The benchmark compared single parts, so do I. Don't tell me that inferring overall performance off this as I know that this is wrong. Moreover it is wrong to infer real world performance by specs as often done even on this forum. Hardware specs on paper don't have a direct relationship to performance in games. This has been done here so many times and it is so wrong that it hurts to even read it.

But no matter what, the slides don't offer enough information to really start a discussion here, anyway.

"The three "major modifications" Sony did to the architecture to support this vision are as follows, in Cerny's words:

• "First, we added another bus to the GPU that allows it to read directly from system memory or write directly to system memory, bypassing its own L1 and L2 caches. As a result, if the data that's being passed back and forth between CPU and GPU is small, you don't have issues with synchronization between them anymore. And by small, I just mean small in next-gen terms. We can pass almost 20 gigabytes a second down that bus. That's not very small in today’s terms -- it’s larger than the PCIe on most PCs!

• "Next, to support the case where you want to use the GPU L2 cache simultaneously for both graphics processing and asynchronous compute, we have added a bit in the tags of the cache lines, we call it the 'volatile' bit. You can then selectively mark all accesses by compute as 'volatile,' and when it's time for compute to read from system memory, it can invalidate, selectively, the lines it uses in the L2. When it comes time to write back the results, it can write back selectively the lines that it uses. This innovation allows compute to use the GPU L2 cache and perform the required operations without significantly impacting the graphics operations going on at the same time -- in other words, it radically reduces the overhead of running compute and graphics together on the GPU."

• Thirdly, said Cerny, "The original AMD GCN architecture allowed for one source of graphics commands, and two sources of compute commands. For PS4, we’ve worked with AMD to increase the limit to 64 sources of compute commands -- the idea is if you have some asynchronous compute you want to perform, you put commands in one of these 64 queues, and then there are multiple levels of arbitration in the hardware to determine what runs, how it runs, and when it runs, alongside the graphics that's in the system."

"The reason so many sources of compute work are needed is that it isn’t just game systems that will be using compute -- middleware will have a need for compute as well. And the middleware requests for work on the GPU will need to be properly blended with game requests, and then finally properly prioritized relative to the graphics on a moment-by-moment basis."

This concept grew out of the software Sony created, called SPURS, to help programmers juggle tasks on the CELL's SPUs -- but on the PS4, it's being accomplished in hardware.

The team, to put it mildly, had to think ahead. "The time frame when we were designing these features was 2009, 2010. And the timeframe in which people will use these features fully is 2015? 2017?" said Cerny.

"Our overall approach was to put in a very large number of controls about how to mix compute and graphics, and let the development community figure out which ones they want to use when they get around to the point where they're doing a lot of asynchronous compute."

Cerny expects developers to run middleware -- such as physics, for example -- on the GPU. Using the system he describes above, you can run at peak efficiency, he said.

"If you look at the portion of the GPU available to compute throughout the frame, it varies dramatically from instant to instant. For example, something like opaque shadow map rendering doesn't even use a pixel shader, it’s entirely done by vertex shaders and the rasterization hardware -- so graphics aren't using most of the 1.8 teraflops of ALU available in the CUs. Times like that during the game frame are an opportunity to say, 'Okay, all that compute you wanted to do, turn it up to 11 now.'"

http://www.gamasutra.com/view/feature/191007/inside_the_playstation_4_with_mark_.php

You don't need to post old news to reply to my post. If you read and understand what's in the text you took as a reply to me you would see that my post is still correct. There is no "unused" hardware in PS4 lying around, waiting to be used by GPGPU. It is the same hardware that is now doing rendering tasks.

I find the fact that the consoles are way, way faster than last gen more interesting than the fact that the PS4 is better than the XOne.

Obviously I have to - if you actualy understood what is in the text, you would know what async compute is and how it runs in paralel with graphics tasks. Maybe you should re-read whole interview again bit more carefully.

I agree with you.

''Playstation 4 chief architect Mark Cerny expressed a similar vision over the potential of GPU compute during past interviews leading up to the console’s launch. He highlighted asynchronous fine-grain compute as a key feature that will give developers the headroom to explore the platform for years to come. Cerny explained that there are many situations during game applications where functions pertaining to graphics take a backseat, and that the PS4 hardware has been designed to make the most of the under-utilized portions of the GPU during such frames.''

''“If you look at the portion of the GPU available to compute throughout the frame, it varies dramatically from instant to instant. For example, something like opaque shadow map rendering doesn’t even use a pixel shader, it’s entirely done by vertex shaders and the rasterization hardware — so graphics aren’t using most of the 1.8 teraflops of ALU available in the CUs. Times like that during the game frame are an opportunity to say, ‘Okay, all that compute you wanted to do, turn it up to 11 now.”

“If you look at how the GPU and its various sub-components are utilised throughout the frame, there are many portions throughout the frame – for example during the rendering of opaque shadowmaps – that the bulk of the GPU is unused. And so if you’re doing compute for collision detection, physics or ray-casting for audio during those times you’re not really affecting the graphics. You’re utilising portions of the GPU that at that instant are otherwise under-utilised. And if you look through the frame you can see that depending on what phase it is, what portion is really available to use for compute.”

Both Katsman and Cerny are essentially talking about the same process of performing non-graphics related tasks in parallel by utilizing wasted compute resources in the GPU.

http://gearnuke.com/mantle-to-bring-ps4-asynchronous-fine-grain-compute-to-pc/

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

It's not enough to read something, it must be understood.

It will take years before Devs will show what PS4 can really do.  Uncharted4 will be the 1st Game which will show something. And the 2nd Game by Naughty Dog on PS4 will be the witness of what the console can do.

Not at all, the rendering tasks, both can be done at the same time. This is why Cerny worked so hard on this

Someone already explained that  it seems.

You do know Async is not a magic bullet.  GPGPU and the GPU are fighting for the same resources.  Yes both can run at the same time but if resources are limited or there is a bottleneck in the process you might not see great gains or performance.   Here is a very important part of that interview that needs to be understood.

""The reason so many sources of compute work are needed is that it isn’t just game systems that will be using compute -- middleware will have a need for compute as well. And the middleware requests for work on the GPU will need to be properly blended with game requests, and then finally properly prioritized relative to the graphics on a moment-by-moment basis.""

Short statement but for the developer its a lot of work and would be very specific for each game.  Using GPGPU compute in parallel is not something that developers just code for it and it just works.  From the statement above, there is a lot of understanding of game state where you setting priorities on what get resources and when.  Scheduling those calls can be a real pain and syncing everything together probably is a nightmare unless you really need to go down that route for more gains in performance.  From this point of view, it definitely something that will need to be planned out way ahead of time and developers probably have to prototype some stuff first before wasting time in development if there are no real gains in performance.

PS4 is built by game developers, and made for gamers.

Async compute is nothing new - CPUs do it all the time when context switching. But still you are going to use the same resources in rendering, as there is not much time to "context switch". All the time frames want to be rendered, where do you want to get the time for doing GPGPU with the same resources?

What they are mostly saying is that you can use the resources for rendering and GPGPU at the same time but if you do GPGPU then these resources do GPGPU and not rendering work. There is no duplicated hardware present.

As Nate4Drake already quoted:

"Cerny explained that there are many situations during game applications where functions pertaining to graphics take a backseat, and that the PS4 hardware has been designed to make the most of the under-utilized portions of the GPU during such frames.''

When will this be? In a graphical demanding game this will not likely happen. And when it does what do you want to compute then? A game has to be developed that "if close to nothing happens" to compute but what for?