AMD Volcanic Islands to Implement CPU Cores

Add Your Comments

AMD’s upcoming generation of graphics cards, dubbed Volcanic Islands, is to include CPU cores on-chip.

AMD's Volcanic Islands - Image Courtesy ChipHell

AMD’s Volcanic Islands – Image Courtesy ChipHell

According to website TechPowerUp!, engineers at AMD have been working on implementing CPU cores in their new GPU design. The result would be a clear departure from previous VLIW4, VLIW5 and GCN, which were all evolutionary architectures instead of complte redesigns. Chinese website Chiphell has leaked what is claimed to be a block-diagram of what could be, by current nomenclature, the AMD HD8970. Codenamed Hawaii, the new Graphics Processing Unit is shown implementing two distinct processing modules: A Parallel Compute Module, which looks like a standard GPU, and a Serial Processing Module, which would take care of serial compute tasks as a CPU would.

The classic GPU part would be based on 16 SPUs for a total of 4096 shader processors, along with 256 TMUs (textue mapping units) and 64 ROPs (render output units). Those numbers represent exactly twice as much theoretical processing power as was present in Tahiti XT, the chip behind Radeon HD7970. It is quite unclear how much CPU power will be present on the chip, though, as the block-diagram is fuzzy and no detailed specifications have been posted. What we can observe are 8 modules that resemble AMD’s current high-performance x86 design, with a shared central floating point unit in each module and shared L3 cache, which would add up to a 16-cores CPU. The whole chip, including CPU and GPU parts, is connected to memory through a 512-bit wide bus.

It is worth noting that AMD also signed a licensing agreement with ARM last year, which could mean that the possibility of ARM cores on AMD chips is not to be excluded. We know from previous announcements from Nvidia that their upcoming all-new architecture will also feature on-die CPU, this time confirmed as an ARMv8 part. Given the timeframe, both Maxwell and AMD’s Volcanic Islands series of GPUs are expected to be built on TSMC’s 20nm bulk process. They should bring enhanced power efficiency, thanks in part to the new 20nm process, and ease of programmability through integrated CPU cores; although we can’t know for sure which one of the designs will prevail in each of the new features.

The next round of the exciting and seemingly never-ending graphics war is to unfold in the last weeks of this year, TSMC’s technology allowing.

dostov

Leave a Reply

Your email address will not be published. Required fields are marked *

Discussion
  1. ATMINSIDE
    So, they've been including GPUs with their CPUs, now they're including CPUs with their GPUs?
    Yo dawg, I heard you like...

    lol i was just about to post this
    Boulard83
    +1
    Just hope they can make drivers that works and manage it correctly now ...

    pfffttt keep dreamin!
    scary thing is nvidia is doing it too hopefully one of them can figure it out!
    perhaps it handles the extra computations that the games need such as location of objects and leaving the cpu free for more taxing things? or perhaps its a more advanced way of thermal / power management?
    I thought ARM chips had trouble with something, Windows, X86, X64, something, what was it? It was the reason Microsoft had to make Windows 8 RT for tablets that cannot run 'legacy' (i.e. normal desktop) apps. What would stop the same issue from occuring here?
    And what is the point of having CPU cores on a GPU? I missed that about this article.
    Theres alot of potential uses for a multi purpose core on something as complex as these modern GPUs. Its also not important for them to be able to run standard code as they are likely not intended to be used for anything outside of the GPU. They are being labled as "serial processors" I imagine that these are for internal scene management/iterative type of code while the 4k parallel processors spend all of there time pumping out pixels.
    Culbrelai
    I thought ARM chips had trouble with something, Windows, X86, X64, something, what was it? It was the reason Microsoft had to make Windows 8 RT for tablets that cannot run 'legacy' (i.e. normal desktop) apps. What would stop the same issue from occuring here?
    And what is the point of having CPU cores on a GPU? I missed that about this article.

    See Bobnova's post below.
    Culbrelai
    I thought ARM chips had trouble with something, Windows, X86, X64, something, what was it? It was the reason Microsoft had to make Windows 8 RT for tablets that cannot run 'legacy' (i.e. normal desktop) apps. What would stop the same issue from occuring here?
    And what is the point of having CPU cores on a GPU? I missed that about this article.

    ARM chips can't run standard Windows because they are not x86 chips. Windows runs exclusively within an x86 environment. But that is not a problem here and those cores could be anything.
    As for the uses of a CPU-type multi-purpose processor on GPU, they are numerous as others have stated, but will also tremendously streamline programming and increase performance of GPGPU applications (general purpose computation on GPU).
    These new GPUs will shine in everything from gaming to computing programs like Folding@Home, and might even take a more important place in HPC and servers.
    CPU cores, and twice as powerful as the Tahiti core to boot!
    I'd like to see AMD use those CPU cores to improve microstuttering and frame pacing in Crossfire setups.
    And I'd like to see some modders set up CPU Physx on the GPU cores... :P
    Theres alot of potential uses for a multi purpose core on something as complex as these modern GPUs. Its also not important for them to be able to run standard code as they are likely not intended to be used for anything outside of the GPU. They are being labled as "serial processors" I imagine that these are for internal scene management/iterative type of code while the 4k parallel processors spend all of there time pumping out pixels.

    Oh... then the normal CPU would be doing what... controlling AI?
    It's the fundamental difference between ARM and desktop CPUs. They're both processors, they both calculate things, but desktop CPUs use the x86 instruction set (which is horribly archaic) while ARM uses the RISC instruction set (ARM standing for Advanced RISC Machines).

    Why doesn't ARM make desktop CPUs if RISC is so much 'fresher'
    ARM chips can't run standard Windows because they are not x86 chips. Windows runs exclusively within an x86 environment. But that is not a problem here and those cores could be anything.

    What does 64 bit windows do differently, then?
    Does the CPU/GPU combo chip have to interact with the normal main CPU at all? Pretty interesting stuff it just intriuiges me is all.