Fiji’s Architecture: The Grandest of GCN 1.2

We’ll start off our in-depth look at the R9 Fury X with a look at the Fiji GPU underneath.

Like the Hawaii GPU before it, from a release standpoint Fiji is not really the pathfinder chip for its architecture, but rather it’s the largest version of it. Fiji itself is based on what we unofficially call Graphics Core Next 1.2 (aka GEN3), and ignoring HBM for the moment, Fiji incorporates a few smaller changes but otherwise remaining nearly identical to the previous GCN 1.2 chips. The pathfinder for GCN 1.2 in turn was Tonga, which was released back in September of 2014 as the Radeon R9 285.

So what does GCN 1.2 bring to the table over Hawaii and the other GCN 1.1 chips? Certainly the most well-known and marquee GCN 1.2 feature is AMD’s latest generation delta color compression technology. Tied in to Fiji’s ROPs, delta color compression augments AMD’s existing color compression capabilities with additional compression modes that are based around the patterns of pixels within a tile and the differences between them (i.e. the delta), increasing how frequently and by how much frame buffers (and RTs) can be compressed.

Frame buffer operations are among the most bandwidth intensive in a GPU – it’s a lot of pixels that need to be resolved and written to a buffer – so reducing the amount of memory bandwidth these operations draw on can significantly increase the effective memory bandwidth of a GPU. In AMD’s case, GCN 1.2’s delta color compression improvements are designed to deliver up to a 40% increase in memory bandwidth efficiency, with individual tiles being compressible at up to an 8:1 ratio. Overall, while the lossless nature of this compression means that the exact amount of compression taking place changes frame by frame, tile by tile, it is at the end of the day one of the most significant improvements to GCN 1.2. For Radeon R9 285 it allowed AMD to deliver similar memory performance on a 256-bit memory bus (33% smaller than R9 280’s), and for Fiji it goes hand-in-hand with HBM to give Fiji an immense amount of effective memory bandwidth to play with.

Moving on, AMD has also made some changes under the hood at the ALU/shader level for GCN 1.2. Many of these changes are primarily for AMD’s Carrizo APU, where task scheduling improvements go hand-in-hand with the AMD’s Heterogeneous System Architecture initiative and deliver improvements to allow the CPU and GPU to more easily deliver work to each other. Similarly, 16-bit instructions are intended to save on power consumption in mobile devices that use lower precision math for basic rendering.

More applicable to Fiji and its derivatives are the improvements to data-parallel processing. GCN 1.2 now has the ability for data to be shared between SIMD lanes in a limited fashion, beyond existing swizzling and other data organizations methods. This is one of those low-level tweaks I’m actually a bit surprised AMD even mentioned (though I’m glad they did) as it’s a little tweak that’s going to be very algorithm specific. For non-programmers there’s not much to see, but for programmers – particularly OpenCL programmers – this will enable newer, more efficient algorithms where when the nature of the work requires working with data in adjacent lanes.

But for gamers, perhaps the most significant architectural improvement to GCN 1.2 and thereby Fiji are the changes made to tessellation and geometry processing. There is no single silver bullet here – after going with a 4-wide geometry front-end in Hawaii, AMD hasn’t changed it for Tonga or Fiji – but AMD has put in quite a bit of effort in to improving how geometry data moves around within the chip and how it’s used, on the basis that at this point the limitations aren’t in raw geometry performance, but rather the difficulties in achieving that performance.

Much of this effort has been invested in better handling small geometry, whether it’s large quantities of small batches, or even small quantities of small batches. The inclusion of small instance caching, for example, allows the GPU to better keep small batches of draw calls in cache, allowing them to be referenced and/or reused in the future without having to go to off-cache memory. Similarly, AMD can now store certain cases of vertex inputs for the geometry shader in shared memory, which like small instance caching allows for processing to take place more frequently on-chip, improving performance and cutting down on DRAM traffic.

More specific to Fiji’s incarnation of GCN is how distribution is handled. Load balancing and distribution among the geometry frontends is improved overall, including some low-level optimizations to how primitives generated from tessellation are distributed. Generally speaking distribution is a means to improve performance by removing bottlenecks, however AMD is now catching a specific edge case where small amplification factors don’t generate a lot of primitives, and in those cases they’re now skipping distribution since the gains are minimal, and more likely than not the cost from the bus traffic is greater than the benefits of distribution.

Finally, AMD has also expanded the vertex reuse window on GCN 1.2. As in the general case of reuse windows, the vertex reuse window is a cache of sorts for vertex data, allowing old results to be held in waiting in case they are needed again (as is often the cases in graphics). Though they aren’t telling us just how large the window now is, GCN 1.2 now features a larger window, which increases the hit rate for vertex data and as a result further edges geometry performance up since that data no longer needs to be regenerated.

As with our R9 285 review, I took the time to quickly run TessMark across the x8/x16/x32/x64 tessellation factors just to see how tessellation and geometry performance scales on AMD’s cards as the tessellation factor increases. Keeping in mind that all of the parts here have a 4-wide geometry front-end, the R9 285, R9 290X, and R9 Fury X all have the same geometry throughput on paper, give or take 10% for clockspeeds. What we find is that Fury X shows significant performance improvements at all levels, beating not only the Hawaii based R9 290X, but even the Tonga based R9 285. Tessellation performance is consistently 33% ahead of the R9 290X, while against Tonga it’s anywhere between a 33% lead at high factors to a 130% lead at low tessellation factors, showing the influence of AMD’s changes to how tessellation is handled with low factors.

The AMD Radeon R9 Fury X Review The Fiji GPU: Go Big or Go Home
Comments Locked

458 Comments

View All Comments

  • Innokentij - Sunday, July 5, 2015 - link

    Dont know how i missed this, thanks alot!
  • toyotabedzrock - Friday, July 3, 2015 - link

    You know I think you could fit a few more ads as content is still visible.
    That said, the interposer could be made in two or for smaller dies and just fab the phy on the interposer.
  • Oxford Guy - Friday, July 3, 2015 - link

    "There is no getting around the fact that NVIDIA’s Maxwell 2 GPUs are very well done, very performant, and very efficient"

    Not the 970. 28 GB/s VRAM bandwidth with XOR contention is the antithesis of efficiency.
  • samer1970 - Friday, July 3, 2015 - link

    If AMD sells this Card for $550 they will win this round granted.

    besides this is the only card with high performance that fits in a small case.

    as for the power , we have 700 watts SFX from Silverstone and it is enough for 2 cards in small mAtx case.

    The ricing isnt just right thats all . $650 is too much for this card for competition.
  • samer1970 - Friday, July 3, 2015 - link

    AMD can still win this round by making a 3xGPU single card , given he space Advantage that allows them to make a single card 3 or 4 GPU onboard.

    the only problem would be the power for the 4 GPU single card ... so I guess they will stop at 3xGPUs on single card.

    nvidia cant do that .
  • Ian Cutress - Friday, July 3, 2015 - link

    If you have multiple GPU silicon on a card, you still have to have a PCIe switch in order to navigate the PCIe 3.0 x16 input into the card. You can see that the dual GPU card has a PLX8747 on it, which splits the PCIe 3.0 x16 lanes from the CPU into two lots of x16, one of each GPU (PLX chips do multiplexing with a FIFO buffer). Having three or four GPUs means you have to look into more expensive PCIe switches, like the PLX8764 or the PLX 8780, then provide sufficient routing which most likely adds PCB layers.
  • Oxford Guy - Friday, July 3, 2015 - link

    You really should have made the point that, in order to get similar load noise level out of the 980 Ti, you're looking at greater expense because the cooler will need to be upgraded. That extra expense could be quite significant, especially if buyers opt for a closed loop cooler.
  • Oxford Guy - Friday, July 3, 2015 - link

    The other point is that AMD had better offer this card without the CLC so people who run water loops won't have to buy a fancy CLC that they're not going to use.
  • meacupla - Friday, July 3, 2015 - link

    But you would just get an air cooler that you're not going to use.
    And I doubt there would be any price difference, because they'll get beefy air coolers.
  • Oxford Guy - Friday, July 3, 2015 - link

    sell the card bare

Log in

Don't have an account? Sign up now