Happy* IINA user here. I use youtube keybindings + 'video speed controller' (chrome extension) keybindings for quick speed up/down/reset and skipping back/forward through videos).

* It did have a problem with crazy power consumption, but I think that's been fixed (haven't noticed the problem for a few months).

I am glad there is an OSS player available, but I am never going to be able to remember this name.

hmm I'm confused, whats the sell with inaa compared to vlc?

Not that new, I've been using IINA for years! It's pretty good.

IINA has been around for quite a while, it's a front end for mpv. Pretty much Just Werks, it's my standard video playback client.

It's a player of the ffmpeg lineage; a GUI frontend for mpv. If you use mpv or some frontend (e.g., SMPlayer, Celluloid, or Haruna) on other platforms, IINA is a natural choice on macOS.

It's free software, it's keyboard friendly, it supports more of less every format, and it blends into the OS's native desktop environment. I don't have an eye for design, but it seems good to me. If there's a better choice for local multimedia playback on macOS, I don't know what that is.

(It seems the only other real contender on the backend is VLC, which is likewise excellent software.)

Another vote for Infuse. In my experience Infuse 8 handles HDR and color management correctly (1:1 with Safari or QuickTime), while IINA does not (too dark).

I think QuickTime still doesn’t play MKV? Anyways, I usually use IINA just for the keybindings.

I was really hoping that butterchurn was a hi fidelity rendering of butter churning to the pacing and sound range of any given media that was playing

A lesson learned over and over again is that you can build amazing things with tightly coupled plugins, but once you have about three of them upgrading or changing anything gets kind of impossible - Hyrum's law. On the other hand, loosely coupled out-of-process plugins are a lot less flexible, but tend to work a lot better and more reliably, at the cost of more up-front engineering and overall investment. Assuming that what they can do is sufficient to work.

Consider that e.g. kubernetes has basically just one actual core component (the API server) and everything else is loosely-coupled plugins. Alternatively, consider any of the projects stuck for 15 years on Python 2 because that's what their plugin system was in 2009. These are two points on a spectrum.