Don't know if this is the right thread to put this in but I came across an interesting (but not surprising) response from the developer about not bringing their game, in this case Trine 5, to Mac (and Linux). The first three games in the series did have Mac versions but the fourth didn't and neither will the upcoming fifth entry in the series. You can play the fourth one on GeForce Now though. Perhaps the fifth one will be available there as well? I wonder if these would work with the GPTK?
Here are the developer's comments.
Unfortunately, Linux and Apple user bases are too small for us to target. This is especially true these days, as the platforms are almost completely different. Porting to Linux won't help porting to OS X and porting to OS X won't help porting to Linux. While Linux development is somewhat easier than it used to be (thanks to Steam OS), OS X development has only gone more expensive.
We did consider a Linux port for a while, but in the end didn't have resources to attempt it. It's possible in the future, if Valve keeps up the good work with Steam Deck, but I wouldn't suggest waiting for it. Trine 5 should run fine on Proton, so native Linux port doesn't really have much to offer.
-JLarja / Frozenbyte
-----------
Every developer and publisher is in a different situation. It is possible to imagine a situation where it would be beneficial for us to make the ports (especially the Linux, or Steam Deck, one).
Please also note that it's not about whether doing a port would be profitable by some calculation. We don't have a pool of (Linux) programmers that we could tap to do a port and then put them back to storage, while we wait couple of years for the next time they are needed. If we were to do a port, then we aren't doing something else that is (as far as we can tell) more profitable. Considering all the money Frozenbyte has so far made has been put back to develop more games, it would effective mean less games from us.
All this, obviously, has lots of variables, and it's impossible to say what would make sense in the long run and what wouldn't. If it was up to me, I would spent the resources on Linux port, as the development environment for Steam Deck seems pretty good and we do have (more or less) working Vulkan renderer from Stadia days. I can't come up with any numbers that say it's a good idea though. it's mostly just that I like Linux and it's close enough call that I can't come up with numbers to show it's a bad idea either.
About Proton: Using a translation layer isn't really a bad idea at all, if instead of hoping for perfect solution you embrace good enough. Much of the programming is building layers upon layers, abstracting the details (like on what hardware the code is running) until the details don't matter anymore (for the code at hand. They obviously matter for the absraction layer).
Badly done native port is certainly much worse than many Windows games currently running on Proton are (I think Trines 1-3 actually demonstrate this quite well). Again, I don't have any numbers, but if we were to make a Linux port, it couldn't receive much optimization. Maybe being native would offset that, maybe not. Maybe the game would run worse in 30-60 FPS range but better around 120 FPS. I would personally consider it a failure (players at 30-60 FPS need the extra frames much more), but some would paint it as evidence of greatness of native port.
All that said, we are open for third party porting. If THQ (publisher of Trine 5) wishes to outsource Linux or OS X porting to third party, it will likely happen (obviously we want something out of it, because otherwise it would be a net loss to us). So even though I'm pretty sure we won't do it, it could still happen.
-JLarja / Frozenbyte