Maynard Handley (name99.delete@this.name99.org) on November 19, 2020 8:22 pm wrote:
> OK guys, M1 time is over. What's next?
>
> Something I just realized, and I think the timing works out beautifully:
> TSMC started N5P testing/Risk production in 2Q2020.
> Meaning that (and now that I think about it, it is so obvious!) the M1X will be on N5P!
> THAT is what's determining its schedule.
>
> There is a precedent. Remember that the A10 came out in Sept
> 2016, on 16nm. But the A10X came out in June 2017 on 10nm.
>
> This makes so much sense. It gives time Apple to improve some of the rushed bits of the M1,

What bits of M1 are rushed? What's your basis for saying that? This is a derivative core of a now very solid core microarchitecture. Does their new hypervisor implementation have bad errata, or their x86 emulation sub-par, or something?

The power, memory, IO interfaces or possible periphery around slightly higher thermal power headroom could be at a less mature point though I suppose.

and gives
> a free 5% speed boost (which isn't much, sure, but may mean that M1X clocks at, say 3.5GHz)
> And gives Apple a second round of (totally justified!) "OMG, Apple is king, everyone else is doomed" publicity,
> say maybe in April or May, which should sustain them till the A15/iPhone next reveals in September.

M1X is for Macbook Pro and larger higher TDP computers isn't it? I don't really follow, how is this related to smartphone market?

I am very interested to see higher power headroom, more cores, and bigger IO and memory etc. regardless of the process it's on, surely it would allow for some decent performance gains.

>
> Well, six months or so to see if I'm right. tick tick tick

It will be interesting to see which way they go with their CPU program now that it's getting more devices now sharing same/similar core.

On one hand it would make a lot of sense for this M1X to retain the same process and practically the same core (and much of the IOs) in order that logic and physical designers have mostly done with it and most of the work goes into putting the next core on the next process.

The other line of thought is that it is beneficial for a known good core to be ported to the next process to better characterize it and tweak and tune your physical design and circuits etc.

I could see it going either way, I don't know details of economics and benefits of different approaches to know what is more likely.