MIDI Passthrough

Locked New Topic
RELATED
PRODUCTS

Post

dluther wrote:
Once a strategic decision is made poorly (like midi pass through in the midi routing table) it can't be easily taken back because the data occurs in sessions then and has to be supported. On a change you have to support both to be compatible with legacy data... not a very great situation.
This is interesting. As a new Cantabile user, I came up 'against' MIDI passthrough as I was learning the system and came up with the following solution which works for me - I was wondering if anyone else had come to the same conclusion:-

I found that it wasn't easy to see at a glance whether MIDI passthrough was enabled for a particular routing and what the 'flow' of MIDI data through the racks was. I was also getting situations where I couldn't tell via what path a sound was occurring - whether it was happening as a result of a MIDI routing or simply due to 'spare' MIDI data coming out at the end of the list (dunno if any of that just made sense...). I also found that not being able to choose between passthrough or no passthrough for any particular sub-session was causing me to have to create more routings than was necessary. And in any case, having the option at all seemed overly complex. I felt that it would be better to ALWAYS have passthrough on, but then have a way of blocking the end of the list up so that only the active routings in a sub-session play sounds.

Therefore, I realised the following. If I always set MIDI passthrough to ON for every MIDI routing and then just create a routing called -=END=- (or whatever) with NO passthrough at the end of the list which took all incoming MIDI data and sent it nowhere (ie. destination NONE), then everything works brilliantly. I can create routings with splits, filters, transpositions and everything and if a routing is turned on in a sub-session then it always works independently and is never affected by another routing. Also, if there are areas of the keyboard range where no routings are assigned then those notes are silent and don't play on every other channel not involved with the routings. Just make sure -=END=- stays at the end of the list to stop un-routed MIDI data from reaching the racks. This behaviour seems much more like that of a workstation type keyboard to me and suits my style of programming.

Dunno if that is of any relevance to anyone or maybe it seems so basic that it's obvious to everyone, but it pleased me to find a solution and with the comment above (from another thread), I thought I'd share.

Post

I also voted against midi pass through especially in the midi routing table. Here in conjunction with subsessions and deactivating routings midi pass through is really annoying...

Not sure but maybe inverting pass through would be better to handle? Opinions??
Best regards, TiUser
...and keep on jamming...

Post

TiUser wrote:I also voted against midi pass through especially in the midi routing table. Here in conjunction with subsessions and deactivating routings midi pass through is really annoying...

Not sure but maybe inverting pass through would be better to handle? Opinions??
I think this is what having the -=END=- routing idea basically achieves. It means that only active MIDI routings will ever produce any sound and that all MIDI routings will work independently of one another [EDIT] as long as you always set passthrough ON in all other routings.

Maybe an idea for a future version might see it hard-coded so that you don't have to have it lurking at the end of the list?

Post

muzicole wrote:If I always set MIDI passthrough to ON for every MIDI routing and then just create a routing called -=END=- (or whatever) with NO passthrough at the end of the list which took all incoming MIDI data and sent it nowhere (ie. destination NONE)
Yup, that's what I did too. Maybe debugging the flow of MIDI data would be easier if we had more indicators about which filters and racks the data is flowing through?

Post

bburtin wrote:Maybe debugging the flow of MIDI data would be easier if we had more indicators about which filters and racks the data is flowing through?
Hopefully, the new MIDI indicators in 2.1 will help. I know there will be activity indicators for racks. Dunno about MIDI routings tho. It'd be great if you could have them, maybe as an option to save on CPU time.

Post

...referring to Cantabile's development blog there will be midi indicators in the rack slots only. No info yet about the racks itself - means no midi info on collapsed racks. No info about midi indicators in midi routing table entries either...

I've criticized the midi pass through in the midi routing table a while ago and know that Brad has played around with changing that but didn't roll it out in a build for compatibility issues with existing sessions - also it seems that he has found a working technical solution...

I think the midi routing table does not need explicit midi pass through at all - it should be simply active as you say. Unwanted midi events can be filtered instead... and when using subsession related routing activations the explicit need for pass through is very confusing. It's not so long that reordering midi routing entries messed up the total behavior of the table - so it is fixed already right now after I complained some time ago...

Guess I've already done what I can do, maybe you should write to Brad and tell your ideas again, multiple voices are always better than one... :oops:
Best regards, TiUser
...and keep on jamming...

Locked

Return to “Topten Software”