The Call

My love/hate relationship with AMD PR continued last year. But lately, it’s been far less hate. Let’s rewind back to the Summer of 2009. I’d been waiting for AMD to call for weeks.

We all knew that the RV870 was going to launch sometime before the end of the year, and we’re normally briefed on new GPUs around a month or so before we get hardware. The rumors said that the launch had been pushed back, but just like clockwork I got a call in June or July of last year. It was my old friend, Chris Hook of AMD PR.

This time he wanted me to come to a press event on a carrier off the coast of California. Sigh.

It’s not that I have anything against carriers. It’s just that all I cared about at that time was the long awaited successor to the RV770. The RV770 was the GPU that unequivocally restored my faith in ATI graphics, an impact shared by others last June. But that’s not how the game is played I’m afraid. AMD promises its management and its partners that they can fill a room (or carrier) up with important press. We get promised access to engineers, useful information and free drinks.


The USS Hornet. GPUs are in there.

I’m not easily swayed by free drinks, but Chris Hook knows me well enough by now to know what I’d appreciate even more.

The Dinner - September 2009

I had to leave dinner earlier than I wanted to. ASUS’ Chairman Jonney Shih was in town and only had one opportunity to meet me before I left Oakland. Whenever either of us happens to be in the same town, we always make our best effort to meet - and I wasn’t going to let him down. In the same vein that Steve Jobs is successful because he is a product guy at heart, running a company best known for its products. Jonney Shih is an engineer at heart, and he runs a company who has always been known for their excellence in engineering. This wasn’t just another meeting with an executive, this was a meeting with someone who has a passion for the same things I do. His focus isn’t on making money, it’s on engineering. It’s a rare treat.

My ride was waiting outside. I closed the lid on my laptop, making sure to save the 13 pages of notes I just took while at dinner. And I shook this man’s hand:

Before I left he asked me to do one thing. He said “Try not to make the story about me. There are tons of hardworking engineers that really made this chip happen”. Like Jonney, Carrell Killebrew has his own combination of traits that make him completely unique in this industry. All of the greats are like that. They’ve all got their own history that brought them to the companies that they work for today, and they have their own sets of personality traits that when combined make them so unique. For Carrell Killebrew it's a mixture of intelligence, pragmatism, passion and humility that's very rare to see. He's also a genuinely good guy. One of his tenets is that you should always expect the best from others. If you expect any less than the best, that’s all you’ll ever get from them. It’s a positive take on people, one that surprisingly enough only burned Carrell once. Perhaps he’s more fortunate than most.

Mr. Killebrew didn’t make the RV870, but he was beyond instrumental in making sure it was a success. What follows is a small portion of the story of the RV870, the GPU behind the Radeon HD 5800 series. I call it a small portion of the story because despite this article using more than eight thousand words to tell it, the actual story took place over years and in the minds and work of hundreds of engineers. This GPU, like all others (even Fermi) is the lifework of some of the best engineers in the industry. They are the heroes of our industry, and I hope I can do their story justice.

As is usually the case with these GPU backstories, to understand why things unfolded the way they did we have to look back a few years. Introducing a brand new GPU can take 2 - 4 years from start to finish. Thus to understand the origins of the Radeon HD 5800 series (RV870) we have to look back to 2005.

Sidebar on Naming

AMD PR really doesn’t like it when I use the name RV870. With this last generation of GPUs, AMD wanted to move away from its traditional naming. According to AMD, there is no GPU called the RV870, despite the fact that Carrell Killebrew, Eric Demers and numerous others referred to it as such over the past couple of years. As with most drastic changes, it usually takes a while for these things to sink in. I’ve also heard reference to an RV870 jar - think of it as a swear jar but for each time someone calls Cypress an RV870.

Why the change? Well, giving each member of a GPU family a name helps confuse the competition. It’s easy to know that RV870 is the successor to the RV770. It’s harder to tell exactly what a Cypress is.

AMD PR would rather me refer to RV870 and the subject of today’s story as Cypress. The chart below shows AMD’s full listing of codenames for the 40nm DX11 GPU lineup:

GPU Codename
ATI Radeon HD 5900 Series Hemlock
ATI Radeon HD 5800 Series Cypress
ATI Radeon HD 5700 Series Juniper
ATI Radeon HD 5600/5500 Series Redwood
ATI Radeon HD 5400 Series Cedar

 

Given that we still haven’t purged the RVxxx naming from our vocabulary, I’m going to stick with RV870 for this story. But for those of you who have embraced the new nomenclature - RV870 = Cypress and at points I will use the two names interchangeably. The entire chip stack is called Evergreen. The replacement stack is called the Northern Islands.

The Best Way to Lose a Fight - How R5xx Changed ATI
POST A COMMENT

132 Comments

View All Comments

  • devene - Sunday, February 14, 2010 - link

    Just like many others, I've been a long time reader and I just couldn't carry on without leaving a comment:

    This has been an article, just like the RV770 one. It may not reveal many facts but is tremendously insightful and inspiring. Thank you for bringing this deeply hidden information out to the public and to the "fans". Please do everything in your power to continue this trend.

    Once again, thank you Anand,
    devene
    Reply
  • medi01 - Sunday, February 14, 2010 - link

    Germans say "lange Rede kurzer Sinn". So many pointless sentences that do not tell anything even remotely interesting. Reply
  • TGressus - Sunday, February 14, 2010 - link

    Even the home team could not be sold on Eyefinity... Reply
  • William Gaatjes - Sunday, February 14, 2010 - link

    Fantastic article.

    "
    First, it massively increased the confidence level of the engineering team. There’s this whole human nature aspect to everything in life, it comes with being human. Lose confidence and execution sucks, but if you are working towards a realistic set of goals then morale and confidence are both high. The side effect is that a passionate engineer will also work to try and beat those goals.
    "

    Finally, someone accepting and using human nature.
    And see it works out...

    The fun part is that a requested functionality that is desired but can not make it within the expected timeframe, can still be worked on and can be ready for the next "bulge" in the market. This way you relieve your engineers form stress, you have the time to sort errors and bugs out, you have time to solve unforseen consequences that always happen( people can get sick, a bug in software, machines breaking down) and you have a feature for the market department to market to the consumer for the next iteration of the product. This way you can use the free market to build an in the end perfect device. It is all about balance. If you have to invest to much energy in situation a, you will have less energy for situation b in a certain timeframe. We are bound by laws of nature meaning there is no "perpetuum mobile" in this universe. Nothing comes for free...
    Reply
  • aegisofrime - Sunday, February 14, 2010 - link

    Anand, you have taken an article that is really technical in nature, and turned it into something entertaining to read and yet informative for non-engineer types. My hats off to you. This is really the right balance of information and readability. If only all the Scientific Papers I have to read were written like this! Reply
  • dukeariochofchaos - Sunday, February 14, 2010 - link

    i wonder if you will give fermi the same drama queen touch?

    i hope so.

    Reply
  • Jamahl - Sunday, February 14, 2010 - link

    I don't think anyone wants to read nvidia's marketing department tell us how awesome PhysX and CUDA is again tbh. Reply
  • TGressus - Sunday, February 14, 2010 - link

    I suspect Fermi will be able to stand on it's technological innovation.
    Reply
  • RJohnson - Sunday, February 14, 2010 - link

    ...and it's exorbitant price/die size will exclude mere mortals from owning one. Reply
  • Spoelie - Sunday, February 14, 2010 - link

    That depends entirely on the openness of NVIDIA on the subject, historically not one of their strong points.

    In fact ATi's take on NVIDIA's design process has been more informative than what has come out of NVIDIA itself.

    But here's to hoping..
    Reply

Log in

Don't have an account? Sign up now