The RV770 Story: Documenting ATI's Road to Success

The Last Hiccup: Boards Went on Sale Too Soon

The RV770 products were finished in May of 2008, production started by June. Even up until the day that the embargo lifted there were some within ATI who felt they had made a mistake with the smaller-die strategy, but they were going to find out how right the strategy was, even sooner than expected.

The last bump in the road to the Radeon HD 4800 came just a week before launch. We had literally just gotten our first Radeon HD 4850 cards when Chris Hook called and told me that some 4850s had started selling in Europe.

In order to salvage the launch ATI was proposing the following: we could talk about Radeon HD 4850 performance, but we couldn’t talk about the 4870 or the RV770 architecture.

Within 30 hours we had our first preview up and made it already clear that ATI was on to something. The GeForce 9800 GTX got an abrupt price drop to remain competitive and even then it wasn’t enough, the Radeon HD 4850 was the card to get at $199.

The last hiccup in ATI’s launch ended up not being bad at all, ATI got some extra PR, drummed up some added excitement and in the end did justice to a product that deserved it.

Recon from Taiwan

One thing I wondered was how well ATI knew NVIDIA’s plans and vice versa, so I asked the obvious: where do you guys get your information from? The answer was pretty much as expected: Taiwan. All of the board makers know one another and are generally open with sharing information, once information hits Taiwan it’s up for grabs. Then there’s a bit of guesswork that’s done.

ATI planned to put its best foot forward, looking at the roadmaps it seemed like NVIDIA wasn’t going to do much in the G92 space in the time period that ATI would launch RV770. NVIDIA had its sights set on another G80-esque launch with GT200, it would introduce this honkin new chip, price it out of the reach of most and not worry about the peasants until sometime in 2010. The existing product line would be relied on to keep the masses at bay.

ATI was lucky that NVIDIA only had GT200 for the end of 2008 and that NVIDIA’s GT200 performance wasn’t exactly where it needed to be, because it created an opportunity that ATI has only had a couple of times in the past decade.

With the Radeon HD 4850 the initial goal was to make a product that was certainly better than the 8800 GT. It was never a goal for the 4850 to be competitive with the 9800 GTX, after all that was a $300 part and this would sell for $200.

The Radeon HD 4870 was targeted to be faster than the 9800 GTX, which again would make a lot of sense since this was ATI’s $300 part and the GTX was NVIDIA’s. What ATI didn’t expect was for the 4870 to do so well against the GeForce GTX 260. When NVIDIA finally launched the GeForce GTX 280/260 ATI looked at the results and let out a collective “wait a minute”. It worked out perfectly, not only did ATI hit the competitive points it wanted to but thanks to GT200 performance being lower than ATI expected and the RV770 doing better than expected, ATI now had a $300 card that was competitive with NVIDIA’s brand new $400 GTX 260.

For ATI, RV770 was the cake that came out unexpectedly well. Everyone could smell it, and they knew it would be good, but no one expected it to be perfect. NVIDIA responded extremely quickly and honestly no other company would be able to handle such competition so well, but that doesn’t change what ATI was able to accomplish.

These days no one questions Carrell’s thinking about RV770 any longer, everyone agrees that he was right about the strategy. My question is, how long until ATI has to re-evaluate its GPU strategy once more? The first time was in 2001 with R300, again in 2005 with the RV770, which would point to next year as to when some tough decisions may be made again - the results of which we wouldn’t see until 2012/2013.

The next question is how will NVIDIA respond to ATI’s strategy? Jen Hsun runs a very tight ship over there and does not take kindly to losing, especially not like this. NVIDIA continues to have very strong engineering talent and over the next couple of years we’ll see how RV770 has impacted NVIDIA’s development. It’s possible that NVIDIA too realized that the smaller-die strategy made more sense without having been impacted by RV770, perhaps NVIDIA will stick with making huge GPUs, or maybe a third option exists that isn’t as obvious.