No more bidirectional communication in MS v3.2.x?

Dang. Ok, thanks CCT

Actually, I've just been checking this using TouchOSC with both MS v3.1.1 and v3.2.4; I don't get any bidirectional feedback in TouchOSC with either version of MS i.e. I've programmed a couple of buttons in TouchOSC and mapped these to Mute buttons in MS. When I tap the TOSC button, it activates the Mute button in MS and this remains lit but it doesn't stay lit in TOSC, which it should if there is bidirectional communication. In fact, I'm sure I read some articles ages ago when I was researching iOS apps to use with MainStage and they all stated that there's no way to get bidirectional feedback in TOSC from MainStage...it's one-way only.

CCT, I don't suppose you happen to own MIDI Designer on iOS, do you?

Wow, OK. I'm having no problems with MIDI Designer buttons staying lit with MS 3.1.1 but that's definitely not happening with 3.2.4. For example: if I touch a mute button in MIDI Designer, the button will stay lit in both MD and MS 3.1.1 but with 3.2.4, the MD button lights momentarily but doesn't remain lit even though it stays lit in MS. I'm even using exactly the same concert in both 3.1.1 and 3.2.4, and my MIDI Designer template is the same it's been for a year. This is driving me crazy.

Mmm, interesting. I rebuilt my gig template from scratch the other day with the exception of the screen controls: I copied/pasted them from the template I've been using for years. I'll try things with a brand new concert. Thanks so much for checking things at your end for me.

Was out all day today but managed to fit in some testing before leaving home this morning. Bidirectional testing DOES seem to work after all but it's not quite the same as it is in v3.1.1 i.e. some buttons in iOS MIDI Designer that used to remain lit as per their MS counterparts now only light momentarily, or light up when other buttons are activated. I tested this by using the same MIDI Designer "concert" and using it to operate a v3.1.1 concert in MS v3.1.1 and then the same concert in v3.2.4. The communication is definitely slightly different between the two versions.

Thanks so much for replying with this info. It's nice to know I'm not the only one experiencing this!

gld282 wrote:

Update, used MIDI monitoring tool - MS does not send the message back to the controller after receiving a message. It does that only when the button is pressed from within the app

Yep, that's definitely what I'm experiencing. Bidirectional MIDI communication in MS v3.2.4 is working to a degree but it's not full bidirectional communication as per previous iterations of MainStage e.g. MS v3.1.1. MS v3.2.4 only sends MIDI messages out to external controllers if it initiates the communication. MS v3.2.4 does respond to MIDI messages from external controllers/apps and can send MIDI messages to those devices, but it's not sending feedback out to those external devices confirming that the message/s was received. However, MS v3.1.1 definitely DOES work correctly in this respect. I'm not sure if this issue started with MS v3.2.4 or with a version sometime between v3.1.1 and v3.2.4 as I haven't used the interim updates because they were too buggy.

Blueberry wrote:

3.2.4 seems the correct behavior. If it would reply to any incoming message with the same one as feedback, then motor faders will have a problem: you are trying to move them down and the feedback would actively work against you. They faders vibrate.

I'm not sure that MS echoes the MIDI message because that would indeed be counterproductive. However, I can't see how it would be natural behaviour for a status LED not to stay lit on an external device if the status LED in MS remains lit. One of the reasons to use external devices is so that you have the freedom to move away from the computer; if your external device isn't showing you the status of a control, what's the point of bidirectional MIDI commands? In my opinion, MS v3.1.1 seems the correct behaviour and given all the bugs that continue to plague the v3.2.x series of updates, this particular issue just seems to be further evidence of that. I've spent countless hours over the past 2 weeks or so trying to get v3.2.4 to replace v3.1.1 for my concerts because it does seem to handle audio better than previous iterations, but the ****** bugs are absolutely driving me nuts. I found another crashing bug in MS v3.2.4 the other day that means I can't open the GUI of a third party plugin if I want to enter Perform Mode, and the plugin dev says it's not the plugin.