I assume that this is not the best time to inquire the following point, but is there any long term incentive of integrating Fst-host into Ardour?
My thought is that it could improve a certain workflow in restoring sessions and keeping the worspaces compact.
I am interested, if there is a current community opinion on that.
Isn’t fsthost just a WinVST host for linux? - if so, Ardour can already be compiled to support WinVSTs natively (using the fsthost code I think) - either way, if you try to use a plugin compiled for Windows, you end up using WINE - which makes the whole ardour / plugin stack quite fragile - so in fact, using a separate plugin-host is a better way to do it (actually the better way to do it, is not to do it at all, and run Windows plugins on Windows).
i was actually thinking a bit in the way of having a checkbutton before install to bridge fsthost to ardour in some way as mentioned here.
http://sourceforge.net/p/fsthost/discussion/general/thread/6b1caea3/
couldnt then vst crash as much as they wanted without affecting ardour?
the advantage of bridging this from running a pluginhost or a standalone vst within wine would be not to have to deal with wineasio for audio and using the midi out capabilties from ardour directly to drive the plugins.
further thoughts greatly appreciated
I can’t speak for the ardour devs, but I see in the thread you linked to that it mentions something like dssi-vst (personally I think dssi-vst is completely the wrong way to do things, and should only ever have been considered a short term fix - if you can even call it that). If you want to ‘sandbox’ the WinVST / WINE plugin host code in some way, then I guess the logical way would be to use JACK - as the core IPC method - which essentially is what already happens when you use fsthost, albeit with some very clunky session management, as is the current ‘state of the art’ on linux.
Stepping back from the problem a bit, it seems to me that having to work out a method to integrate with ardour without really integrating with ardour (i.e. using a separate process and IPC) implies that the plugins are a little volatile, and the reason for that is mostly to do with trying to run binaries compiled for a completely different OS. Once you factor in all of the reliability issues (and time lost if / when you then come to depend upon some Windows plugin - which subsequently fails after a WINE ‘upgrade’), have you really gained anything over simply running your Windows software on Windows?
Yes you do, but not a whole lot, and nothing good session management and flexible osc and/or midi control couldn’t solve to be honest.
To the op, see Carla and the recent work to get it running as a lv2 plugin, which is a bit different way of doing this with integrating and maintaining the code direct in ardour. I am curious to see if it will be able to recover from a plugin crash in Carla without taking down Carla and thus Ardour as it will be similar to any other plugin crashing.