I've been thinking about this for some time, and I have some observations and suggestions. The thing that strikes me most about problems and issues with MC is that it is built on an uncontrollable foundation. What I mean by this is that MC is built on computer hardware with a computer OS underneath it, and MC is constantly at the mercy of the OS developers. For example, when Microsoft forces updates of their OS, then the updates can cause issues with MC that did not exist previously, thus causing needed changes in MC to adjust for changing OS updates, video/audio drivers, subtitle renderers, etc. ---OR--- MC must wait until the OS developer "fixes" the problem that they caused, which may or may not ever happen. Basically, things that work fine this week may be broken next week simply because of poor or inconsiderate OS development and not the fault of MC.
So, how do you fix this problem? One answer is to build your own hardware, install an OS which you can control (like one of the Linux flavors), and then putting MC on top of it. In this way JRiver can now completely control their own environment from foundation to roof without having to deal with unexpected, unknown changes in the OS/drivers. The underlying OS should be mostly invisible to casual (or maybe even all) users in order to keep things as simple as possible to the end user, as well as preventing users from screwing things up.
For example, this is the route that the Vero 4K has gone. They sell a box which is basically a Raspberry Pi with their own variation of Linux running on it, the OSMC engine. They don't have a product anywhere near as good as MC, but they do have their own "modified for them" version of KODi running on top. When updates are released, they update the lower level OSMC, as well as their custom modified KODi, as a package in order to make sure (as much as possible) that everything will work correctly. The limitations for them are whatever the hardware is capable of doing, along with the imagination, creativity, and raw abilities of the coders/programmers.
The biggest problem with this approach is that hardware keeps changing, being improved, fixed, etc. and that alone might be reason not to go in that direction, but if the cost of the underlying hardware was relatively low, then every couple of years or so, or whenever the hardware changes SIGNIFICANTLY, then release a new hardware box which can perform all of the latest improvements in audio/video performance, as well as new bells and whistles in the front end, cloud access, streaming, etc. If a user is happy with his "old" unit, then it will still perform fine, as there will not be any OS developer making changes to screw it up. If the end user wants all the new stuff, then just buy the new box...hopefully everybody wins.
This is just a suggestion for your consideration. Like I repeat often, I am not a techy guy who has a clue as to whether this approach would be desirable, reasonable, or feasible, or simply too much work, but it is a new direction for JRiver II, and that's what the thread title asked for.
Personally speaking, if JRiver offered a hardware box like this, I would RUN, not walk, to buy it, as I consider MC to be by far the best software of its kind, with plenty of room to make it even better by developing it on all of its many fronts.
I want to buy the "JRiver Media Monster" or "JRiver Media Octopus" as soon as possible!