Hi there, I am an incredibly happy owner of the new motorised knobs device called Roto-Control by Melbourne Instruments. The haptic feedback in it makes it a real pleasure to use, its like having reconfigurable, bi-directional pots instead of endless encoders. You can even change their behaviour so that they feel like they have up to 16 stepped values, with clicks you can really feel, instead of one smooth pot sweep.
In standard MIDI mode (as opposed to two Ableton modes that it also supports), Roto-Control is already working with the ipad, but there is one obvious missing piece to this picture. An ipad app that can act as an auv3 host and that has its own bi-directional MIDI mapping which works not just for its own controls, but also for any auv3 plugin parameters that the plugins expose to the host app.
Drambo is very much on my radar in this regards, so I purchased it today. Unfortunately someone on another forum made it sound like the new feature being discussed here was already available, and it was only upon finding this thread that I discovered the reality. Never mind. I would be happy to test the beta with the Roto-Control if that is an option, if not I shall just have to be patient and wait.
I have to apologize first. Since I'm deep in testing and reviewing, I have to admit that I've lost track what feature set has made it into the Appstore version and which features didn't. The difference has become quite substantial in some areas, but it's in the nature of substantial features that they need a lot of testing and a lot of bug fixes. I have more than 10 different MIDI controllers here, some with and some without feedback, and with a wide selection of supported and unsupported MIDI messages. One of them was, by the way, that most controllers support Note Off messages in the form of Note On with velocity = 0, but one controller made by Arturia only switched off its LEDs when receiving an actual Note Off message (with any velocity)...
So I have a simple question for you: What kind of MIDI messages does the Roto-Control expect for setting knob positions in feedback?
The same message type as the knob sends? In that case, it will be easy to map.
Thanks, and dont worry about the confusion, but thanks very much for the apology anyway.
Yes the Roto-Control has a nice simple implementation, it responds to the same messages that a knob sends.
The knobs (and buttons) send and receive cc's by default. You can tell it to use 14 bit cc's (pairs of cc's) or NRPNs instead of single cc's, but I am used to working with apps that dont necessarily support those. And current firmware has some bugs when receiving NRPNs so I wouldnt use those at this stage anyway, and even when fixed I'd likely only use those for direct communication with synths that use NRPNs. In any case, whatever it sends is what it expects to receive back for that control.
edited to add extra detail - everything sends/receives stuff just like absolute value pots do, rather than like relative encoders.
Yes, this seems to be one of the best concepts. Relative encoders are a fairly good workaround to avoid value jumps but knobs and faders that jump to the real value are certainly the better option.
Here's what you can already do now, even before MIDI feedback is officially released:
You would MIDI map the KNOBS, choose your Roto-Control as MIDI output and also use the KNOBS to control parameters on your rack, anything to the right of these modules.
There will naturally be a feedback loop that might give you a slightly strange feeling when turning knobs (same with motorized faders), and I'd e happy to hear your experiences in this regard.
Thanks very much. I was able to understand the logic of your screenshot & instructions and I got it working, even though I am new to Drambo.
In regards whether the feedback loop makes the knobs on the Roto-Control feel weird, I need to do some more testing. Because sometimes it does, and sometimes it doesnt. Im not sure if this is because the Roto-Control itself tries to take care of this potential issue, eg whether it stops this unwanted feeling based either on whether you are touching the knob at the time (as each knob can detect fingers touching it) or whether its based on timing of incoming messages vs actual knob position at that moment in time. I cannot currently say whether my fluctuating experience in this respect is down to the Roto-Control itself not always managing to detect and compensate for this 'tug of war' or whether its down to fluctuating timing of MIDI messages that might be improved on the Drambo side of things when the proper feature is available. When I've done more testing I will also compare this experience to the one I've had when using Roto-Control with other apps, both on iPad and with various stuff on a desktop computer.
Yes, that's exactly the point. As long as you're only using MIDI in one direction (either Sending CC values OR positioning the knob from received CC values), all seems good.
When using MIDI feedback on a motorized knob or fader though, that control needs to synchronize motorized position and sending value precisely. Message delays, tiny offsets and intermediate knob positions can all cause audible noise or a knob that seems to resist turning (or a fader to resist moving).
I wouldn't worry too much about it, I just mentioned it because it's kind of expected behavior to some extent.
Could you be a bit more specific please? What kind of functionality, workflow and MIDI feedback do you have in mind?
Does the Oxi One expect some specific message format? Are the feedback messages different from the sent messages?
MIDI mapping will be as generic as possible. Supporting specific models is too much work but of course, the mapping engine should be generic enough to create mappings for most MIDI controllers out there. I already have about 15 different controllers here but that's still only a fraction of what is and was available on the market.
Will it be be possible to send this type of sysex messages? (I know it is not exactly what you’ve asked, and it is not really about bidirectional midi)
@ndnr Sending sysex for initialising the controller will be supported, but from the spec it looks like every LED will need its own sysex message to be controlled, which will require a 3rd party app like Streambyter for now. You'd use it to convert MIDI note numbers (as received from the grid buttons and sent by Drambo in return) to the Oxi's proprietary sysex messages for controlling the grid LEDs.
Ok. Sounds interesting! I am looking forward to see your result and see if i can create some seamless integration where i can just use the controller like i would like to :) Thanks for your effort to make Drambo better!
Oh maaaaaan was hoping this would be coming soon the video they posted on YouTube got me excited. Want to pull the trigger on a launchpad but kind of holding off until this comes out.
As mentioned, there is no deadline. It might be weeks, it might be months. Based on the previous experience with the developer, I would expect at most few months from today for next release.
Of course, he may always decide to omit the controller support if it would still feel unstable.
So basically nothing is guaranteed. If you want to buy Launchpad, just buy it and have fun with other apps in the meantime, there are already quite some iOS music apps supporting it.
Comments
Hi there, I am an incredibly happy owner of the new motorised knobs device called Roto-Control by Melbourne Instruments. The haptic feedback in it makes it a real pleasure to use, its like having reconfigurable, bi-directional pots instead of endless encoders. You can even change their behaviour so that they feel like they have up to 16 stepped values, with clicks you can really feel, instead of one smooth pot sweep.
In standard MIDI mode (as opposed to two Ableton modes that it also supports), Roto-Control is already working with the ipad, but there is one obvious missing piece to this picture. An ipad app that can act as an auv3 host and that has its own bi-directional MIDI mapping which works not just for its own controls, but also for any auv3 plugin parameters that the plugins expose to the host app.
Drambo is very much on my radar in this regards, so I purchased it today. Unfortunately someone on another forum made it sound like the new feature being discussed here was already available, and it was only upon finding this thread that I discovered the reality. Never mind. I would be happy to test the beta with the Roto-Control if that is an option, if not I shall just have to be patient and wait.
Thanks for making this happen!
Hi and welcome on this forum Steve,
I have to apologize first. Since I'm deep in testing and reviewing, I have to admit that I've lost track what feature set has made it into the Appstore version and which features didn't. The difference has become quite substantial in some areas, but it's in the nature of substantial features that they need a lot of testing and a lot of bug fixes. I have more than 10 different MIDI controllers here, some with and some without feedback, and with a wide selection of supported and unsupported MIDI messages. One of them was, by the way, that most controllers support Note Off messages in the form of Note On with velocity = 0, but one controller made by Arturia only switched off its LEDs when receiving an actual Note Off message (with any velocity)...
So I have a simple question for you: What kind of MIDI messages does the Roto-Control expect for setting knob positions in feedback?
The same message type as the knob sends? In that case, it will be easy to map.
Thanks, and dont worry about the confusion, but thanks very much for the apology anyway.
Yes the Roto-Control has a nice simple implementation, it responds to the same messages that a knob sends.
The knobs (and buttons) send and receive cc's by default. You can tell it to use 14 bit cc's (pairs of cc's) or NRPNs instead of single cc's, but I am used to working with apps that dont necessarily support those. And current firmware has some bugs when receiving NRPNs so I wouldnt use those at this stage anyway, and even when fixed I'd likely only use those for direct communication with synths that use NRPNs. In any case, whatever it sends is what it expects to receive back for that control.
edited to add extra detail - everything sends/receives stuff just like absolute value pots do, rather than like relative encoders.
Yes, this seems to be one of the best concepts. Relative encoders are a fairly good workaround to avoid value jumps but knobs and faders that jump to the real value are certainly the better option.
Here's what you can already do now, even before MIDI feedback is officially released:
You would MIDI map the KNOBS, choose your Roto-Control as MIDI output and also use the KNOBS to control parameters on your rack, anything to the right of these modules.
There will naturally be a feedback loop that might give you a slightly strange feeling when turning knobs (same with motorized faders), and I'd e happy to hear your experiences in this regard.
Thanks very much. I was able to understand the logic of your screenshot & instructions and I got it working, even though I am new to Drambo.
In regards whether the feedback loop makes the knobs on the Roto-Control feel weird, I need to do some more testing. Because sometimes it does, and sometimes it doesnt. Im not sure if this is because the Roto-Control itself tries to take care of this potential issue, eg whether it stops this unwanted feeling based either on whether you are touching the knob at the time (as each knob can detect fingers touching it) or whether its based on timing of incoming messages vs actual knob position at that moment in time. I cannot currently say whether my fluctuating experience in this respect is down to the Roto-Control itself not always managing to detect and compensate for this 'tug of war' or whether its down to fluctuating timing of MIDI messages that might be improved on the Drambo side of things when the proper feature is available. When I've done more testing I will also compare this experience to the one I've had when using Roto-Control with other apps, both on iPad and with various stuff on a desktop computer.
Great to hear @SteveElbows!
Yes, that's exactly the point. As long as you're only using MIDI in one direction (either Sending CC values OR positioning the knob from received CC values), all seems good.
When using MIDI feedback on a motorized knob or fader though, that control needs to synchronize motorized position and sending value precisely. Message delays, tiny offsets and intermediate knob positions can all cause audible noise or a knob that seems to resist turning (or a fader to resist moving).
I wouldn't worry too much about it, I just mentioned it because it's kind of expected behavior to some extent.
I guess you don’t know when the testing will be over. But could you say if the Oxi One is one of those controllers?
Could you be a bit more specific please? What kind of functionality, workflow and MIDI feedback do you have in mind?
Does the Oxi One expect some specific message format? Are the feedback messages different from the sent messages?
MIDI mapping will be as generic as possible. Supporting specific models is too much work but of course, the mapping engine should be generic enough to create mappings for most MIDI controllers out there. I already have about 15 different controllers here but that's still only a fraction of what is and was available on the market.
I would like to implement a DAW (resp. Drambo) controller using it. This is what I’ve found yesterday in the documentation:
Will it be be possible to send this type of sysex messages? (I know it is not exactly what you’ve asked, and it is not really about bidirectional midi)
@ndnr Sending sysex for initialising the controller will be supported, but from the spec it looks like every LED will need its own sysex message to be controlled, which will require a 3rd party app like Streambyter for now. You'd use it to convert MIDI note numbers (as received from the grid buttons and sent by Drambo in return) to the Oxi's proprietary sysex messages for controlling the grid LEDs.
Ok. Sounds interesting! I am looking forward to see your result and see if i can create some seamless integration where i can just use the controller like i would like to :) Thanks for your effort to make Drambo better!
Does anyone know when this update will be released? Some kind of time scale?
I am afraid not even the almighty developer himself knows :)
There is some other big feature expected in the next release so I think this is currently swallowing the most time.
Oh maaaaaan was hoping this would be coming soon the video they posted on YouTube got me excited. Want to pull the trigger on a launchpad but kind of holding off until this comes out.
When is the next release you speak off?
As mentioned, there is no deadline. It might be weeks, it might be months. Based on the previous experience with the developer, I would expect at most few months from today for next release.
Of course, he may always decide to omit the controller support if it would still feel unstable.
So basically nothing is guaranteed. If you want to buy Launchpad, just buy it and have fun with other apps in the meantime, there are already quite some iOS music apps supporting it.