[Kuinox]

The other CPU catch back after several months, the Apple Silicon is a good marketing prowess, the reality is that Apple CPUs wins because they are the first on to be produced on a new lithography node, because Apple can book a larger quantity than anyone else. When the other CPU manufacturer produce CPUs on the same CPU node, the performance is matched.

Here Jim Keller explain why CPU architecture doesn't matter a lot in performance now: https://www.anandtech.com/show/16762/an-anandtech-interview-...

[criddell]

CPU architecture probably doesn't matter.

But then I'm back to my original question: where are the Linux or Windows machines with great battery life and good performance. Is there anything like what Apple sells at Walmart: a 13.3" MacBook Air with the M1 chip, retina display, and 8 GB of RAM for $700. This is a very old machine at this point and still very hard to beat.

[bryanlarsen]

He does talk about why Apple is doing so well. Architecture, but Apple has IC process, top-level designers, a wide well balanced design and RAM bandwidth advantages.

[criddell]

I'm saying all this as a frustrated ThinkPad buyer. It's been 3.5 years since the first M1 machines went on the market and I still can't buy a ThinkPad as good as the Apple computers of that generation. I have to remember to power off completely before putting my ThinkPad in my bag otherwise I will have a very hot laptop, fans at full, with 50% battery remaining by the time I get to work.

Sometimes it feels like Apple is the only company even trying. It sucks.

[bryanlarsen]

Is it still true? My Ryzen desktop is a lot faster than my M1 Mini. A laptop Ryzen won't be as fast as my desktop, but it should be close?

Regardless, it certainly feels like an own goal that AMD or Intel hasn't yet released a CPU with at least 256 bit wide on package RAM. They'd be able to capture a higher percentage of the BOM cost of the laptop if they did so on top of the speed benefits.

[criddell]

The M1 used in the original MacBook Air has a TDP of around 10 watts. I don't track this closely, but AFAIK there aren't any good performing chips on the PC side of the world that can beat that.

I don't really track this directly, but periodically I look for passively cooled (fanless) laptops. I haven't looked into Ryzen, but it would be a very nice surprise if there are fanless Ryzen laptops out there.

[Kuinox]

TDP is a lie. M1 can draw up to 100W. https://www.anandtech.com/show/17024/apple-m1-max-performanc...

[lonjil]

> The other CPU catch back after several months, the Apple Silicon is a good marketing prowess, the reality is that Apple CPUs wins because they are the first on to be produced on a new lithography node, because Apple can book a larger quantity than anyone else.

Uh, but M2 on TSMC's N5 node is faster and more efficient than AMD's products fabbed with TSMC's N4 node. Edit: more efficient in general, but I meant "faster" in comparison with AMD's latest laptop chips.

> Here Jim Keller explain why CPU architecture doesn't matter a lot in performance now: https://www.anandtech.com/show/16762/an-anandtech-interview-...

Let's see what he has to say:

> But instruction sets only matter a little bit - you can lose 10%, or 20%, [of performance] because you're missing instructions.

20% is quite a lot in the current competitive landscape. Jim Keller is stuck thinking like it's 10+ years ago.

[Kuinox]

> but I meant "faster" in comparison with AMD's latest laptop chips.

The 7940HS is way faster than the M2. Efficiency is harder to compare since TDP doesn't mean anything theses days.

[paulmd]

In cinebench, or in (eg) compiling or JVM/Javacode/other bytecode workloads etc?

because even in cinebench the m3 max is already ahead in cinebench 2024 let alone R23...

https://www.notebookcheck.net/R9-7940HS-vs-Apple-M3-Max-16-C...

[Kuinox]

We were talking about the M2.

The M3 max is on TSMC N3B, which have a +30% of transistor density compared to the TSMC nodes AMD is using (N4).