Super nerdy post but wondering if TR will be porting over desktop TR app to Universal Mac app for native support of the new M1 Apple Silicon?
Just did a work out on my new MacBook Pro and it’s consuming 3x less battery (even though it’s running using Rosetta 2 emulation) than my 2015 13" MBP. Speed on opening is significantly improved and it syncs much quicker.
I’d beg to differ. Granted a 5 yr old battery has lost a bit of capacity, I’d disagree that TR isn’t computer intensive especially if you factor in group rides. It’s constantly pulling and pushing data through BT. I barely made it through an hr group ride on battery under those conditions.
I totally forgot about the group rides. That is definitely a different story, especially on an older MacBook. I too have a 13" 2015 MacBook Pro which could use a battery replacement.
That said, I have no doubt that TR will have a native M1 version in the near future.
I reckon they are using some cross-platform frameworks, so perhaps it is a bit more difficult than that. Especially if some of the libraries are only available in the form of binaries.
At some point Apple will set a date for the sun setting of Rosetta 2 just like it did with Rosetta 1. Rosetta 1 lasted 3.5 years before it was not included in Snow Leopard.
That is surprising: Apple is moving its entire line-up of computers to ARM, and by now they are more than half-way there. The move was announced in 2020, so TR has had >1.5 years. Don’t you have some proper Mac heads at TR?
It is a tremendous amount of work to rebuild something for a whole new architecture. It might be more effective to just have them “port” the iPad OS TR to allow it to run on the new M1 Macs.
I think that would be the most economical way of doing it.
I suspect the actual dev work is fairy minimal. The Mac app is an Electron app so I’m guessing most of the work to port the code is already done. (I’ve never shipped an Electron app, so could be mistaken) It will be the typical overhead of shipping such a change that costs the most. e.g., testing, PMing, etc.
And whilst this work is being done the PO will be wondering why their features aren’t being worked on.
This won’t be prioritised unless there’s something significantly broken whilst running on Rosetta or until Apple announces a sunset date for Rosetta.
Given Apple is in the middle of a two year transition to arm64, that means there will still be Intel Macs shipped over the next year. Earliest date for Rosetta deprecation is maybe 2 years from next year’s macOS release i.e., 2024.
Not on modern platforms and software that does not rely on lower-level APIs or platform-specific optimizations, and especially for software based on the cross-platform framework Electron. That already runs on ARM: if you have been using the iOS or Android app, you are running a native ARM app.
Agreed, I would expect that there is no or next-to-no lower-level code and no e. g. hand-coded high-performance code that uses e. g. AVX instructions (I couldn’t see what for). So it should be straightforward.
Sure, but keeping up with platforms is just the cost of doing business. With Electron apps, this is fairly minimal anyway. Probably changes in iOS’s Bluetooth stack will cause TR much more headache than an ISA transition.
I hope not, that would be recipe for disaster. I’m much more worried that TR staff in charge of the app apparently hadn’t heard of that (at least it wasn’t on their to do list according to @IvyAudrain).
Apple has the ROSETTA2 emulator on the M1 computers. That will allow running non m1 specific apps for a long time.
Also, m1 is a ARM chip, which means that the any app that runs on ARM can potentially also run on m1.
Apple iDevices have run on ARM architecture for a long time. And the move to ARM on the rest of the lineup is partially because is hard to maintain 2 different eco system. Their idea is that any app that run on iDevice and also run on regular computers.
What all of this have to do with TR?
Well, they do have an iDevice app. That probably can be migrated to be use on Mac computers.