Hi, I believe the Hapax is sending out notes when I play them on the grid on channel #1 on a MPE track. However, the Hydrasynth doesn’t register these. If I play notes on any other channel to the Hydrasynth while in MPE mode, I can get a sound. But it seems like Hapax, by default, sends notes from the grid on channel 1. Is there any way around this? Like, in MPE mode, can I have it normally sending notes on channel 7? Does this have something to do with the UPPER and LOWER modes in MPE, and why doesn’t the Hapax have a way to distinguish between these modes?
1 Like
Its not hugely unusual for some MPE equipment to not give users choices about which zone to use, although it would be better if everything did.
Stuff that doesnt, tends to behave as if the lower zone is being used. Channel 1 in the lower zone is the ‘Manager Channel’ where MIDI messages that are not attached to specific MPE voices, ie dont offer per-note expressivity, are allowed to be sent/received. Example uses of this channel are for a global pitch bend wheel, mod wheel.
But it is also the case that instruments are supposed to listen for notes on that channel too, they are not supposed to ignore notes sent on that channel. In the official MPE specification, it is absolutely mandatory for devices to receive note on/off messages on the Manager Channel. If the Hydrasynth is ignoring these then it is failing to properly adhere to the official MPE specification.
So technically, its Hydrasynth that is at fault for the issue you are experiencing. However its probably not the only synth that behaves like this, and so it would be better if Hapax gave us some choice in this matter.
If there is no intention to change this Hapax behaviour now, then I would suggest that it would at least be a good idea to change it at the same time as other MPE features are added. For example, if we are going to be able to create and edit MPE expressive events directly from the Hapax itself in a future firmware release, then that would be a very good time to allow the ‘live mode’ of the grid to generate notes that go to the MPE voice channels rather than the Manager Channel.
Some of the development work to achieve that probably gets a bit complicated because even when Hapax is just receiving stuff from a MPE controller and recording it into a MPE track, Hapax already tries to intelligently handle MPE voice channel allocation itself, rather than just use whatever channels the MPE controller was already assigning to notes. I expect the reasons for this very much include the reasons certain DAWs behave this same way, most obviously so that MIDI record overdubbing on a MPE pattern can work with MPE data without getting into a big mess of multiple notes likely ending up on the same channel at the same moment in time, which is what would happen if the device just relied on whatever channel the MPE controller was using and a pattern was then recorded via multiple overdub takes. If/when Hapax allows fancy MPE note creation and editing directly from the Hapax, this same principal of MIDI channel allocation should apply there too, and it should really apply to the ‘live mode’ of the grid too.
There are other reasons why ideally we should have more MPE configuration choices available too, but some of these will impact on the implementation of the intelligent dynamic MPE channel allocation code. For example, even if we do not get full blown support for both MPE Lower zone and Upper zone, and we end up with something that is equivalent to just the MPE Lower Zone, it really would be best if we could at least specify the highest channel that should be used for MPE voices. Reasons why that is important for some configurations is that some synths only support enough MPE voice channels to cover the number of actual voices the synth has, eg there are a bunch of 6 voice synths by Sequential that only listen to channels 2-7 when in MPE mode. DAWs such as Ableton already give us the ability to specify MPE channel range for MPE MIDI output in order to support this scenario, and their dynamic MPE channel allocation routines work hand in hand with this option. When done right, this is another good reason for using dynamic allocation in the first place, because you can target those synths without needing to worry about what the MPE controller or existing recorded MPE patterns were using for their MPE channel range. So hopefully Squarp will be able to do a similar thing using the dynamic foundation they put into their original MPE implementation, once they add more features on top of this.
1 Like
very helpful reply, thanks. i can only hope that squarp implements a way to adjust these zones.
I suppose the temporary workaround for your scenario involves using an additional track. The additional track would not be in MPE mode, and you would use this additional track for playing via the live grid. Its been a while since I checked what sort of MPE voice channels the Hapax dynamically uses the most often for MPE tracks, but if it always starts from the lowest channel 2 and then works its way up, you should use a channel like 16 on the additional live grid track, in order to reduce the chances of notes played in this way clashing with proper MPE notes you may be playing back from recorded MPE data in the MPE track.
I have not tested this workaround so am only assuming the MPE + non-MPE tracks both targeting the same MIDI destination wont cause other issues.