A Bit More On Graphics Core Next 1.1

With the launch of Hawaii, AMD is finally opening up a bit more on what Graphics Core Next 1.1 entails. No, they still aren’t giving us an official name – most references to GCN 1.1 are noting that 290X (Hawaii) and 260X (Bonaire) are part of the same IP pool – but now that AMD is in a position where they have their new flagship out they’re at least willing to discuss the official feature set.

So what does it mean to be Graphics Core Next 1.1? As it turns out, the leaked “AMD Sea Islands Instruction Set Architecture” from February appears to be spot on. Naming issues with Sea Islands aside, everything AMD has discussed as being new architecture features in Hawaii (and therefore also in Bonaire) previously showed up in that document.

As such the bulk of the changes that come with GCN 1.1 are compute oriented, and clearly are intended to play into AMD’s plans for HSA by adding features that are especially useful for the style of heterogeneous computing AMD is shooting for.

The biggest change here is support for flat (generic) addressing support, which will be critical to enabling effective use of pointers within a heterogeneous compute context. Coupled with that is a subtle change to how the ACEs (compute queues) work, allowing GPUs to have more ACEs and more queues in each ACE, versus the hard limit of 2 we’ve seen in Southern Islands. The number of ACEs is not fixed – Hawaii has 8 while Bonaire only has 2 – but it means it can be scaled up for higher-end GPUs, console APUs, etc. Finally GCN 1.1 also introduces some new instructions, including a Masked Quad Sum of Absolute Differences (MQSAD) and some FP64 floor/ceiling/truncation vector functions.

Along with these architectural changes, there are a couple of other hardware features that at this time we feel are best lumped under the GCN 1.1 banner when talking about PC GPUs, as GCN 1.1 parts were the first parts to introduce this features and every GCN 1.1 part (at least thus) far has that feature. AMD’s TrueAudio would be a prime example of this, as both Hawaii and Bonaire have integrated TrueAudio hardware, with AMD setting clear expectations that we should also see TrueAudio on future GPUs and future APUs.

AMD’s Crossfire XDMA engine is another feature that is best lumped under the GCN 1.1 banner. We’ll get to the full details of its operation in a bit, but the important part is that it’s a hardware level change (specifically an addition to their display controller functionality) that’s once again present in Hawaii and Bonaire, although only Hawaii is making full use of it at this time.

Finally we’d also roll AMD’s power management changes into the general GCN 1.1 family, again for the basic reasons listed above. AMD’s new Serial VID interface (SIV2), necessary for the large number of power states Hawaii and Bonaire support and the fast switching between them, is something that only shows up starting with GCN 1.1. AMD has implemented power management a bit differently in each product from an end user perspective – Bonaire parts have the states but lack the fine grained throttling controls that Hawaii introduces – but the underlying hardware is identical.

With that in mind, that’s a short but essential summary of what’s new with GCN 1.1. As we noted way back when Bonaire launched as the 7790, the underlying architecture isn’t going through any massive changes, and as such the differences are of primarily of interest to programmers more than end users. But they are distinct differences that will play an important role as AMD gears up to launch HSA next year. Consequently what limited fracturing there is between GCN 1.0 and GCN 1.1 is primarily due to the ancillary features, which unlike the core architectural changes are going to be of importance to end users. The addition of XDMA, TrueAudio, and improved power management (SIV2) are all small features on their own, but they are features that make GCN 1.1 a more capable, more reliable, and more feature-filled design than GCN 1.0.

The AMD Radeon R9 290X Review Hawaii: Tahiti Refined
Comments Locked


View All Comments

  • Pontius - Tuesday, October 29, 2013 - link

    Some good points Jian, I would like to see side by side comparisons as well. However, I've seen some studies that implement the same algorithm in both OpenCL and CUDA and the results are mostly the same if properly implemented. I've been doing GPU computing development in my spare time over the last year and OpenCL does have one advantage over CUDA that is the reason I use it: run-time compilation. If at run-time you are working with various data sets that involve checking many conditionals, you can compile a kernel with the conditionals stripped out and get a good performance increase since GPUs aren't good at conditionals like CPUs are. But in the end, I agree, more apples to apples comparisons are needed.
  • azixtgo - Thursday, October 24, 2013 - link

    the titan is irrelevant. I can't figure why the hell people think a $1000 GPU is even worth mentioning. It's not for sane people to buy and definitely not a genuine effort by nvidia. They saw an opportunity and went for it
  • Bloodcalibur - Thursday, October 24, 2013 - link

    It's $350 more because of it's compute performance ugh. It benchmarks 5-6x more than the 780 on DGEMM. This is why the card is priced a whopping $350 more than their own 780 which is only a few FPS lower on most games and setups. The only people that should've bought a Titan were people who both GAME and do a little bit of computing.

    To compare it to the 290x are what retarded ignorant people are doing. Compare it to the 780 which it does beat out. Now we have to wait for nvidia's response.
  • Cellar Door - Thursday, October 24, 2013 - link

    Read the review before trolling. It's $549
  • azixtgo - Thursday, October 24, 2013 - link

    technically it's a good value. I think. I despise the higher prices as well but who really knows the value of the product. Comparing a GPU to a lot of things (like a ps4 that has a million other components or a complete PC ), maybe not. but comparing this to nvidia... well...
  • Pounds - Thursday, October 24, 2013 - link

    Did the nvidia fanboy get his feelings hurt?
  • superflex - Thursday, October 24, 2013 - link

    Yes, and his wallet got shredded.
    Validation is a bitch.
  • piroroadkill - Thursday, October 24, 2013 - link

    Huh? You can go over to Newegg right now and buy one for $580.
  • Wreckage - Thursday, October 24, 2013 - link

    He did not say "Mantle" 7 times so he might not be from their PR department.

    Either way the 290 is hot, loud, power hungry and nothing new in the performance department. It's cheap but that won't last. Looks like we will have to wait form Maxwell for something truly new.
  • chrnochime - Thursday, October 24, 2013 - link

    You OTOH look like you can't RTFA.

Log in

Don't have an account? Sign up now