Intel to further delay its first 10nm processors to late 2019
Some interesting remarks have been made by Intel during a quarterly financial discussion, an Intel representative mentioned that data center products will be followed by desktop processors fabbed at 10nm at the earliest by 2019. So, long story short, all procs for another year will be based on 14nm.
This means that 10nm Cannon Lake desktop products will not be arriving until late 2019. Intel’s 10nm “Cannon Lake” in small volumes have spotted in early stages back in January 2017. However products based on the chip were slated to arrive by the end of 2017, that never happened. Intel instead pushed back its Cannon Lake rollout until 2018 due to manufacturing difficulties originating from the 10nm process. The delay on 10nm chips is not only limited to desktop and notebook computers, as it also applies to servers parts. It seems Intel is fighting yields in the 10nm process.
Micro-architecture | Core Generation | Fab Process | Year |
---|---|---|---|
Sandy Bridge | 2nd | 32nm | 2011 |
Ivy Bridge | 3rd | 22nm | 2012 |
Haswell | 4th | 22nm | 2013 |
Broadwell | 5th | 14nm | 2014 |
Skylake | 6th | 14nm | 2015 |
Kaby Lake | 7th | 14nm+ | 2016 |
Coffee Lake | 8th | 14nm++ | 2017 |
Whiskey lake | 8th | 14nm++ | 2018 |
Cannon Lake | 8th | 10nm | 2019 |
Ice Lake | 9th | 10nm+ | 2019 |
Realistically, the smaller the processor components, often the more problematic the manufacturing process becomes. We've seent hat in the past, and it seems to apply to complex 10nm designs in their yields, e.g. producing a higher number of defective chips per wafer. Interestingly enough, 14, 12 and 7nm seem to be going strong, AMD is on track with 7nm for ZEN2 and VEGA. This year the company will continue to focus on 14nm based processors of which the first generation was released in 2014. By the end of this year, Intel’s fifth and last generation 14nm technology based CPUs should be introduced, which are code-named Whiskey Lake.
Intel to Acquire eASIC to Bolster FPGA Talent and Solutions - 07/13/2018 07:49 AM
Intel is acquiring eASIC, a 120-person custom chip company based in Santa Clara. This team will join Intel’s Programmable Solutions Group. You can read more in this editorial by Dan McNamara...
Intel to halt Extreme Edition branding for its processors (updated and debunked) - 07/10/2018 07:53 PM
Intel might be dropping the Extreme Edition branding for their processors. The news was posted by François Piednoël, an ex Intel employee who has worked there for 20 years working on architecture ...
Confirmed: AMD Zen designer Jim Keller joins Intel to lead silicon engineering department - 04/27/2018 08:27 AM
Yesterday we posted that CPU architecture designer Jim Keller might make a move to Intel. Word actually is that he was fired at Tesla. Intel decided to not delay any announcements and issued a press ...
Intel to use Graphics processor for antivirus scanning - 04/17/2018 07:48 AM
Yes, you read that right,m basically Intel will be using the GPU, to offload the CPU for virus and malware scanning. The first program to utilize the new features will actually be Windows Defender fo...
Intel To Release affordable Optane 800P series SSDs in March - 01/11/2018 09:23 AM
Intel announced the release of a new Optane SSDs series, the 800P Optane SSDs are positioned between the Optane Memory caching modules and the Intel Optane 900P SSDs....
Senior Member
Posts: 3095
Joined: 2008-01-06
Can't see them skipping 10nm at all lol They've spent god only knows how many billions developing it, we'll be using their 10nm for years to come

Senior Member
Posts: 896
Joined: 2008-07-10
Intel's 10nm is the competition's 7nm. They have always been more ambitious in trying to get closer to ITRS node standard. Broadwell was the first sign of lithography size issues delaying a new processor launch by more than a year (where they had to start going to optimization steps like Devil's Canyon).
This time, the long-term wait to see substantial gains isn't paying out well for them as the competition catches up and plays it safe getting there (putting out products more often with smaller process improvements. Intel isn't the only one that does this).
Senior Member
Posts: 5591
Joined: 2012-11-10
Frankly, I blame Krzanich why Intel is in this mess in the first place. Intel pretty much stopped innovating since 2013 (they instead were either buying out companies, or only did minor upgrades to their products). The only especially interesting thing Intel did since then was Optane, which had a rough launch due to those anemic M.2 models. Anyway, Krzanich was CEO since 2013, so if he isn't to blame, it sure is a strange coincidence that Intel became "boring" since he was in charge.
As far as I'm concerned, he deliberately caused that in-company relationship so he could get out while Intel is still holding strong. It's too convenient - it wasn't sexual assault, so it doesn't make him look especially bad. Meanwhile, he gets to leave the company without putting investors in a scare. It sure doesn't look good if a CEO willingly steps down when the future is questionable, so, this allows him to run without throwing the company under the bus. In other words, he knows the future is bleak and wants to leave as a result, but, he also wants to ensure he's not the one blamed for tarnishing Intel's reputation.
Senior Member
Posts: 1219
Joined: 2009-08-19
Frankly, I blame Krzanich why Intel is in this mess in the first place. Intel pretty much stopped innovating since 2013 (they instead were either buying out companies, or only did minor upgrades to their products). The only especially interesting thing Intel did since then was Optane, which had a rough launch due to those anemic M.2 models. Anyway, Krzanich was CEO since 2013, so if he isn't to blame, it sure is a strange coincidence that Intel became "boring" since he was in charge.
As far as I'm concerned, he deliberately caused that in-company relationship so he could get out while Intel is still holding strong. It's too convenient - it wasn't sexual assault, so it doesn't make him look especially bad. Meanwhile, he gets to leave the company without putting investors in a scare. It sure doesn't look good if a CEO willingly steps down when the future is questionable, so, this allows him to run without throwing the company under the bus. In other words, he knows the future is bleak and wants to leave as a result, but, he also wants to ensure he's not the one blamed for tarnishing Intel's reputation.
Could be true.
Senior Member
Posts: 2252
Joined: 2005-03-26
^ I was thinking along similar lines. Is the 10nm still the same architecture BTW? If not that's what Intel needs to keep up in the core wars, especially in the HEDT and server market.