The idea was simple, with a two year upgrade cycle for new processes, a two year cadence for new silicon architectures, and a yearly release of updated chips, Intel has fairly regular introductions to plan around. New process and new architectures both tend to have teething problems, and doing both at the same time is essentially asking for trouble. In case you can’t see where this is going, the idea behind tick-tock is to spread out your pain points so you never have two at the same time.
With that in mind, Intel’s approach was pretty basic, never introduce a new architecture on a new process, and vice versa. With new architectures such as Nehalem, Sandy Bridge, and Haswell coming every other year, and minor updates like Westmere, Ivy Bridge, and Broadwell every other year, new processes were introduced with the updates. That means you never get two big potential problems on the same chip, something that has bitten AMD very publicly several times in the past. For Intel anyway, the whole concept has worked out pretty well until recently.
Note: The following is analysis for professional level subscribers only.
Disclosures: Charlie Demerjian and Stone Arch Networking Services, Inc. have no consulting relationships, investment relationships, or hold any investment positions with any of the companies mentioned in this report.
Latest posts by Charlie Demerjian (see all)
- SemiAccurate has Skylake-SP die shots, sizes, and more - Jul 21, 2017
- Skylake-SP has a diverged core - Jul 19, 2017
- Intel’s Purley platform architecture is a step forward - Jul 13, 2017
- Intel launches Purley aka metal Xeons - Jul 11, 2017
- AMD’s Epyc has lots of connections - Jun 26, 2017