Building Tahiti & The Southern Islands

Now that we’ve had a chance to go over the basis of the Graphics Core Next architecture, let’s talk about the finished products.

Today AMD will be launching Tahiti, the first GPU of the Southern Islands family. Southern Islands will initially be composed of 3 GPUs: Tahiti, Pitcairn, and Cape Verde. Tahiti is the largest and most powerful member of the Southern Islands family, while Pitcairn and Cape Verde get progressively smaller.  AMD has not yet announced the branding or launch dates for Pitcarn and Cape Verde, but it typically takes AMD around 6 months to launch a complete family. As such it’s reasonable to expect that all 3 GPUs will have launched by the end of June although there’s a good likelihood of it happening sooner than that.

All 3 GPUs are based on the GCN architecture, and as family members will have similar features while varying the number of functional units accordingly. Along with the architecture change Southern Islands brings with it a slew of additional features that we’ll get to in the following pages, including Partially Resident Texture (PRT) support, PCIe 3.0, FastHDMI, Direct3D 11.1, and AMD’s fixed-function H.264 encoder, the Video Codec Engine.

But today is all about Tahiti, so let’s get down to business.

As we quickly covered in our introduction, Tahiti is a 4.31B transistor GPU based on the GCN architecture and built on TSMC’s new 28nm High-K process.  Due to TSMC canceling their 32nm process last year AMD has had to wait over 2 years for the next full node rather than taking advantage of the half-node process as they typically do, and as a result the jump from Cayman at 40nm to Tahiti at 28nm is much bigger than with past product launches. Whereas Cayman had 2.64B transistors and a die size of 389mm2, Tahiti has a whopping 63% more transistors than Cayman and yet it’s still smaller, coming in at a slightly more petite 365mm2.


GPU Die Size Comparison

At this point AMD hasn’t provided us with the typical board power values for 7970, but we do know that PowerTune is limited to 250W. In terms of design 7970 is clearly intended to work in similar environments as the 6970, in which case power consumption should be similar to the 6970.

Interestingly enough however we’re hearing that 7970 cards are proving to be very overclockable, which is a good sign for the state of TSMC’s 28nm process, and at the same time a bit distressing. Moore’s Law has continued to hold with respect to transistor density, but the power consumption benefits of using smaller nodes has continued to wane. Having a lot of overclocking headroom means that the 7970 has the potential to be much faster, but it also means that the 7970 (and 28nm GPUs in general) are going to be bottlenecked by power. In which case seeing as how we’re already approaching 300W with single-GPU video cards, the performance gains realized from future fabrication processes would be limited to the ever diminishing returns on power consumption improvements.

Diving deeper into Tahiti, as per the GCN architecture Tahiti’s 2048 SPs are organized into 32 Compute Units. Each of these CUs contains 4 texture units and 4 SIMD units, along with a scalar unit and the appropriate cache and registers. At the 7970’s core clock of 925MHz this puts Tahiti’s theoretical FP32 compute performance at 3.79TFLOPs, while its FP64 performance is ¼ that at 947GFLOPs. As GCN’s FP64 performance can be configured for 1/16, ¼, or ½ its FP32 performance it’s not clear at this time whether the 7970’s ¼ rate was a hardware design decision for Tahiti or a software cap that’s specific to the 7970. However as it’s obvious that Tahiti is destined to end up in a FireStream card we will no doubt find out soon enough.

Meanwhile the frontend/command processor for Tahiti is composed of 2 Asynchronous Command Engines (ACEs) and 2 geometry engines. Just as with Cayman each geometry engine can dispatch 1 triangle per clock, giving Tahiti the same theoretical 2 triangle/clock rate as Cayman. As we’ll see however, in practice Tahiti will be much faster than Cayman here due to efficiency improvements.

Looking beyond the frontend and shader cores, we’ve seen a very interesting reorganization of the rest of the GPU as opposed to Cayman. Keeping in mind that AMD’s diagrams are logical diagrams rather than physical diagrams, the fact that the ROPs on Tahiti are not located near the L2 cache and memory controllers in the diagram is not an error. The ROPs have in fact been partially decoupled from the L2 cache and memory controllers, which is also why there are 8 ROP partitions but only 6 memory controllers. Traditionally the ROPs, L2 cache, and memory controllers have all been tightly integrated as ROP operations are extremely bandwidth intensive, making this a very unusual design for AMD to use.

As it turns out, there’s a very good reason that AMD went this route. ROP operations are extremely bandwidth intensive, so much so that even when pairing up ROPs with memory controllers, the ROPs are often still starved of memory bandwidth. With Cayman AMD was not able to reach their peak theoretical ROP throughput even in synthetic tests, never mind in real-world usage. With Tahiti AMD would need to improve their ROP throughput one way or another to keep pace with future games, but because of the low efficiency of their existing ROPs they didn’t need to add any more ROP hardware, they merely needed to improve the efficiency of what they already had.

The solution to that was rather counter-intuitive: decouple the ROPs from the memory controllers. By servicing the ROPs through a crossbar AMD can hold the number of ROPs constant at 32 while increasing the width of the memory bus by 50%. The end result is that the same number of ROPs perform better by having access to the additional bandwidth they need.

The big question right now, and one we don’t have an answer to, is what were the tradeoffs for decoupling the ROPs? Clearly the crossbar design has improved ROP performance through the amount of memory bandwidth they can access, but did it impact anything else? The most obvious tradeoff here would be for potentially higher latency, but there may be other aspects that we haven’t realized yet.

On that note, let’s discuss the memory controllers quickly. Tahiti’s memory controllers aren’t significantly different from Cayman’s but there are more of them, 50% more in fact, forming a 384bit memory bus. AMD has long shied away from non-power of 2 memory busses, and indeed the last time they even had a memory bus bigger than 256bits was with the ill-fated 2900XT, but at this point in time AMD has already nearly reached the practical limits of GDDR5. AMD’s ROPs needed more memory bandwidth, but even more than that AMD needed more memory bandwidth to ensure Tahiti had competitive compute performance, and as such they had little choice but to widen their memory bus to 384bits wide by adding another 2 memory controllers.

It’s worth noting though that the addition of 2 more memory controllers also improves AMD’s cache situation. With 128KB of L2 cache being tied to each memory controller, the additional controllers gave AMD 768KB of L2 cache, rather than the 512KB that a 256bit memory bus would be paired with.

A Quick Refresher, Cont Tahiti: The First Direct3D 11.1 GPU
Comments Locked

292 Comments

View All Comments

  • CeriseCogburn - Sunday, March 11, 2012 - link

    We'll have to see if amd "magically changes that number and informs Anand it was wrong like they did concerning their failed recent cpu.... LOL
    That's a whole YEAR of lying to everyone trying to make their cpu look better than it's actual fail, and Anand shamefully chose to announce the number change "with no explanation given by amd"... -
    That's why you should be cautious - we might find out the transistor count is really 33% different a year from now.
  • piroroadkill - Thursday, December 22, 2011 - link

    Only disappointing if you:

    a) ignored the entire review
    b) looked at only the chart for noise
    c) have brain damage
  • Finally - Thursday, December 22, 2011 - link

    In Eyefinity setups the new generation shines: http://tinyurl.com/bu3wb5c
  • wicko - Thursday, December 22, 2011 - link

    I think the price is disappointing. Everything else is nice though.
  • CeriseCogburn - Sunday, March 11, 2012 - link

    The drivers suck
  • RussianSensation - Thursday, December 22, 2011 - link

    Not necessarily. The other possibility is that being 37% better on average at 1080P (from this Review) over HD6970 for $320 more than an HD6950 2GB that can unlock into a 6970 just isn't impressive enough. That should be d).
  • piroroadkill - Friday, December 23, 2011 - link

    Well, I of course have a 6950 2GB that unlocked, so as far as I'm concerned, that has been THE choice since the launch of the 6950, and still is today.

    But you have to ignore cost at launch, it's always high.
  • CeriseCogburn - Thursday, March 8, 2012 - link

    I agree RS, as these amd people are constantly screaming price percentage increase vs performance increase... yet suddenly applying the exact combo they use as a weapon against Nvidia to themselves is forbidden, frowned upon, discounted, and called unfair....
    Worse yet, according to the same its' all Nvidia's fault now - that amd is overpriced through the roof...LOL - I have to laugh.
    Also, the image quality page in the review was so biased toward amd that I thought I was going to puke.
    Amd is geven credit for a "perfect algorythm" that this very website has often and for quite some time declared makes absolutely no real world difference in games - and in fact, this very reviewer admitted the 1+ year long amd failure in this area as soon as they released "the fix" - yet argued everyone else was wrong for the prior year.
    The same thing appears here.
    Today we find out the GTX580 nvidia card has much superior anti-shimmering than all prior amd cards, and that finally, the 7000 high end driver has addressed the terrible amd shimmering....
    Worse yet, the decrepit amd low quality impaired screens are allowed in every bench, with the 10% amd performance cheat this very site outlined them merely stated we hope Nvidia doesn't so this too - then allowed it, since that year plus ago...
    In the case of all the above, I certainly hope the high end 797x cards aren't CHEATING LIKE HECK still.
    For cripe sakes, get the AA stuff going, stop the 10% IQ cheating, and get our bullet physics or pay for PhysX, and stabilize the drivers .... I am sick of seeing praise for cheating and failures - if they are (amd) so great let's GET IT UP TO EQUIVALENCY !
    Wow I'm so mad I don't have a 7970 as supply is short and I want to believe in amd for once... FOR THE LOVE OF GOD DID THEY GET IT RIGHT THIS TIME ?!!?
  • slayernine - Thursday, December 22, 2011 - link

    Holy fan boys batman!

    This comment thread reeks of nvidia fans green with jealousy
  • Hauk - Thursday, December 22, 2011 - link

    LOL, Wreckage first!

    Love him or hate him, he's got style..

Log in

Don't have an account? Sign up now