Waiting for AMD’s Zen

Rob Enderle

Zen is the code name for AMD’s coming processor for PCs and servers. What makes it particularly interesting is that it was redesigned from scratch to take into account current workloads, with an architecture that should allow it to compete effectively with two-socket solutions using a single socket. This should provide a number of advantages, including better thermals and performance, since more will be done in silicon. The timing is also fascinating, because I see a similarity to the last time AMD rose up, when Craig Barrett was running Intel and the firm lost track of its core markets.

Zen, which had a coming out party of sorts during IDF, will show up on the desktop in limited numbers before year end, and on servers in a bit under a year. Let’s talk about why the market seems suddenly very excited about a new x86 part from AMD.

Intel/AMD

Both Intel and AMD are still recovering from what is largely a lost decade during which both firms tried to chase the mobile market. AMD unfortunately started its chase after it had sold off ATI’s mobile assets, making the move foolish. Intel didn’t seem to get that moving from a high-margin part to a low-margin part was literally suicidal and there was no viable path to success. Even if it had been successful in taking out Qualcomm, the drop in margins would have crippled the company. Apparently, someone had skipped economics.

Both firms now find themselves with new CEOs -- even though you could argue that the problems largely came from their respective boards -- and suddenly, with very different strategies as well.

Intel is currently pivoting hard from PCs and servers, based on its developer conference, to drones, IoT and autonomous cars. All of these areas are potentially very lucrative and Intel does have the resources to go after them aggressively, but I wonder if Intel has learned its lesson that it doesn’t matter how much you spend, it matters if you spend enough to get the job done. This is where most firms fail in going after new markets: They budget based on what they think they can afford, not on what it will take to be successful, so these efforts are almost globally under resourced.

AMD, on the other hand, has fully buckled down and focused back on its traditional server and PC segments; and using a handpicked team of folks who came out of a variety of firms, including Intel, IBM, and Qualcomm, it has created Zen.

Zen

Zen isn’t available for third-party testing yet, but I have spoken to some of those who have been testing it in AMD and they are beside themselves with excitement. As I noted, this part was not one adapted from one of the earlier offerings, but it represents a clean slate, ground-up design that prioritizes current manufacturing methods, emerging loads like virtual reality, and 4K editing, and high-performance computing to create a uniquely tuned part that should be attractive to systems builders, individuals and firms that build their own hardware.


I’m a big fan of clean slate efforts, even if they are just an exercise, because often things get built into a design for the wrong reasons or as a result of mistakes. Then, they are left there because nobody challenges them. You can lose some efficiencies created over time by starting this way, but if you don’t do this regularly, you likely end up carrying an increasing number of inefficiencies and false assumptions as time reveals what the market actually wants. Zen is effectively a reset for a market that was largely built on a distributed world but is now shifting hard to a mixed model where much of the actual work is done centrally.

Wrapping Up: Unusual AMD Excitement

I should say this is unusual excitement for any x86 part, because with all of the focus on mobile, there just hasn’t been that much interest in x86 technology for some time, even though it is still central to a market with ASPs (average selling prices) that are multiples of their mobile counterparts (which are often under even heavier price and volume pressures). In the end, Zen brings excitement and interest to a segment that is in desperate need of it and further showcases the need to rethink a product in order to better zero in, not on where it would be going, but where it should be going. And that last makes Zen a very interesting and special offering indeed.

It is actually kind of interesting how many people seem to be very excited about this part. Like many, until I see my first evaluation system, I’ll be waiting for Zen.

Rob Enderle is President and Principal Analyst of the Enderle Group, a forward-looking emerging technology advisory firm.  With over 30 years’ experience in emerging technologies, he has provided regional and global companies with guidance in how to better target customer needs; create new business opportunities; anticipate technology changes; select vendors and products; and present their products in the best possible light. Rob covers the technology industry broadly. Before founding the Enderle Group, Rob was the Senior Research Fellow for Forrester Research and the Giga Information Group, and held senior positions at IBM and ROLM. Follow Rob on Twitter @enderle, on Facebook and on Google+



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.