I’m at the AMD event in San Francisco this week and find it kind of ironic that I’m only a few blocks from where Intel used to hold IDF, the conference that assured its dominance and that its terminated CEO has killed. Currently, in the PC and server space, while Intel still leads in market share, its lead in mindshare has dropped dramatically over the last several years due to a combination of mistakes, not the least of which are efforts to maximize quarterly results by sacrificing strategic initiatives like IDF. On the other hand, and competing with Intel, are AMD and Qualcomm, both in very different situations. Qualcomm is defending against Intel, trying to literally steal its market (and it isn’t going well outside of Apple) and AMD is moving on Intel ever more aggressively.
Intel’s financial performance remains strong despite its serous security issues, the accusations of intellectual property theft, and its inability to hire either a CEO or CMO in a timely manner. The firm is badly broken, but so dominant that it is holding, until now. AMD, at this event, which is being held on election day here in the U.S. (a bit poetic because it is making a strong case for voting with your dollars against Intel’s painful dominance), is providing a counterpoint in that it isn’t focused on CEO compensation, covering up security problems, or abusing customers, and instead is talking about its focus on customers and performance.
The AMD Alternative
As I’m writing this, Mark Papermaster, CTO for AMD, is talking about how AMD’s processor architecture is vastly different from Intel’s. It is far more resistant to security problems, using a concept called “chiplets” and a new 7nm process, which increases performance and sharply reduces cost (which AMD does pass on to customers), and demonstrating a multi-year history of meeting expectations.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=i
Most of this is focused on Zen2, the next-generation processor, which should not only have the expected jumps in performance but excel in flexibility, so it can provide more value for the dollar and the strongest alternative to Intel’s offerings so far.
What I think is fascinating is that at prior events, Papermaster has been reserved because he was still dealing with some bad mistakes made by his predecessors. No longer. Now he feels confident in AMD’s ability to execute and he is effectively throwing down the gauntlet to Intel. Suddenly, this got a lot more interesting.
Papermaster was followed by David Wang, SVP, Radeon Technologies Group. On graphics, AMD, which acquired ATI around a decade ago, has always outperformed Intel. But here it faces a far less troubled competitor in NVIDIA. Unlike Intel, NVIDIA is hitting on all cylinders as well, but the market is more balanced. NVIDIA’s dominance isn’t as absolute as Intel’s. Wang is newer and more experienced with AMD/ATI than Papermaster in that he used to work for AMD, left, and then returned a year ago. His predecessor left AMD involuntarily and now works at Intel in what will likely be another failed attempt by that firm to close the technology gap.
The reason I expect that graphics effort to fail is because Intel has a habit of stabbing external hires in the back, and catching AMD and NVIDIA, which are racing each other and moving (in graphics) at a far faster pace (historically) than the CPU market, from behind is likely impossible. They are showcasing a new concept, Hardware Based Virtualization, which promises far greater security and, possibly, far higher potential performance than the more common Software Based Virtualization. This appears to be mostly focused on AI and deep learning and security, because the concern of biased and potentially hostile AIs resulting from breaches remains extremely high. (We apparently don’t want to experience Terminators in real life, go figure?)
Another competitive advantage is AMD’s near rabid focus on open source, which is being presented as one of the key reasons for the design wins being highlighted at the event. Baidu was on stage in support of this effort, but they showcased several other key players that are apparently also now embracing AMD. The new graphics parts are performing at 3x to 9x the performance of the prior generation (recognize that performance jumps tend to be closer to 10 percent, or .1x historically, so this kind of a performance jump is very unusual). I should point out that NVIDIA is also showcasing this kind of jump, further showcasing that it is virtually impossible for Intel to catch either firm from behind.
They brought “highwai” up on stage. Highwai is a simulation firm specializing in deep and machine learning. The example is on using simulation to train autonomous cars. The AMD-based system is being used by firms like Waymo (one of the leaders in the race to autonomous taxis) to assure that these robotic cars are safe. The operation is extremely performance intensive because they run the simulations at computer speeds, allowing the virtual cars to travel millions of virtual miles in a tiny fraction of the time it would take on real-world roads.
Wrapping Up: Who Is Executing and Who Is Not
Intel is the weakest I’ve ever seen it. Even though this hasn’t really hurt the firm financially yet, I know the OEMs are upset with it, it still lacks a CEO or a CMO right now, and I’ve heard from many who simply won’t buy an Intel-branded product anymore (due to a variety of reasons including some nasty layoff behavior). It killed its most powerful effort to assure its future: IDF. It is interesting to note that most of us think Pat Gelsinger, who currently runs VMware, would be the best fix for the company, but even he won’t touch it with someone else’s 10-foot pole. And VMware was largely created to mitigate Intel’s bad behavior regarding Intel customers.
AMD is executing well, is much more focused, has an executive team that is focused on the business and not on creating drama for customers or investors, and has the most solid product line it has ever had. In this world of customer, employee and investor abuse, that is refreshing. You just want a company like this to win. Apparently, Wall Street is impressed, and most of the folks here with me appear to be as well.
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+