Apple Silicon M1 chip in MacBook Air outperforms high-end 16-inch MacBook Pro

I wonder what Linus thinks of Geekbench's accuracy now:

https://www.realworldtech.com/forum/?threadid=136526&curpost...

https://www.realworldtech.com/forum/?threadid=185109&curpost...

Personally I'm not convinced, until I see something like SPECint/SPECfp results.

anandtech has specint numbers for the A14, which should serve as the floor for M1 performance. M1 will (likely) clock higher, has more cores and sustain thermals longer.

https://www.anandtech.com/show/16226/apple-silicon-m1-a14-de...

I’m also interested in seeing what M1 can do once people get their hands on real hardware running on Mac OS where so many more details can’t be hidden like in iOS, but all signs point to it being an absolute monster.

> We've seen this before: cellphones tend to have simpler libraries that are statically linked, and at least iOS uses a page size that would not be relevant or realistic on a general purpose desktop setup.

This is an interesting thing to say considering that 1. most apps on iOS use large dynamic libraries and 2. Apple silicon Macs run on 16K pages, just like iOS.

He'd probably say it's garbage, since all the reasons it is garbage remain true. Getting yet another garbage result hardly changes the argument.

Here is a video of what the other Linus thinks: https://youtu.be/ljApzn9YWmk

The M1 runs with 4k pages, unlike the iPhone chips.

I believe that is only for Rosetta, and A14 shares the capability but does not use it.

we don't know Big Sur's optimizations, it is possible they used a number of iOS techniques within it.