MP4 Playing Issue on Toshiba Gigabeat MEGF40

[Llorean]

My understanding from the guys using libmtp to access the Gigabeat S is that right now, it's far from working as well as one would want it to if trying to actually sync a device using it.

I still think a better option would just be some sort of standard "device capabilities" file that UMS sync tools could use.

There's absolutely no reason why a UMS sync program couldn't do everything he wanted to see a program do, if designed properly.

[saratoga]

My understanding from the guys using libmtp to access the Gigabeat S is that right now, it's far from working as well as one would want it to if trying to actually sync a device using it.

I think this is because MS plays fast and lose with the spec though (or even ignores in in the case of the Zune), not because the spec is bad.  Presumably if someone targeted libmtp, this wouldn't be an issue for us.

I still think a better option would just be some sort of standard "device capabilities" file that UMS sync tools could use.

This is basically what the iTunes format is.  UMS + binary database file.  Since theres already libraries out there for it, IMO it'd make more sense to just use that.

[Llorean]

I'm not suggesting a database file (creating the Rockbox database would be a second, independent feature).

Rather, a standard for a sort of features.txt that can list supported audio streams, what containers they may be in, video formats, what containers they may be in, etc. Then if you point a supporting sync program to a UMS drive that contains this file, it would know which files can be synced, and which would need transcoding, and even the "optimal" video settings, etc.

Of course, this sort of thing would be more suited for fishing around Mediamonkey and various other music management tools to see if they could agree on something, but if they ever did, there's no reason Rockbox couldn't have its own version of said file covering the new features of the hardware for use with that sort of program.