Is Pro Tools behind the curve?

[JFreak]

Even back then it was mostly ticking a box and recompiling, most of the work was dealing with random bugs. I bet they (in-house development a.k.a. alpha team) already have it working on ARM but sure it is still long way to be ready for beta testing. Beta testers do not have hardware yet.

[Darryl Ramm]

Originally Posted by JFreak

Even back then it was mostly ticking a box and recompiling, most of the work was dealing with random bugs. I bet they (in-house development a.k.a. alpha team) already have it working on ARM but sure it is still long way to be ready for beta testing. Beta testers do not have hardware yet.

It is almost certainly not just as simple as checking a box and recompiling, and I've already pointed out some of the reasons why out to you in past threads. I would expect there needs to be performance optimization work, and there might be significant compatibility stuff to be done. I'd be very curious what happens with Metal graphics and Apple GPUs on these systems as OpenGL gets further deprecated (and plugin GUIs today still depend on OpenGL.. and so do other lots of other third party apps.. so we'll see). And there are all the Avid interface device drivers, and they may also require more effort. It will be a time where I expect lots of stuff close to its end of life to not make it across. Avid currently has problems shipping a HD Driver with an up to date installer compatible with latest macOS releases... I don't take that as a good sign they are going to breeze through this stuff. And then they have to drag along plugin vendors, and are probably not in a position to fund them with NREs/incentives to make that transition.

[JFreak]

Will see. Drivers is the biggest problem for sure, expect some stuff get left behind.

[JFreak]

Remember, driver development and PT development go different route. Avid may already have ARM hardware running their software with generic built-in output. PT itself is "easy", but drivers are low-level and need more time. Anything that is not supported now will not be working in ARM platform, pretty much given.

[deanrichard]

Re-compiling code for a different processor is never easy. It will be a challenge. You can all but guarantee it will be much more than a straight recompile, and some code will need to be modified. So if anyone has been on the fence about switching to PC from Mac, this might be a good time to consider it. The first ARM version is likely to be a work in progress.

Originally Posted by JFreak

Remember, driver development and PT development go different route. Avid may already have ARM hardware running their software with generic built-in output. PT itself is "easy", but drivers are low-level and need more time. Anything that is not supported now will not be working in ARM platform, pretty much given.

[deanrichard]

You sound like some of the marketing people at places I worked, back in my software engineering days. "Nothing to it" they would be telling upper management and customers. Meanwhile the SE's have to work out the details. Some ports are easier than others, but I have to think going to a different processor will be a significant task for an app like PT, with all of its time critical code.

Originally Posted by JFreak

Even back then it was mostly ticking a box and recompiling, most of the work was dealing with random bugs. I bet they (in-house development a.k.a. alpha team) already have it working on ARM but sure it is still long way to be ready for beta testing. Beta testers do not have hardware yet.

[Philthy]

Will the switch from Intel to ARM have any impact on RTL? Like, how does latency through ARM processors compare to x86?

[Darryl Ramm]

Originally Posted by Philthy

Will the switch from Intel to ARM have any impact on RTL? Like, how does latency through ARM processors compare to x86?

The CPU is not the issue there.

[deanrichard]

Do you know that from technical expertise, or are you making an assumption? From the outside looking in it seems like there could have some impact. A different CPU will have different scheduling algorithms and require different drivers, which seems relevant. Of course it could be a positive impact, and latency might be better, depending on the drivers.

Originally Posted by Darryl Ramm

The CPU is not the issue there.

[JFreak]

RISC usually has less realtime problems


Sent from my iPhone using Tapatalk