The RV770 Story: Documenting ATI's Road to Success
by Anand Lal Shimpi on December 2, 2008 12:00 AM EST- Posted in
- GPUs
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...
Facebook
Twitter
RSS
116 Comments
View All Comments
d0nnie - Wednesday, December 3, 2008 - link
Like many said before, this is truly one of the best readings ive had in a long time. Keep up the good work man!Zar0n - Wednesday, December 3, 2008 - link
Great article, it's excellent to have the back story of the GPU wars.It's kind like the planes/tanks on the history channel :)
5* Anandtech
ViperV990 - Wednesday, December 3, 2008 - link
Awesome article. Did NOT miss the sea of charts. At all.Loved every bit of the tale.
Easily the best PR a company can get.
However I was kinda hoping for a little bit more info on the RV670. What did *that* team go through when making the part that is RV770's direct (or no?) predecessor?
Frallan - Wednesday, December 3, 2008 - link
Im acctually concidering buying a GFX to support the company that allowed this interview. Its easily one of the best Ive read.
hrishi2das - Wednesday, December 3, 2008 - link
A great article... very well written and personal...Can Anand with his great contacts get us a backstory on the Core 2 arch. That would be another great article.
malmal - Wednesday, December 3, 2008 - link
As an NVDA shareholder, having a *huge* amount of unrealised loss in the shares, I bought the 4870 a month ago.A testament to how good the 4870 is at its price point. It just makes sense.
piesquared - Wednesday, December 3, 2008 - link
All preferences aside, I have to say that was probably, hands down, the best article i've read in a long, long time. Talk about capturing your audience.Tremendous insight.
rocky1234 - Wednesday, December 3, 2008 - link
Well I must say this is a good article because it is something different good work on this one. I do have to ask this has AMD been making GPU's for 8 years really..okI know ATI has for sure but they were not part of AMD then if AMD has been then oops I guess I never owned one of their GPU's before they bought ATI.
nowayout99 - Wednesday, December 3, 2008 - link
I have to echo the other comments here, wonderful article! One of the best ever on Anandtech.And thank you, ATI, for giving the community this kind of access to your hard workers.
leonxki - Wednesday, December 3, 2008 - link
What a great story. Really shows how long it can take to make a significant change towards a process that is right for the end user. All those external factors e.g no GDDR5 at the time and G80s prosperity can really suck out an engineers motivation. Good thing those few engineers stuck to the task.The card itself looks so neat and well designed from a bird's eye view. To see how the internals were pateiently designed too is awesome!
The article showed how the RV770 came to life using just words. Time to introduce my pc to life using this card I say :-)
I also join the best-article-I've-read-wagon on this one.