Ke10g

A lot of us are using single cycle waveforms, so it might be worth updating the sample audition function (when browsing samples) so that they audition in loop mode or loop a certain number of times if they are shorter than a certain length. This would be in addition to the current functionality which automatically sets these shorter samples to loop mode once they are loaded. Right now it is very difficult to know what you are going to get from auditioning a single cycle sample: it would be really helpful to hear the actual timbral/harmonic characteristics of the sample if it looped while auditioning.

Dear Jamie and fellow Downrushers,

I've been toying with MPE (MIDI polyphonic expression) control of the deluge. The Deluge is capable of MPE, it just is very tedious to implement. I'm wondering if you, or anyone else listening, might be able to help. I'm not much of a programmer and I feel it would be fairly straight forward to streamline the process with an XML editor. Given that there are very few hardware MPE synths out there, there is a great incentive to get this going. The basic idea for using the current Deluge as an MPE synth is that you use each track as a synth voice. So you basically have the same synth preset copied onto as many tracks as voices you want, and map each track to a different midi channel. It would be great to have an editor for this, as it is REALLY tedious to set up. Ideally, I think, it should be added as an extra feature of Downrush in the form of an enhanced song editor.

Here is my process for doing MPE with Deluge (I use a couple Roli Seabord Blocks as MPE controllers, so the CCs mentioned here correspond to those, but the process would be similar with different instruments):

-make a synth sound you like
-map aftertouch to some parameter
-copy the preset to as many tracks as voices you want
-on each track, learn the midi note channel for that voice (for 8 voices I use channels 2-9)
-on each track, learn the 'Y' (vertical slide) parameter (cc 74 for each channel, by default on seaboard)
--- the other "dimensions" of the 5D controller are "automatically" mapped: pitch bend and aftertouch and note-number-to-pitch.
--- you can then do global modifications of your synth patch for all voices by modulating the knobs on the song level.
------I use Max MSP to "learn" the midi in channels and the cc74 Y parameter to each of the tracks... in any case you need a device that allows you to send a single message precisely to the Deluge.

I've toyed with modifying the song XML directly to do this, but it is still a major pain.

I'm thinking of an interface similar to Douwnrush kit editor for Songs that allows you to:
-select an existing preset and apply it to different tracks
-assign midi in channel number to those tracks
-map aftertouch, velocity, and cc74 to the parameters to each of those tracks/synths
---One issue is that while aftertouch, pitch bend, and note number, is handled by Deluge at the "track" level, CCs are handled at the "sound" level, in the Deluge file format.

A more sophisticated MPE-oriented version of this might allow you to select the number of voices you want, choose a midi start channel, and select your "patch cable" connections (mappings), and the program would automatically create a Song with the appropriately configured tracks for use as an MPE synth.

What do you think?

Thanks for your time!

MPC LIVE and MPC X now have this feature. It would be great to have it on Deluge.

the ability to get a separate 'click/metronome' output - perhaps through a c/v jack or something?

This is not important to me.

I just want a simple way of arming record for sampling of the input (not the master out) in a kit track.
All the elements are in place already, the Synthstrom team would only need to add one simple interface element:

Right now, I can create a new kit and press record while the Deluge is playing and record whatever I feed through the input. The only problem is that I have to time it just right, ie: press record at the beginning of the pattern and press again at the end of the pattern, because it starts recording as soon as you press the button, rather than arming and starting only at the next section start point (and stopping at the end).

Here is my idea for a quick fix to the issue:

-When recording a new sample into a kit, by first pressing the kit audition pad and the load button, then record:
--If deluge is not currently playing, recording starts immediately, and stops when you press record again, as it does now.
--If deluge is playing, the recording is armed, and starts only the next time the playhead reaches the beginning of the pattern, and stops again at the end of the pattern, unless the user presses stop before that point.

@muleskinner said:
It's probably worth pointing out that before we have a looper (which I would really, really like) - we're going to need the ability to actually loop

What you are describing would definitely be useful, but this is a different issue. We certainly do not "need" the feature you are describing in order to have basic quantized recording of the input. All we need is for the recording on and off points be quantized to the pattern beginning and end points.