By: Doug S (foo.delete@this.bar.bar), July 29, 2021 5:44 pm
Room: Moderated Discussions
James (no.delete@this.thanks.invalid) on July 29, 2021 2:52 pm wrote:
> Blue (Blue.delete@this.blue.com) on July 27, 2021 9:15 am wrote:
>
> > Just wish Intel has gone right to “xxA” naming.
>
> I can't help thinking that "Intel 7", "Intel 4" and "Intel 3" are unambiguous (or are ambiguous in
> a good way for Intel). But "20" was always going to be confused with 20 nm, so they did the old Intel
> trick of adding an "A" to the end, like with the Celeron 300 and 300A¹. Then someone asked what
> the "A" in "20A" meant, and someone else was rather too creative with their explanations...
I highly doubt that. No one at Intel would randomly volunteer "A is for Angstrom" on their own without it being official policy. Their press events are precisely planned and rehearsed to every detail, they don't have rogue agents running around making stuff up off the cuff.
They know they look weak talking about having to "catch up" and being forced to rename their processes to avoid looking even further behind than they actually are. They wanted to get a jump on everyone else by being the first to talk about dimensions below a nanometer because being first to go there makes it sound like they're going to regain leadership.
Not that it is necessary, we used to talk about 0.25u, 0.18u, and 0.13u and it wasn't until 90nm that everyone started talking nanometers.
None of it really matters, since the process names have nothing to do with a physical dimension anywhere in the design. It is just a placeholder for "2x the transistors in the next generation" but we aren't even seeing that lately as TSMC only got 1.8x scaling on N5 and 1.7x on N3 - but TSMC wasn't calling those 5nm and 3nm, it is mostly outsiders doing so (maybe TSMC does as well, but probably only because outsiders referred to them that way)
Who knows what TSMC will call the stuff below N2, will it be N1.4 or P1400 or just choose another letter at random, multiply by 10, so X14 then X10 and so on.
> Blue (Blue.delete@this.blue.com) on July 27, 2021 9:15 am wrote:
>
> > Just wish Intel has gone right to “xxA” naming.
>
> I can't help thinking that "Intel 7", "Intel 4" and "Intel 3" are unambiguous (or are ambiguous in
> a good way for Intel). But "20" was always going to be confused with 20 nm, so they did the old Intel
> trick of adding an "A" to the end, like with the Celeron 300 and 300A¹. Then someone asked what
> the "A" in "20A" meant, and someone else was rather too creative with their explanations...
I highly doubt that. No one at Intel would randomly volunteer "A is for Angstrom" on their own without it being official policy. Their press events are precisely planned and rehearsed to every detail, they don't have rogue agents running around making stuff up off the cuff.
They know they look weak talking about having to "catch up" and being forced to rename their processes to avoid looking even further behind than they actually are. They wanted to get a jump on everyone else by being the first to talk about dimensions below a nanometer because being first to go there makes it sound like they're going to regain leadership.
Not that it is necessary, we used to talk about 0.25u, 0.18u, and 0.13u and it wasn't until 90nm that everyone started talking nanometers.
None of it really matters, since the process names have nothing to do with a physical dimension anywhere in the design. It is just a placeholder for "2x the transistors in the next generation" but we aren't even seeing that lately as TSMC only got 1.8x scaling on N5 and 1.7x on N3 - but TSMC wasn't calling those 5nm and 3nm, it is mostly outsiders doing so (maybe TSMC does as well, but probably only because outsiders referred to them that way)
Who knows what TSMC will call the stuff below N2, will it be N1.4 or P1400 or just choose another letter at random, multiply by 10, so X14 then X10 and so on.
