My long term bet is on M1 and ARM/RISC5 for the future.

x86 is on the way out and can't compete on power or performance. In the end RISC won, it just took a while to get there.

As things trend that direction AMD and Intel don't really have much to offer, they're competing on legacy tech.

With TSMC providing fabs for designs from anyone, the serious players don't need AMD or Intel. Apple has a massive lead here, but others watching this will follow it.

> x86 is on the way out and can't compete on power or performance. In the end RISC won, it just took a while to get there.

What exactly do you think is the fundamental limitation of x86? Most chips do lot and lots of crazy logic to go from instruction set to microcode, it's hard to imagine that the variable nature of x86 instructions are the limiting factor.

At very least having a big fat decoder costs you power and space, measuring the tradeoff in raw decoder performance is probably dependent on being an engineer working on it at Intel or AMD (or Apple). At very least, Zen 3 is a cutting edge microarchitecture on a cutting edge node with lots of transistors to play with, and it still decodes only half as many instructions per cycle as M1 - an unbelievably wide CPU, which is probably the way forward.

The "x86 tax" (if it exists, I guess) is usually estimated at somewhere on the order of 5% to 10% - which is a lot of money at scale but probably not enough for a total rethink.

Judging from the M1, it seems the most important thing the big fat decoder costs you is not being able to manufacture billions of processors for small devices (where the decoder would be a much larger fraction of the CPU), develop experience and optimizations and capital based on that, and then have those benefits carry over to your high-end processors.

There’s nothing that restricts AMD to compete solely using x86. I think they’re sitting on an ARM core design from the first Zen arch that they never released.

You’re right, there’s nothing to restrict them except inertia and that they haven’t done it so far when the writing is on the wall.

There’s also the coordination problem with windows and third parties. Without vertical integration and tight OS support they’ll fall behind.

Something will have to give, but the future doesn’t look good.

The reason that AMD didn't release their ARM design I think is that they know it is better to only have to compete with Intel than being one of the many ARM designs.

That reasoning is a mistake.

It's like train companies thinking they shouldn't compete with cars because they're in the train business (without realizing they’re actually in the transportation business).

AMD is in the chip business, not the x86 business - and x86 just became a whole lot less relevant for the future.

I'm sure amd could make a monster arm chip but without an oem who would put it in a laptop or desktop and without killer arm apps for windows whats the point?

I think this is why they're in trouble.

Apple has the vertical integration to do it properly. Amazon and Microsoft have the server cloud computing to do it server side.

Without proper coordination from Microsoft, third party OEMs, and AMD they'll all fall behind on computing on end devices.

I think that's part of the reason AMD/Intel will be less relevant, the incentives and market moves don't look good for them.

I don’t think x86 is on the way out. Apple won’t dominate desktop/server market anytime soon.

Not apple - see the graviton cpu's AWS is putting out?

I think the server market will move to RISC faster than people think.