and 1485x1080 in orbis (+16%) and never spot the difference
or you can do 1080P @40FPS (capped to 30) in durango and 1080P46 FPS (+16%) capped to 30 in orbis and again, no difference visible

 

[Averagejoe]

it's already stated sooooo many times, that 4 CU in orbis are not for graphic, because they can improve only slightly, this is why they are destinated to other tasks

so are 12 vs 14 CU,

the
same
thing

1080P is 2.073.600 pixels
720P is 921.600 pixels

14 CU over 12 CU = +16%
1080P over 720P = + 125%

so can finally end there the false statement "orbis is 1080p where durango is 720p" or we have to explain why a total workload of 225% can't be done with only more 16% CU units PLEASE?

and don't talk about bandwidth, because unltil contrary trial, they have both 170 GB/s, one from gddr and the other with combined/tiled ram-esram



NO, their ROPs don't work in the same way, if I remember well orbis can't resolve in eSRAM and can't do hardware tiling on this

and durango runs physics on CPU.

Now i debate this with you already,those 4 cu are no useless or close,any developer using those for physics,animations,lighting and so on will have to use CPU and GPU resources con durando to keep up.

You don't know how capable those cu are,and trying to mean they will not be an advantage is silly.

Anything done on those 4 CU will require resources from durando to be able to duplicate.

 

[Xenio]

Now i debate this with you already,those 4 cu are no useless or close,any developer using those for physics,animations.

they can be useful, maybe will be used to extract 3D data of the envirorment from the dual camera, while kinect have its compute block, who knows, glad if they do amazing physics or wonderful animations but

but will be NOT used for graphics, the rumors says clearly that those CU will not give a noticeable contribute to graphics
so please stop here the stupid statement that orbis is 1080P where durango is 720P, because doesn't have any sense at all

 

[Delta9]

it's already stated sooooo many times, that 4 CU in orbis are not for graphic, because they can improve only slightly, this is why they are destinated to other tasks

so are 12 vs 14 CU,

the
same
thing

1080P is 2.073.600 pixels
720P is 921.600 pixels

14 CU over 12 CU = +16%
1080P over 720P = + 125%

so can finally end there the false statement "orbis is 1080p where durango is 720p" or we have to explain why a total workload of 225% can't be done with only more 16% CU units PLEASE?

and don't talk about bandwidth, because unltil contrary trial, they have both 170 GB/s, one from gddr and the other with combined/tiled ram-esram



NO, their ROPs don't work in the same way, if I remember well orbis can't resolve in eSRAM and can't do hardware tiling on this

and durango runs physics on CPU.

In that blog post by Lottes he said the 32mb of esram wouldn't be enough for good AA at 1080P.

 

[Betanumerical]

again?
doesn't matter how many times a thing will be explained.

try to explain what do you don't understand of "the 4 CU's will be not used for graphics because they can help only slightly."

and for a 1000th time, what they do can be done more simple in cpu (rumored customized version of the orbis one, how much custom nobody knows yet)

I think thats a bad quote, and or written badly. They would not take out anything from the CU's that would not make any sense, instead they would add stuff to them. That would make them just as good for graphics.

Also you cannot simply add up bandwidths like that in all scenarios it doesnt work like that at all.

Orbis has twice rops (fill) off Durango, and the same amount of bandwidth, that ones a pretty simple equation.

 

[Xenio]

In that blog post by Lottes he said the 32mb of esram wouldn't be enough for good AA at 1080P.

strange, 32 MB are enought for a 1080P 2xMSAA framebuffer
and durango do tiling in hardware, so it's not limited to physical 32 MB

search "tile based deferred engines"
but forward engines too

 

[Averagejoe]

again?
doesn't matter how many times a thing will be explained.

try to explain what do you don't understand of "the 4 CU's will be not used for graphics because they can help only slightly."

and for a 1000th time, what they do can be done more simple in cpu (rumored customized version of the orbis one, how much custom nobody knows yet)

this
don't
change
the
target
resolution

you can see 1280x1080 in durango and 1485x1080 in orbis (+16%) and never spot the difference
or you can do 1080P @40FPS (capped to 30) in durango and 1080P46 FPS (+16%) capped to 30 in orbis and again, no difference visible

Wow dude they have mínima impact rendering,now use for any other type of thing like physics will eat durando resources is imposible that you ignore this imposible.


You talk like if those 4 CU units were good for nothing but durando ones will be increíble forma no apparent reason that is wishful thinking.

 

[Xenio]

I think thats a bad quote, and or written badly. They would not take out anything from the CU's that would not make any sense, instead they would add stuff to them. That would make them just as good for graphics.

maybe are you misinformed?

"About 14 + 4 balance:

- 4 additional CUs (410 Gflops) “extra” ALU as resource for compute

- Minor boost if used for rendering"

http://www.vgleaks.com/world-exclusive-orbis-unveiled/

Also you cannot simply add up bandwidths like that in all scenarios it doesnt work like that at all.

have you missed all the talk about this, have you?

 

[Delta9]

strange, 32 MB are enought for a 1080P 2xMSAA framebuffer
and durango do tiling in hardware, so it's not limited to physical 32 MB

search "tile based deferred engines"
but forward engines too

Here is the quote.

A fast GDDR5 will be the desired option for developers. All the interesting cases for good anti-aliasing require a large amount of bandwidth and RAM. A tiny 32MB chunk of ESRAM will not fit that need even for forward rendering at 1080p. I think some developers could hit 1080p@60fps with the rumored Orbis specs even with good AA. My personal project is targeting 1080p@60fps with great AA on a 560ti which is a little slower than the rumored Orbis specs. There is no way my engine would hit that target on the rumored 720 specs. Ultimately on Orbis I guess devs target 1080p/30fps (with some motion blur) and leverage the lower latency OS stack and scan out at 60fps (double scan frames) to provide a really great lower-latency experience. Maybe the same title on 720 would render at 720p/30fps, and maybe Microsoft is dedicating a few CPU hardware threads to the GPU driver stack to remove the latency problem (assuming this is a "Windows" OS under the covers).

 

[Betanumerical]

maybe are you misinformed?


http://www.vgleaks.com/world-exclusive-orbis-unveiled/



have you missed all the talk about this, have you?

Awesome.

1.84TFLOPS from 14 CU's, the entire GPU is most certainly custom then.

It probably says minor boost for rendering because it only represents a small amount of the power of the GPU over all, or maybe it means a minor boost over the rest of the CU's, per CU.

Explain to me how when you have a really really big assets that just don't fit in ESRAM at all how you are going to get the new data from the DDR3 to the ESRAM without using the DDR3 to transfer it, cause that would stop other things accessing the DDR3 and also eat up the DDR3 bandwidth.

The bandwidth in the two is NOT equal, you are spouting nonsense please stop it this is a technical site.

 

[XpiderMX]

Awesome.

1.84TFLOPS from 14 CU's, the entire GPU is most certainly custom then.

It probably says minor boost for rendering because it only represents a small amount of the power of the GPU over all, or maybe it means a minor boost over the rest of the CU's, per CU.

Explain to me how when you have a really really big assets that just don't fit in ESRAM at all how you are going to get the new data from the DDR3 to the ESRAM without using the DDR3 to transfer it, cause that would stop other things accessing the DDR3 and also eat up the DDR3 bandwidth.

The bandwidth in the two is NOT equal, you are spouting nonsense please stop it this is a technical site.

1.84TFlops from 18CUs, not 14CUs.

 

[Xenio]

1.84TFlops from 18CUs, not 14CUs.

we have two option, or he doesn't want to accept it or he doesn't understand it.

 

[Betanumerical]

we have two option, or he doesn't want to accept it or he doesn't understand it.

You are saying they give a minor boost to rendering, and that they represent 4, of the 18 CU's.

Therefore we can conclude that the majority of the power comes from the 14 CU's meaning they probably provide around 1.6TFLOPS just from the 14.

Or you can concede they provide the same power as the rest. Your choice.

 

[Averagejoe]

maybe are you misinformed?


http://www.vgleaks.com/world-exclusive-orbis-unveiled/



have you missed all the talk about this, have you?

Rendering minor boost if use for rendering.. You know what means?

What if the 4CU are use for cumpute? Say physics with what durando will compensate.? Oh yeah with the cpu,dude those 4 CU units alone probably have more gFLOPS than durando CPU by a good margin.

Before you keep downplaying those 410 GFLOPS mind you the complete xbox 360 doesn't have 410 GFLOPS dude so don't tell me they are no difference makers.

 

[temesgen]

maybe are you misinformed?


http://www.vgleaks.com/world-exclusive-orbis-unveiled/



have you missed all the talk about this, have you?

Have you ever considered taking a break? Your arguments are almost predictable at this point. You are always glass half full for anything pro XBOX and almost anything which lends itself to Sony having a more powerful machine is shot down as hearsay, speculation, bad info. To put it another way 1+1 = 3 for for one side and 1+1 = show me the proof that 2 is the answer or else if its Sony. Its almost comical.

IMO both Machines are based around the priorities/business plans of the their respective companies and we will have to see who ends up gauging the market better. MS showed last time a better overall vision and their execution was better so they grew market share. Sony appears to have learned some hard lessons from last time but even if their hardware is superior they will need to build out better OS and platform for online and more importantly figure out how to make money and even then if they release another machine like the PS3 which has constant software updates, doesn't offer a consistent baseline of features for MP they are going to struggle.

Graphics and overall power didn't dictate success or failure this round and it isn't likely that either will be the deciding factor next time either.