9 comments

[ 0.26 ms ] story [ 34.7 ms ] thread
Wow - I wasn't really excited about the ARM change, but I can see with the SoC changes it enables the progress can be amazing the new few years.
It is interesting at least, main concern for me is running software that doesn't have an ARM compiled version.
thanks to ever added [video] sorry I forgot!

You can find some of Gavin's prior work on ioccc :D

Purpose-built hardware makes sense for battery performance, and battery performance makes sense for 2020.

x86 is a trainwreck.

I wonder how Mac's with dedicated GPU's will be supported with the new unified memory architecture
It'll probably behave as it does now if you try using unified memory in Metal (i.e., you will get an error). It's currently only valid in iOS/iPad/iEtc devices.

Have a look here: https://developer.apple.com/documentation/metal/mtlstoragemo...

I guess there will be a compile-time flag that you can #ifdef to check if you are on a new Apple Silicon mac or not.

It looks like Apple will be going it's own way with GPU, as tight integration between GPU and CPU enables much better performance. Think vias rather than buses.

The higher end Macs should see some large on-chip (or part of chiplet array) GPU implementations.

It'll largely be transparent at the Metal API level.

Interestingly, from a business perspective this is mainly bad for AMD.

Really well thought system level engineering.

I was thinking to shift to Surface Book 3 but i think i should wait until end of this year.

At 13:50, is he saying that x64 code running under Rosetta will enjoy TSO memory ordering? Is Apple implementing something stricter than the ARM memory model in their own silicon, maybe with a mode bit that Rosetta gets to set?