Just One Small Problem: We Need a New Memory Technology

The R600 GPU had an incredibly wide 512-bit memory interface, the problem with such a large interface is that it artificially makes your die bigger as you’ve got to route those interface pads to the memory devices on the board. For RV770 to have the die size ATI wanted, it needed to have a 256-bit memory interface, but using (at the time) current memory technology that wouldn’t give the GPU enough memory bandwidth to hit the performance targets ATI wanted.

When the options were either make the chip too big or make the performance too low, ATI looked elsewhere: let’s use a new memory technology. Again, put yourself in ATI’s shoes, the time was 2005 and ATI had just decided to completely throw away the past few years of how-to-win-the-GPU-race and on top of that, even if the strategy were to succeed it would depend on a memory technology that hadn't even been prototyped yet.

The spec wasn’t finalized for GDDR5 at the time, there were no test devices, no interface design, nothing. Just an idea that at some point, there would be memory that could offer twice the bandwidth per pin of GDDR3, which would give ATI the bandwidth of a 512-bit bus, but with a physical 256-bit bus. It’s exactly what ATI needed, so it’s exactly what ATI decided to go with.

Unfortunately whether or not GDDR5 shipped by the summer of 2008 wasn’t all up to ATI, the memory manufacturers themselves had a lot of work to do. ATI committed a lot of resources both monetarily and engineering to working with its memory partners to make sure that not only was the spec ready, but that memory was ready, performing well and available by the summer of 2008. Note that the RV770 was going to be the only GPU that would use GDDR5, meaning that it was ATI and ATI alone driving the accelerated roadmap for this memory technology. It’s akin to you trying to single handedly bring 100Mbps internet to your city; it’ll happen eventually, but if you want it done on your time table you’re going to have to pickup a shovel and start burying a lot of your own cable.

ATI did much of the heavy lifting with the move to GDDR5, and it was risky because even if RV770 worked out perfectly but the memory wasn’t ready in time the GPU would get delayed. RV770 was married to GDDR5 memory, there was no other option, if in three years GDDR5 didn’t ship or had problems, then ATI would not only have no high end GPU, but it would have no performance GPU to sell into the market.

If GDDR5 did work out, then it meant that RV770 could succeed and that it would be another thing that NVIDIA didn’t have at launch. That is, of course, assuming that ATI’s smaller-die strategy would actually work...

The Power Paradigm Dave Baumann Saves the Radeon HD 4850
Comments Locked

116 Comments

View All Comments

  • VaultDweller - Wednesday, December 3, 2008 - link

    I wish you had gotten greedy! I want to know about RV870, and about nVidia's first DirectX 11 part too.

    I had been thinking about building a new gaming rig in Q1 2009, but presently it looks like I'd be spending too much for too little improvement over my current box. I'm hoping that changes by late summer. :)
  • murray13 - Wednesday, December 3, 2008 - link

    Kudos to you Anand for a G R E A T article!!!

    And to AMD/ATI for making this possible!

    Now if only AMD on the cpu side could do something similar...
  • jzodda - Wednesday, December 3, 2008 - link

    This is one of the best articles I have read here and there have been so many over the years. In the more then 10+ years I have been coming here I have always enjoyed this site and Anand continues to produce great content. Here's to another 10 years!
  • pcfxer - Wednesday, December 3, 2008 - link

    Not only does anand do engineering, he writes like mad!

    "Passion has a funny way of being a person’s strongest ally."
  • prophet001 - Wednesday, December 3, 2008 - link

    great article. thank you for writing this :)
  • rqle - Wednesday, December 3, 2008 - link

    awesome article, very good read, thanks
  • wingless - Wednesday, December 3, 2008 - link

    Anand, you and all the hard working people at this website have just outdone yourselves. You raised the bar yet again.

    Your readers are probably as amazed as you are that AMD/ATI came out with such personal and intimate information as to what goes on behind closed doors. Your conclusion is on point as well. Without competition, we know these other companies will run wild with their prices. Unfortunately for us, the fate of competition in both the GPU and CPU market falls on AMD which needs a little financial lovin' right about now.

    My strongest desire is that the CPU team over at AMD pulls out all the stops with their next CPU to Bulldoze the competition (or just their prices). We need to make $1000 CPUs a thing of the past. Maybe your site and others can put pressure on AMD to bump up their CPU roadmap about a year so we see Bulldozer in 2010.
  • Zoomer - Wednesday, December 3, 2008 - link

    Yup. I have always voted Red with my wallet though! The only two green cards I had were the Riva TNT and the 7600GT. Got rid of that after a couple of months though: it was noisy and the drivers sucked.

    Radeon DDR 32MB, Radeon 8500, Radeon 9500Pro (Unlocked ;)), Radeon X800XT, Radeon x1950 Pro, and finally the Radeon 4850!
  • wingless - Wednesday, December 3, 2008 - link

    I know that is an unrealistic hope. I understand AMD has made it's road map decisions very carefully given their current situation. I can dream though....
  • Bull Dog - Wednesday, December 3, 2008 - link

    Awesome article Anand. I thoroughly enjoyed reading it and I hope for more articles like this in the feature. Many thanks to AMD for letting this happen.

Log in

Don't have an account? Sign up now