Some Thoughts on the Impact of AMD's Layoffs on the Future of AMD GPUs
by Anand Lal Shimpi on November 4, 2011 8:18 PM ESTHere's something I'm concerned about. AMD's Carrell Killebrew, part of yesterday's announced layoffs, was a Director of Product Planning for AMD's GPU division. His job, at least as he explained it to me so many times in the past, was to figure out what the next 3 - 5 years of AMD GPUs were going to look like. He's still technically with AMD today, although that will change in the not too distant future.
Carrell is a dedicated guy, he works hard and generally seems to know the right move for AMD in the GPU space. His track record as of late is a good one. The verdict isn't out on AMD's 28nm GPUs, but the last three years of AMD GPU releases have been excellent. They've been competitive and well executed.
When reducing workforce to cut costs, you don't go after your product planners - unless their vision and your vision don't line up. We all know what Carrell wanted for the future of AMD GPUs (as I wrote before, he wanted to deliver a first generation "Holodeck" by 2016), but what does AMD's new CEO want that conflicts with this goal?
Carrell's vision saw the continued growth of the high-end GPU. On November 9th we're supposed to hear more about Rory Read's new strategy for AMD. I am concerned that it may conflict with Carrell's vision. Maybe I'm reading too much into all of this. What do you all think?
Facebook
Twitter
RSS
48 Comments
View All Comments
fic2 - Friday, November 4, 2011 - link
Seems like he should have been moved over to the CPU side. Or put above both CPU and GPU.deputc26 - Friday, November 4, 2011 - link
With the GPU side of things doing much better (at least from a design standpoint) than the CPU side it is disturbing that AMD would let this guy go. Hopefully this won't dull the wonderful competition that us consumers have enjoyed in the GPU industry in recent years.RussianSensation - Friday, November 4, 2011 - link
This likely means less dedicated engineers working on high-end parts. When you have limited resources, you have to cut costs to fund your new strategy. Unfortunately, I either see AMD's Bulldozer Next or AMD's Graphics R&D budget being redirected towards low-power and more mainstream devices.For instance, they can manufacture more versions of the Brazos CPU to better align with tablet/smartphone market needs. They can focus on creating low-powered server CPUs by massaging Bobcat.
The creation of all-in-one chips (Something that Steve Jobs described in his biography as what Intel didn't want to do) is likely the direction AMD wants to follow. So the focus might be shifting away from high-end GPUs (although I hope not).
It seems even Micron's CEO, Appleton, just announced that they too are moving towards manufacture of chips for mobile devices:
http://www.businessweek.com/news/2011-11-03/micron...
A lot of companies foresee that the next 5-year growth phase will come in smartphone/tablet space. Of course, discrete GPUs will still have a market, but it won't grow anywhere as fast.
AMD's GPU division also doesn't appear to be making that much $, given Q3 earnings report by AMD. AMD keeps competing on price for GPUs, which isn't helping their average selling price/profitability. For example, if HD7970 launches way ahead of Kepler, then AMD should price such as chip at $499, not $369. Their current GPU strategy is good for market share and mobile devices (i.e., manufacture of very efficient GPUs), but it's not helping them sell their GPUs for more $ since they aren't faster than NVs, and hence can't command a premium.
I am sure Read is not happy about selling GPUs at low prices just to keep market share or please us gamers. It will suck for gamers if high-end GPUs are given less priority, but I feel that AMD must raise prices for their GPUs. I mean sure, gamers love it when you can buy an HD6950 2GB for $230 and unlock it into an HD6970, but that's not favourable for business profitability.
I have a feeling Dirk was ousted because he was strongly against the new direction the board wants to pursue, because it likely de-prioritizes the enthusiast server, CPU and GPU markets in favour of low-priced, higher volume designs for low-power devices, and emerging markets.
mckirkus - Saturday, November 5, 2011 - link
The next gen consoles will probably define what GPUs look like going forward. As integrated graphics improve and laptops take over, the market for discreet graphics shrinks in the PC space.I'm pretty sure the next gen consoles will look a lot like what AMD is going to put into laptops next year. Consoles that are huge power hungry beasts may produce better graphics but they're all going to look so much better than XBox 360 and PS3 that it's probably not a good strategy for AMD.
Nvidia is probably going to focus on mobile and turning their chips into CPUs (project Denver) so they're not stuck selling video cards in a dying market.
formulav8 - Saturday, November 5, 2011 - link
The Higher-end gpu market will be more likely to shrink instead of expand. Its really not a very profitable business as you can see with AMD's GPU division not being very strong in making money the past 1/4 even though they sold lots of cards.Were just not big enough anymore, and never will be like the market was in the late 90's through the mid-2000's. Thus, when the next gen consoles come around, the masses (even including people from us) will surely abandon pc gaming even more. Which will make the pc gaming world even less of an incentive.In the long run consoles cost less, is much simplier to program for, and is much better to deal with when you have to try and keep up with gpu/cpu tech.
Don't be surprised if Amd TRULY starts to Only care about the Lower End/Mainstream CPU And GPU markets when it comes to non-server customers. Even the nVidia fannys will have a cow about the cost of future video cards when AMD finally announces what their new strategy will be in the next year or 2. (I'm obviouslly not 100% sure what their stategy will be. It just seems that the writing is on the wall).
Its all about money, period. You can be sure as well that the board is already VERY upset selling their low-powered IP to Qualcomm.Especially as cheap as they sold it. Something like $12 Million iirc?
Belard - Monday, November 7, 2011 - link
I agree. Especially with next gen consoles about A year or so out . The aaa titles are mostly console... I don't have one, but I get the feeling play a game on a 48-60 inch tv may pull u into the game more than a 20-24" monitor.No games = no need for high end gpus.
0ldman79 - Monday, November 7, 2011 - link
I've got my PC connected to my 46 inch Samsung. Runs quite nicely and looks slightly better than my PS3 with only a Radeon 5670.For one, the mid range cards are quite powerful, so the high end card sales are weaker, two, the economy sucks, people are doing without or doing with less.
shin0bi272 - Tuesday, November 8, 2011 - link
by that logic you should just buy a long hdmi cable and a wireless keyboard and mouse and play your pc games on your home tv and you'd save the pc gaming market.Zink - Friday, November 4, 2011 - link
I can't see how letting the high end go would be a good thing. AMD's APUs look good because they have the GPU architecture that their competition doesn't have. With CPU architecture so weak it would make sense to maintain a lead in high end graphics and HPC tech while the CPU guys try to get everything back together. On the other hand I don't know how many high end GPUs AMD sells, maybe being competitive with phone/tablet GPUs and making efficient APUs is just as profitable.sjael - Friday, November 4, 2011 - link
If I read this article right, then there's a chance that AMD will give up on high-end GPUs as well?I'm no industry expert or anything, but as a hardware enthusiast (and thus de facto tech support guy to everyone I know) I've seen most of AMD's success with GPUs come from delivering a high-end part at a great price.
I guess the traditional wisdom is that high-end parts are for advertising, and mid-range is for making money, but I actually don't know *anyone* who has less than a 6870 in their desktop PC.
Of course I don't know anything about how business is going at AMD, just strikes me as absurd that they would can one of their handful of remaining (visibly) successful product brackets.