This site may earn affiliate commissions from the links on this folio. Terms of utilize.

Twenty-one-twelvemonth AMD veteran and HSA (Heterogeneous System Architecture) president Phil Rogers has decamped from AMD to join Nvidia. Rogers will be taking over every bit Principal Software Architect of Nvidia'due south Compute Server division, at a time when Team Green is rolling out features like NVLink and continuing to button frontwards with its ain plans to brand CPU and GPU compute more capable. With HSA one.0 now complete, you could contend Rogers is taking a folio from Jim Keller's volume and moving on at present that he'southward finished his work.

That's possible — but I honestly doubt information technology. Jim Keller has a longstanding reputation as a problem solver, and a history of spending just a few years at whatsoever given company. Over the course of his career, he's worked for DEC, AMD, SiByte / Broadcom, PA Semi, Apple, and AMD (again). Rogers, in contrast, was a longtime AMD employee and the very public confront of the entire HSA initiative. The trouble is, AMD'southward ability to actually create an ecosystem effectually HSA capabilities is extremely limited.

HSA: A neat idea, merely a lousy near-term strategy

When AMD bought ATI back in 2006, it talked about a "Fusion" of product families that would create synergy between the 2 companies. At beginning, that meant ameliorate integrated graphics for desktop and mobile processors, simply AMD had something far more grand in listen. Integrating a GPU straight into the CPU was a tremendous technical accomplishment, only the existent goal of Fusion (subsequently HSA), was to provide a programming model that allowed developers to write lawmaking that would run seamlessly on whatever processor block was all-time able to execute it.

AMD's original HSA roadmap

AMD's original HSA roadmap

Ane thing I want to emphasize is that HSA is a great idea. There's a reason why companies like Qualcomm, ARM, TI, and Samsung all signed on with the HSA Foundation, and why these companies go on to improve the heterogeneous compute capability of their own solutions. Every major semiconductor visitor on Earth has taken steps towards heterogeneous compute, from Qualcomm'due south Snapdragon 820 and its CPU – GPU – DSP triad, to Intel and the Xeon Phi. The idea of matching the correct workload to the right processor is stiff, and AMD wasn't incorrect when it recognized that many-core heterogeneous architectures would be critical to long-term performance improvements.

But technical excellence and early design leadership don't e'er translate into increased market place share. The problem with HSA is that AMD utterly lacked the resources to move the market, every bit a whole, towards adopting it. Kaveri's OpenCL functioning improved on Richland in a number of ways, as we documented when that chip launched, but 18 months after Kaveri launched, there's virtually no HSA software in the marketplace.

This isn't the first fourth dimension AMD has led in technical evolution simply depended on other companies to ultimately drive adoption. Two of the company's previous technologies, HyperTransport and x86-64, took like paths. The departure, however, is that AMD was able to make substantial employ of HT in its own hardware — information technology collection the "glueless" architecture that fabricated Opteron servers and then compelling compared with Xeon alternatives in the 4P space back in 2003 to 2005.

As for x86-64, not only did AMD win huge accolades for stealing a march on Intel, it ultimately forced Intel to adopt its ain standard for the future of 64-bit on x86 CPUs. HSA wasn't intrinsically useful to AMD APUs without substantial software back up, and AMD simply lacked the funds and evolution resources to drive wide adoption. AMD's APUs may be "good enough" for vast swathes of the market, but that hasn't kept the company'due south sales from collapsing. "Good enough" doesn't kindle the imagination or get developers excited to work on your platform. And while it'south truthful that GPU performance has been a highlight of AMD's APUs for many years, great graphics functioning (relative to market segment) didn't require HSA features in the kickoff place.

Information technology should exist noted that these problems aren't at all unique to AMD. After briefly flirting with consumer-level CUDA applications, like the media encoder Badaboom, Nvidia largely left the space. Yous can notwithstanding find media encoders with GPU support, to exist sure, simply there's been no push button to bring GPU dispatch to casual content or to widely leverage OpenCL in every day applications. GPU acceleration remains the province of workstation-class software, for the virtually part.

So, why move to Nvidia? Because the supercomputer / HSA space offers a much more fertile basis for the kinds of improvements that heterogeneous compute can offer. CUDA is already well-established, AMD doesn't seem to take whatever serious plans to attack the marketplace in the virtually-term hereafter, and Rogers likely wants to run across the improvements he'due south designed bear fruit in bodily aircraft software. In the HPC infinite, building new software capabilities to have reward of hardware is part of the job, and while Nvidia faces its ain headwinds in that market place thanks to Intel and Xeon Phi, it'due south all the same a much meliorate overall position.

I don't think AMD will drop HSA or heterogeneous compute going forrad, simply AMD's marketing materials don't actually focus on information technology these days. That'southward likely wise. Fundamental CPU performance and ability efficiency are far more than critical to the visitor's future than any improvements it tin can incorporate into heterogeneous compute. Boom Zen's debut and future APU launches, and the HSA question tin be dealt with at a later on date.