Intel is rather desperate for acceptance on Coffee Lake so they are launching it with a stunt rather than facts. As SemiAccurate has been saying for years now, Intel has abandoned messaging for social media who won’t question their failures.
So Intel is launching their ‘8th generation’ of CPUs aka Coffee Lake. These ‘new’ parts aren’t, they are quite simply warmed over ‘7th gen’ parts which were simply warmed over 6th gen parts. Note the lack of single quotes around the 6th gen, those were actually a new architecture called Skylake. Because Intel’s 10nm process is abjectly broken as SemiAccurate exclusively reported months ago, the Tick-Tock process became Tick-Tock-Sproing with the emergency addition of the Kaby Lake rehashes and Tick-Tock-Spoing-Thud with Coffee.
If you read Intel’s official explanations at Manufacturing Day all is not just well but better than expected. If you look into their explanation in more than a cursory fashion you have to wonder what is going on at the SEC. In any case it is pretty clear that Intel’s 10nm process doesn’t work, they can only manufacture low power, low-speed, small die devices on it for the reasons we explained above, and even then yields are frightening.
Back to Coffee Lake, the warmed over Kaby Lake still on 14nm. Intel is promising a 30% increase in performance, or more, which doesn’t seem to be much of a trick when your core counts are going up between 50% and 100% depending on the model. Much of the model lines have been leaked already but lets dig in to what SemiAccurate independently found and confirmed. Others may have a bit more here and there, and we have more here and there, but we know where our data came from. The numbers show quite clearly why Intel is choosing stunts over a real launch.
Note: The following is for professional and student level subscribers.
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)
- HyperX ships it’s 60 millionth enthusiast memory module - Oct 15, 2018
- Bittware/Nallatech water cools 300W of Xilinx FPGA - Oct 12, 2018
- More on Intel’s 10nm process problems - Sep 17, 2018
- Intel puts out another 14nm 2020 server platform - Sep 11, 2018
- Why Can’t Intel Supply Enough 14nm Xeons? - Sep 10, 2018