Various session box issues

Submitted by johnqpublic on Sun, 09/10/2017 - 14:05
johnqpublic
pro user
Hi I'm trying to add an NI F1 to my setup. I previously had two XoneK2s, but I'm wanting to try out the F1. I'm also updating some settings now that I don't have to provide scene stop buttons in the session box setup. Or I can provide ones on a different controller if I want. My aim is to use the F1 for transport and device control and the XoneK2 to continue as my track and clip controller. The first problem I encountered is that when I deleted the values from the scenes and track stop editors, they weren't all removed from the script. I think the scene stop ones remained. So I couldn't use those control values for anything else until I manually modified the script and then imported it. The second problem I encountered is that I can't set up session boxes in combination. The app complained that I have to set the session boxes in different modes. But when I did that the first box stopped working. My session box navigation controls either had no effect or, those that shared a mode with the second session box moved the first one, but multiple steps rather than the single steps I'd set up. The third problem is related to the second - after deleting the second session box to try to get the first one working again, the session box was not removed from the script, so it still appeared and the navigation of the first session box still didn't work. So how are combination boxes supposed to be set up?

15 Responses

Comments

johnqpublic
pro user

I'm now getting told that a control I haven't used in the session box is used in the session box. The session box has CC32 on channel 15 (the XoneK2) for one of the track launch buttons. I had the transport stop already assigned to CC32 on channel 13 (the F1). The transport play, stop, etc., set up on the F1, no longer work. When I opened the stop in the editor and simply tried to save it without making changes I got an error stating that I'd already used the setting combination already. It turned out this one is because the app had the midi controller values set to have channel 13, even though the script had 15 as the channel on the session box.

There are other weird clashes going on. The track stop buttons don't work after changing them back to the XoneK2 in the session box settings (the JSON kept the channel set to 13 even though I'd used midi learn to set channel 15). The track 1 arm button doesn't work at all (I've checked the values). And the LEDs on display don't match the settings - the arm buttons all light up when I shift to mode 2 and back to mode 1. An LED in the top row randomly turns on or off when I shift to 2 and back to 1 as well.

It took hours to get the initial setup working. With the recent changes, trying to get LEDs to work as expected again and trying to add the F1 has cost most of the last two days and I'm no closer to having working scripts. Is there perhaps some way to determine whether settings will work and, when they don't, why not?

John Carney
pro user

Hi there,

Sounds like you have a lot going on there :)
Could you upload the json file for one or both of your scripts so I can check.

John

johnqpublic
pro user

I'm attaching the latest attempt with the K2 and Twister. Various bits have started and stopped working, seemingly randomly. It's difficult to tell what is changing each time the scripts generate because bits move around even when no changes are made. The script has also behave inconsistently in Ableton. The python generated from this JSON worked, at least for the K2. I put in a previous version and then switched back to this one and the K2 controls no longer worked properly - moving the session box up and down no longer worked and when shifting to mode 2 the session box changed colour, indicating there were 2 boxes when there'd previously only been 1.

upload files: 
johnqpublic
pro user

Here's an example with the XoneK2 and the F1. Here the session left and right work, but not up and down. The device bank up and down are doing something - the LEDs go on and off. The transport is working too. The only thing obviously not working is the session box up and down.

upload files: 
johnqpublic
pro user

And this one still has no working up and down on the session box, but now the device bank buttons also do nothing.

upload files: 
John Carney
pro user

Cool, will take a look at all of them and get back to you.

John

John Carney
pro user

For all 3 scripts you have the session navigation on mode 2 but the session box is on mode 1 which is why you can't move it up / down.
If you want the up / down controls on mode 2, duplicate the session box and put the copy on mode 2.
See the attached image.

John

John Carney
pro user

attached here

upload files: 
johnqpublic
pro user

It used to work. The only thing I changed was the LED settings in the hope that for some reason that had caused the problem. And in intermediate versions with the twister the box did move up and down with the modes set as they are. There's evidently some odd inconsistency in the script generation. Unfortunately it's difficult to track down because it's not generated deterministically. That other controls have randomly started and stopped working (like the device bank and transport buttons) doesn't bode well either. I haven't kept all the json and script versions, but it seems unlikely that having them would be useful.

johnqpublic
pro user

I'm also curious to know what would generate two session boxes when only one is configured in the json. Sometimes when I run Ableton I see two session boxes (the second only appears when I use one of the shift buttons I've set up).

johnqpublic
pro user

OK, I've partially figured out the previously working session box (other than the one case where it worked yesterday with most of the most recent k2andf102 scripts) - I modified the python, creating a single set of session box settings used to create a session box in each mode activated so that the box position, colour, etc., persisted across modes, but could be moved differently in different modes. I did remember that I'd made changes, but I wasn't sure how much was functional and how much was just cleaning up the code. Probably the best use of my time now is to do similar cleaning up in the python and then figuring out how the bits fit together so I can manage the script manually. Otherwise any changes I make in remotify will require redoing all the manual changes.

John Carney
pro user

Yes, custom coding in the script can lead to unexpected behaviour :p

FYI, the way code is generated by Remotify, its not possible for 2 session boxes to be loaded at the same time (in the same script).

If you want a session box to persist across multiple modes, set it to mode 0. This is the "global" mode which always remains active alongside another mode (modes 1 through to 16). So you can set controls on different modes to control a single instance of session box, which is what it appears you're trying to achieve.

Hope this helps.

John

johnqpublic
pro user

I did try mode 0 for the session box, but that didn't fix the problem, unfortunately.

John Carney
pro user

Not sure why that is, just make sure you only have one session box in your script if you set it to mode 0.

John

ananaphone
pro user

@johnqpublic
Regarding the arm buttons: If you can edit the parameters in your midi controller, try setting the on value of a toggle button to send 127 and the off value to 126.

-A