I’m having an issue with midi remote feedback in Ableton live and wondering if anyone using the E1 has had a similar problem. I considered posting a message on the Ableton forum, but it’s such a mess, and the bug report system will take a few months to give me an answer (if any…). So I’ll try my luck here !
I have a pretty complex Ableton project to play live. And I use a few midi controllers connected to a Macbook pro.
Midi in/out is setup to receive remote midi feedback and reflect the state of the Ableton project on the E1 as soon as I load my Ableton project. I’ve been using this system for a few years (with the E1 for over a year).
But I’ve discovered these last few days that certain controls on the E1 don’t update anymore when I load the Ableton project…
I used “Midi monitor” to inspect the midi data being dumped to the E1 when I load the Ableton project. And apparently Live is missing certain controls. I can force a new dump by entering midi learn mode, and when coming out of midi learn, Ableton sends a new dump of the controls to the connected midi controllers. It will update some controls that it missed the first time, but even if I force 10 dumps, I’ll never get all the E1 controls to fully reflect the state of the Ableton project.
By looking closer at the dump, I discovered that Ableton is sending certain control messages many times instead of the missing ones.
All this is very random, it’s not one specific CC being a problem. And on a new fresh project they seem to work nicely. I’m wondering if it’s a case of having built too much on the same Ableton project file and having issues because of that ? Any ideas ? Anybody had a similar issue ?
I have to say, that the E1 is a GREAT controller, but not having a reliable total recall system is really problematic. Especially with over 400 controls mapped in a project !
When developing my Ableton Live remote script I ran into similar problems. It appears the E1 cannot reliably handle too many parameter updates in quick succession.
Hum… Too bad.
I suppose you got around that issue in your script by delaying the messages ?
I’m not sure what to do about it. It really defeats the purpose if I can’t use total recall.
Thanks for your feedback !
After your feedback, I decided to downgrade to firmware 2.16. And the E1 seems to be having an easier time dealing with the dump sent by Ableton. I can even see on the screen that the faders update much faster.
I still seem to have another problem with a few parameters simply not being sent by Ableton. Probably two separate problems, which always makes debugging harder !
Just an update from my side. I am working on this using the input from @jhh and now from you as well
Thanks Martin !
I’ll stay on version 2.16 for the moment.
I did add small delays yes, order of 10ms (so for large patches this creates a quite noticeable delay upto several seconds). Also, there are commands to stop and reenable display updates on the E1 which make it respond faster.
I finally got around the problem. I had two issues combined, which made debugging more complicated.
Something went wrong with the Ableton file and a series of knobs inside a macro we’re sending the wrong midi CC during the dump. Basically, the first knob in the macro was sending its value four times, instead of sending four different values for each four knobs mapped in the macro (so the last three knobs were not getting updated). I’m not sure what happened, but I had to recreate the macro from scratch to get it working properly again.
The firmware 3.05 was also part of the problem. The E1 running version 3.05 couldn’t cope with the amount of midi data sent during the dump and the faders didn’t update properly. All is fine on version 2.16 and Martin is aware of the issue and working on it.
Hope this can be useful to others in the future.
@Jesse perfect! I am already testing an updated version of the firmware. The handling data bursts seems to be resolved there. I also gave a try to @jhh’s Ableton preset with delays removed. That worked fine too. More on that later today.
ok, it was confirmed by @Jesse and @jhh that the upcoming firmware release resolves this issue.