AMD’s Big Play: Heterogeneous Computing

Rob Enderle
Slide Show

What Can Augmented Reality Do?

I attended a number of sessions at AMD’s developer conference (APU13) this week and I’ve found that AMD is becoming a very different company. No longer satisfied with chasing Intel, it is moving to redefine the market with initiatives like Mantle and HSA that will provide the company a unique advantage due to its new cross-platform focus. While AMD joins a growing chorus of firms arguing “heterogeneous computing,” the company is unique in its efforts to actually provide it from a core technology standpoint. Rory Read, AMD’s charismatic CEO, is on strategic track to shift the company from the heavily traditional PC market to a blend that will hedge its investments and market fluctuations by more effectively spreading company revenue base into areas it hasn’t addressed previously.

Let me discuss some of the big thoughts I got from AMD executives this year.

Thinking Must Change

One of AMD’s best executives is Dr. Lisa Su and she made one of the more provocative statements. Basically she said that PC companies needed to stop thinking like PC companies. This really got me thinking. It is consistent with old MBA advice in regard to thinking about the market you are in, not the product you make. For instance, the reason most buggy companies and blacksmiths went out of business when cars came out is because they thought they were in the horse and buggy business, not the personal transportation business. Those who figured this out started making cars or became car mechanics and were more successful.

PC companies aren’t in the IBM PC business. These companies are in the personal computing business and that actually sits as a superset over most consumer electronics, tablets, smartphones, Web services (like Office 365, Mainframe2, and Box), which is pretty much everything that keeps them up at night. This is much like when the buggy companies were scared to death of cars, since buggy companies and car companies were in the same market segment. To survive, those companies needed to evolve.

You basically get two choices: Figure out a way to compete or shutter the plant. Like it or not, companies that don’t have much PC legacy to support, such as Google and Samsung, have entered the segment and you either play or go home.

One other option would be to think ahead to what could displace the current trend that is hurting your business. When Apple saw that smartphones were starting to displace iPods, rather than trying to defend only the iPod, the company jumped ahead and came up with the best iPod smartphone. And it did it before someone else did. Granted, this took two tries (the ROKR sucked), but you can’t argue it wasn’t a success. And then when Samsung started hitting hard on with similar capabilities, Apple came up with a bigger iPod (the iPad) and convinced folks it was new, different and magical. Apple led the race again, becoming the most valuable company of the decade.


Apple was still in the personal computing market, which had largely absorbed consumer electronics even though consumer electronics companies were doing a better job at making the transition. That last point is particularly interesting, because Samsung actually made the flip as well as Apple did when both firms found the market had shifted.

AMD’s message was that the new PC market is about multiple, different-sized devices using different optimized processors and AMD is going to be uniquely expert at addressing it because, unlike the competitors, it is focused on the way the world is now, not the way it was.

Eye Resolution

Mark Papermaster, AMD’s CTO, also addressed some fascinating ideas about where the market is going. If we move the timeline into the future, we are approaching a time when TVs become far smarter and could easily replace windows, allowing us to appear to live anyplace in the universe, both real and imagined. One of the more interesting concepts was called “Eye Resolution,” which is the resolution needed for you to look at a display and find it indistinguishable from a window. The new 4K standard comes very, very close but it is far from the end of this evolution. Combined with overlapping sound fields, the result falls short of the Star Trek Holodeck (though we apparently are closer than you think), but it nearly gives us the ability to move the in-home experience from Hawaii, to Mars, to Barsoom. The only limitation is our imagination and this too will be part of the future personal computer market.

It is interesting to note that while this conference was going on, Apple put on hold its 4K Apple TV (largely due to a lack of ability to get content), but this showcased that Jobs had been planning to pull another iPad-like stunt and get ahead of the next trend. Tim Cook just couldn’t execute Jobs’ plan (in Cook’s defense, no one else seems to be able to pull this off yet, either). Still, I think Jobs did likely call the next technology jump in line with the iPod, iPhone, iPad, reinforcing Papermaster’s vision of our future.

Projects like HAS and Mantle are critical to AMD being able to play in these coming opportunities, because they allow software to more easily move between vendor visions. Changing hardware form-factor solutions is important, so that AMD resides at the heart of the next technology wave instead of being overmatched in the last one.

Wrapping Up: Creating a Wave, Then Riding It

This is at the core of most big corporate success stories from IBM, to Microsoft, to Google and even Facebook. Create a vision, drive it into a technology wave, and then ride it. The other way to achieve this is like Netscape’s old CEO Jim Barksdale said, “Find a parade and get in front of it.” Granted, Netscape kind of lost track of where its parade was going and lost its way. Whether AMD is creating a wave or making use of one that is building can be disputed, but what can’t be disputed is that the company is working its collective ass off to ride that wave. Based on this conference and the latest financial results, AMD is doing an impressive job. It is always a pleasure to see a company understand that the industry isn’t its product—it’s what people do with it.

AMD isn’t focused on building buggies; it is planning on building engines for Holodecks.



Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

null
null

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.