Hello, I’m thinking to buy a MAC Book pro (I would definitely leave Windows, because even if I’ve an HP intel core i7 8th Gen 16GB RAM buyed 2 years ago, but since Windows updated it need 5-10minutes to load the windows!..I tried to format it some weeks ago, but it didn’t solve, then I’m fed up of Windows)
It would be my first mac…So I don’t have any idea about the Mac world! The Apple assistance said me to buy a M1 processor Mac Book Pro…
then I would like to know
if ROOT works on M1 processor!
For physics software (Geant4, ROOT, MAthematica, etc ) is better a M1 processor MAc or a Intel 7 processor MAC?
BTW I know it isnt’ a ROOT question, but I also have on my windows Geant4, Matlab, Mathematica, if you have M1 processor Mac Book, do you know if they work?
Thank you
There are still a couple of issues with software support for Apple’s M1 ARM, notably ROOT / llvm and exceptions. That’s really a terrible issue. For the rest it mostly works - you can always use the emulation layer if you cannot find an M1 port of a software.
The chip is certainly faster than the Intel generation used for macbook pros before; in that respect I can recommend it.
For any serious work I’d certainly recommend using Linux, though - there’s quite some choice for laptopts with Linux pre-installed, and even more so for laptops that have parts that Linux supports. And exceptions work, too, and you don’t get into the macOS security hassle of " I’m sorry Dave, I’m afraid I can’t let you run this program". I’d personally get a new AMD Ryzen mobile, plenty of RAM, Linux - that’s a lot of fun to work with!
I have MacBook Air with M1 processor and from my experience there’s no problem with building and running ROOT natively on that machine without Rosetta 2. One has to build ROOT version with this commit resolving some bug with M1 and use those options for cmake:
There are still a couple of issues with software support for Apple’s M1 ARM, notably ROOT / llvm and exceptions. That’s really a terrible issue.
Can you please elaborate a little bit on that or provide a link to the corresponding issue?
Because I’m aware only of the issue with computing the correct ABI in arm and it was resolved in the commit mentioned above.