Intel 7nm pushed to 2021-2022 timeframe according to the following article:
http://www.fool.com/investing/2016/09/02/intel-corporation-may-have-pushed-7-nanometer-tech.aspx
By then TSMC will have 2 FULL YEARS of high volume prod on their 7nm and will be well on their way to 5nm, if not shipping it already. So that means TSMC will have completely reversed Intel's original lead of 1 full process node ahead.
The whole TMG has failed .. TSMC / Samsung has caught up and sadly there is little reason for any fabless to consider Intel Foundry. The only reason they do is that Intel is giving away the silicon in a desperate play for PR and to fill rapidly dropping x86 orders. Thanks to BK and his minions from TMG for loosing the leadership in Moore, like ACT would fix it!
https://www.semiwiki.com/forum/content/6160-2016-leading-edge-semiconductor-landscape.html
I cannot help but wonder if things would be different if Dadi Perlmutter was still onboard.
And It will be even more delayed after ACT
Ashrat used to be an intel fanboi.. but one can only be disappointed so many times before they figure it out that Moore's Law is over at Intel and the mantel for two year cadence goes to the ARMy makers who have 300million + business and growing to Intel's shrinking x86.
The CEO and TMG monkeys, were they laid off with this round of ACT, if not why not?
The schedule for 10nm is believable though: 10nm in 2018, 10nm+ in 2019, 10nm++ in 2020. Intel has already confirmed these process steps publicly. If this holds, then 7nm is indeed slated for 2021. For argument's sake, let's say the initial 10nm process has an unusually short run with Cannonlake in 2017-2018, then IceLake moves to 10nm+ in second half of 2018, and Tigerlake in 10nm++ by 2019. We still wouldn't see any 7nm products before 2020, which gives TSMC more than a full year's lead.
Intel labs have no connection with product groups.
Ashrat is just full of sh!t most of the time.