Intel Rehires Nehalem’s Chief Architect to Build a New High-Performance CPU
Intel’s new CEO is already putting his own stamp on the company, even if he doesn’t formally take power until February 15. News broke today that Intel will rehire former Intel Senior Fellow Glenn Hinton to work on an “exciting high-performance CPU project.”
Hinton spent 35 years with Intel and led development on the Pentium Pro (P6) microarchitecture, the Pentium 4 microarchitecture, and Nehalem. He also worked on Intel’s i960 project. Nehalem is what he’s best known for.
A quick bit of history: Nehalem was Intel’s follow-up to the Core 2 Duo family. It debuted in late 2008 (I reviewed it, if you’re curious) and not long before the launch of AMD’s 45nm Phenom II. Nehalem’s micro-architecture was a modest update to Penryn, with improvements to branch prediction, macro-op fusion, and loop stream detection. It was the first Intel CPU to feature an on-die memory controller, and it reintroduced Hyper-Threading at the high end of the product stack.
Finally, Nehalem’s launch represented some major changes to Intel’s product stack. Up until the first Core i7s launched, Intel was still selling a lot of high-end dual-core CPUs positioned for the enthusiast market. With Nehalem, Intel standardized on the idea that a 4C/8T CPU would be a high-end desktop gaming platform going forward, with a 4C/4T CPU positioned as a midrange to upper midrange part. Intel had previously shipped quad-core CPUs into the consumer market, but it had split the space between dual-core and quad-core chips.
Hinton coming back out of retirement to work on a new performance architecture is not a small announcement. It doesn’t sound like he’s coming back just to shepherd whatever project Jim Keller was working on over the finish line. That means we’re probably looking at a new effort. If so, it’ll be 3-5 years before we see the results of this work. It took AMD about 4.5 years from hiring Jim Keller to shipping Ryzen for revenue. Apple bought PA Semi in 2008 and shipped its first custom ARM architecture CPU in 2012. When Intel built Atom, it started the design in 2004 and launched it in April of 2008.
None of this should be read to mean that Intel CPUs won’t continue to improve over the next 3-5 years. It may have taken the company years to break its 10nm logjam, but Tiger Lake offered a solid uplift in terms of CPU performance. AMD’s decision to freeze development on high-performance CPUs during Piledriver was unusual, and the company has since returned to keeping multiple chip designs in flight as well.
Continue reading
In Massive Shift, Apple Announces New Macs With ARM-Based M1 Chip
Apple saw huge success the last time it switched architectures to Intel, but this time? The jury's still out, but one thing is certain: Apple is about to make a lot more money.
Microsoft: Pluton Chip Will Bring Xbox-Like Security to Windows PCs
Intel, AMD, and Qualcomm are working to make Pluton part of their upcoming designs, which should make PCs more difficult to hack, but it also bakes Microsoft technology into your hardware.
Why Apple’s M1 Chip Threatens Intel and AMD
Intel's own history suggests it and AMD should take Apple's new M1 SoC very seriously.
How L1 and L2 CPU Caches Work, and Why They’re an Essential Part of Modern Chips
Ever been curious how L1 and L2 cache work? We're glad you asked. Here, we deep dive into the structure and nature of one of computing's most fundamental designs and innovations.