The past as in before 2021? They committed to “5 nodes in 4 years” plan in 2021, and they seem to be on track to delivering on that. This is an article from 2021:
The implied expectation is that a company with a history of innovation in the semiconductor space, including vertical integration between chip design and fabrication, would continue to innovate and deliver.
For those that are not aware Intel previously highlighted successful milestones for its 10nm process and claimed that it was on track. For example:
> Intel's official Twitter account has tweeted that Ice Lake has been taped in at the 10nm process.
> The company still holds the industry advantage in terms of process that they have always enjoyed (assuming you don't fall for the marketing names that other foundries employ which are not indicative of the physical process) and appears to be confident in its promised delivery of 10nm products.
The difference this time is that they have already shipped or are in the process of shipping products on new nodes (Intel 4, Intel 3) in a relatively short period of time whereas before the 10 nm node they were stuck on 14 nm forever. So they’re executing on new nodes.
Whether they will continue and whether it will be any good is another question but it’s hard to argue the current situation parallels 2017. There’s been demonstrated progress, government money, and new leadership.
I do have to balance that a little with the fact that a lot is resting on the success of these nodes. It's almost the only good news Intel has, so we should be skeptical. That wasn't the case in 2017.
Intel 10nm was delayed repeatedly, however to be fair, this article is from mid-2017 and Intel would launch 10nm in 2018. Years after they had originally road mapped, but by the time the taped in they were largely back on track to launch.
Speaking of which, is there a post-mortem on Intel 10nm somewhere?
I distinctly remember Semi Accurate swinging hard on the idea that it not only didn't work, but would never work (actual analysis behind paywall). I think this was in 2017 or 2018, and even after years of delay I hear that the Intel 10nm process still has yield problems. How did he / his sources know that the problems wouldn't go away?
Do you maybe want to add a sentence or two tying this link to the current discussion? I feel like you could have done that instead of “just leaving this here”.
I like a Reddit cliche as much as the next person, but this cliche is usually used with things that are self-evident. Your link is not.
