Features request for next firmware versions...
  • No I won’t listen to any new feature request for v0.92. Let’s test and deliver it!

    Every new feature request will be considered for v0.93

  • easier random function.
  • How can this be made “easier”? A description of the requested behaviour would be helpful!

  • Okay, maybe i am getting it wrong, if i'm in Browse mode (or combo) and hit the random function everything works as wanted, but i can't save the actual
    generated patch. In order to achieve that, i need to go to save mode first and then do randomization then entering the name. that's a bit tricky.

    For the feature requests:

    - controllable functions by the controller, e.g. MMC (midi machine control), e.g. start and stop of sequencer, midi clock follow (let's say if you stop the sequencer
    and want to restart-it, then hitting the button would restart the midi at the next main step)
    - Joystick controller (or documentation thereof) for one or multiple joysticks.
  • Can’t the joystick controller be done with MIDI?

  • well. probably with some addon PCB, yes. i thought if that's similar doable as with the programmer.
  • I wrote this in another post but here we go. I would
    like the filter envelope amount to be included in the
    modulation matrix. Sometimes (quite often actually)
    it feels more natural and sounds better to modulate
    that instead of the filter cutoff.
  • Longer envelope stage times please. It's okay at the moment while it's mono but for poly-Shruthis 6-8 seconds for A,D or R isn't enough!
  • Note gate as a modulation destination (over 63-> on, under 63-> off)
    Sample & Hold as an modulation operation
    Playing with the chaos, these would open all kinds of more melodic avenues

    a third envelope… with a DAHDSR curve… that can cycle like an LFO... with adjustable slope shapes for both rise and fall :)
    I realize this would break preset storage, but it would be all kinds of awesome.

  • Actually, I should’ve said Sample & Hold as a modulation destination, since the lfos laready have a s&h waveform – which samples the noise generator, I guess? Also, I’d like BPM as a modulation destination. I’d really like to have an LFO shift the arp/seq speed between the MIDI clock multipliers. :D.

    I guess those features are a bit extreme, but all the basics are completely covered already.

    Also, I should’ve said “please”. And “thanks!”. Because you rock!

  • +1 for longer envelope times…i like sounds that take time to evolve :)

  • I would overhaul the way that MIDI out works for chaining purposes. If we could generate MIDI clock from the master device, slaves could set BPM to ext and arp/seq modes would be sync’d. I wonder if that’s too CPU intensive. I also realize that it would be problematic if slaves down the line are generating their own clock (not set to EXTERNAL) – how do you know whether to forward the clock received or the clock generated?

    Also, in the standard chaining mode, it makes sense to send all parameter tweaks (and the entire patch, on load) over the MIDI out so that the slaves follow the master. In split mode, this is not desirable – it’s usually going to be the case (I think) that there are different patches for each split. MIDI clock would be useful here as well, but what I don’t know is whether it would be advantageous to send program change messages instead of the sysex patch dump. In that case, you could have different patches set up for e.g. shruthi #1 program 1, and shruthi #2’s program 1. When you change the master to program 1, the slave’s program 1 automatically loads, but they’re different patches.

    As it currently stands, you’d have to change your program on the master, and then change the program on the slave. The benefit I guess is that you don’t have to worry about keeping patches stored in the right place so that the split configuration is always as you want it. Fortunately with larger patch storage comes greater freedom with this kind of thing.

    It might work better to have a slave ignore CC/NRPN messages rather than stripping them from the MIDI out of the master. I think this would work better because then messages can still be forwarded down the chain.

    I know that currently the MIDI implementation is as elegant as it can be for polychaining, which is already a complex feat. This would make things a little more difficult to comprehend but I think it would make chaining more useful.

  • @ pichenettes & janavolt re joystick

    Wavestation implements this a 2 CCs +/- 64 in value for the xy. I simpy use my novation remote touchpad to control it instead of a joystick and it works great..

  • Hmm, i’m not sure but it seems like patch/program changes already work this way? It seems to me now after playing with this a bit more that program changes are sent in split mode, and the entire patch is sent while polychaining… But I haven’t taken the time to hook it up to a midi monitor to see what is sent, and when.

    Are there more details about how this is designed to work somewhere? The manual is a little slim on info in that arena.

  • @smrl: I plan to do a complete re-check of the MIDI out modes this week-end, I’ll report here how it works, maybe there’s a weird corner case I have missed somewhere…

  • are we already able to make sequence only dumps?
    this I'd like

    for random function: I'd like first to press the two knobs to activate.
    But then, as you never only press only once :-D
    I'd like to be able to continue the random thing via the main encoder!
    pressing again a button takes me back into normal mode to tweak and so one...

    I'd like to see the "performance" parameter names and values in the load menue at least or only by tweaking

    more modulation slots!!!!!!
  • > I’d like to see the “performance” parameter names and values in the load menue at least or only by tweaking

    It’s a good idea, and it’s trivial enough to be added to the v0.92 before I burn the chips! Done.

    > more modulation slots!!!!!!

    I’ve reached the point where I cannot add any new synthesis parameter since the patch data structure is entirely filled – If I want to do so, for example to add new modulation slots, I will break the compatibility with the current patch format. Wait for my next synth :D

  • I haven't seen this mentioned anywhere but I may be wrong: I have been playing with the arpeggiator a lot recently and was wondering how easy it would be, or if it would even be possible, to have the arpeggio restart from the beginning of the sequence in latch mode when the MIDI notes/keys are re-triggered? i.e, re-triggering the chord notes or playing a different chord in latch mode would restart the arp/seq clock. This would be especially good for live performance when the arpeggio clock does not quite match the drum timing - you can get a better approximation to 'loose' drumming or an unknown BPM by playing the chord at the beginning of each bar which restarts the clock from the first beat.

    Also, how about an 'additive' latched arpeggiator mode whereby you play a chord, release the keys and then add other notes to it in an incremental or aggregated fashion (though this may require more than 7 notes in an arpeggio).

    .....not sure I'm making much sense..... burble, burble, glug
  • > Also, how about an ‘additive’ latched arpeggiator mode whereby you play a chord, release the keys and then add other notes to it in an incremental or aggregated fashion (though this may require more than 7 notes in an arpeggio).

    But then, how do you tell the arpeggiator that “this note should be added to the chord” vs “this note is the beginning of a new arpeggio”?

    > to have the arpeggio restart from the beginning of the sequence in latch mode when the MIDI notes/keys are re-triggered?

    Same question as above… What if the arpeggiator is arpeggiating C, E, G ; and you play C, E, A to change the chord. Should it retrigger or not?

  • >> Also, how about an ‘additive’ latched arpeggiator mode whereby you play a chord, release the keys and then add other notes to it in an incremental or aggregated fashion (though this may require more than 7 notes in an arpeggio).

    >But then, how do you tell the arpeggiator that “this note should be added to the chord” vs “this note is the beginning of a new arpeggio”?

    I admit that hair-brained idea was a bit of an afterthought!

    >> to have the arpeggio restart from the beginning of the sequence in latch mode when the MIDI notes/keys are re-triggered?

    >Same question as above… What if the arpeggiator is arpeggiating C, E, G ; and you play C, E, A to change the chord. Should it retrigger or not?

    My idea was that every time you play a new chord, even if identical notes are played, the clock is re-triggered. I suppose this means that even if you play one note, as soon as you let go and play another single note, the clock is re-triggered. So yes, playing C,E,G and then C,E,A would re-trigger the clock, as long as all keys were released in between the chords. If you played C,E,G, then let go of G, keeping C and E down and then played A, this would not re-trigger the clock. This would allow for the player to bring the arpeggiator into sync with the beat of the music (perhaps the drums) at the beginning of each bar (or as many times as the player wants to re-trigger the clock). This would be particularly useful if you had no way of MIDI syncing to the rest of the music (analogue drum machines and real drummers in a live situation). I imagined an arp+lat+rtg mode but there is probably no room left in the firmware?

    If I am the only one who thinks it is a good idea then never mind - I'll give the drummer a kick every now and again ;-)
  • Please, please create an input filter for midi.
    In studio, there's no problem, I can filter out all unwanted midi-events.
    But live, I play the Shruti by a VA synth via midi to make the VA more valuable ;-), stacking it or even playing the Shruti on the keyboard.
    The VA has a lot of pots which send many controller events. e.g. the cutoff pot sends cc101, cc102 and cc103, one for all of the three synth sections.
    You can imagine what happens to the Shruti.

    So it would be very great to have different switches to filter CC (all or above cc10, except cc64 of course or so), AT and so on.

    Also it would be usefull, to do a note priority switch for incoming chords, e.g. lowest note, highest note.
  • How about modulation destinations for filter 1/2 decay time? That would let you do 303-style accented bass notes by making louder notes have shorter envelopes.


  • I think it would be great if the shruthi would diplay changes made via nrpn the same way as it displays changes made via knob.

  • @Manuel: not sure that would work very well. What about several simultaneous changes being made from a MIDI controller or sequencer? For example, using the XStation template I made, I can change filter and envelope parameters at the same time. To display these, the Shruthi display would have to switch between several different pages, which wouldn’t be useful.


  • Yes, this has been discussed before. It wouldn’t work well and would be very disturbing when several values are changed at the same time.

  • I think the way its implemented currently is the best compromise: if you happen to be on the particular page on the Shruthi showing the parameter being edited remotely, you’ll see the value changing.

    Incidentally, pichenettes, did you consider making all the knobs endless encoders, rather than pots with a limited angle of travel. This is probably a personal thing, but I’ve always thought that relative operation with endless encoders made more sense with physical controls that have more than one function. Having said that, I haven’t found the fact that the position of the controls initially bears no relation to he value it’s controlling too annoying in practice.


  • I don’t see the difference to turning two potis at the same time. Sure, it would display garbage if several values are changed at the same time, but it would be easier to finetune a single value using a midi keyboard.

  • If you need to fine-tune, you can always us the controls on the Shruthi itself, surely.


  • @toneburst: I think there’s more “immediacy” with a pot — I just don’t like the feel of rotary encoders. Actually, if they were high-res (say 120 steps per rotation) and clickless, this would be OK, but such parts are not widely available. I could have had some custom-made to those specs, but this would have been quite expensive. From another thread, it looks like people are already having trouble getting a part with the most common specs (24 clicks/turn)!

    @manuel: One use case in which displaying NRPN change would be very annoying and disturbing would be overdubbing the recording of knob changes. You record the NRPN produced by knob changes in a track, then you replay the same sequence and record a change to another knob. But if this really annoys you, this can surely be easily fixed in the firmware.

  • @pichenettes you can relatively easily modify some kinds of encoders to remove the clicks. I’ve got a 95%-completed MIDIbox MB-6582 (long story) that uses encoders modified in this way. Those encoders are much larger though. Not sure if the same trick can be applied to the much smaller units used in the Shruthi.


  • HI,

    would it be possible to have an MIDI CV OUT - GATE OUT ?
    Maybe using a pin free atmega..
    thanks :)
  • There’s no free pin on the ATMega :) And even if there was, you wouldn’t got very far with it. To get a CV/Gate out compliant with what analog synths accept, you’ll need at least a 12 bits DAC to get an accurate pitch CV, and a 9V supply circuit to generate the large pulses and voltage accepted by analog synths.

  • I did not say anything *whistles*
  • *whistles too

  • I would like to suggest the possibility to save the arp settings either with the patch or as an independent sequence. Unless I am missing something, which is perfectly possible, it seems a lost opportunity having such a great arpeggiator and so many variation possibilities in just 4 parameters and being unable to fine tune and then save them for repeatability. It would also be nice to have more midi clock sync subdivisions in the LFO's speed parameter.
  • New Shruthi user here, so I may have missed something, but I would really love to see a down–ramp waveform for the LFOs. I use that way more then the up–ramp wave that's there now.
  • cbm, just set a negative value for the source lfo in the modulation matrix, that turns the up into a down ramp.

  • How would that work for the LFO2 -> filter routing?
  • in the modulation matrix something like this:
    1.. lfo2… cut… -63 (negative value for ramp down)

    with ramp as waveform (lfo2 page)
    and because lfo2 is set to cutoff by default you need to set this to zero in the filter page so the settings in the mod. matrix have effect.

  • @aksmaster. Unfortunately if I decide to save more stuff in the patch, I’ll have to revise the patch data structure and this will break everybody’s patch library ; so I’m reluctant to do that. I won’t do it unless it’s for a very major upgrade.

    @cbm. rosch’s solution is the right one. The env / lfo modulation settings on the filter page are just hardwired shortcut – easy, accessible defaults like on the MS20 or ARP2600 :) Note that in early versions of the firmware, the env and lfo settings on the filter page were in the -63..63 range. But it’s easier to kill the modulation when the modulation range goes from 0 to 63 (turn the pot fully counterclockwise), rather than when it goes from -63 to 63 (very precisely center the pot).

  • @pichenettes. I understand your reluctance about changing the patch data structure, then, would it be feasible to sacrifice some sequence patch memory to save some arp setups, or save the "arp" page within the sequence patch? It is just that the simple interaction of dir/rng/pat/wrp yields some very specific musical results which, especially live, are difficult to set up with speed and precision even using handwritten notes.
  • here is a complicated feature request that just donned on me

    a vocoder to go with the digital FX board

  • @askmaster: yes, it would be feasible to sacrifice some sequencer memory space for arp presets.

    @n8bit: absolutely not possible given that the digital FX can only process one input. And not mentioning the ridiculously low processing power available on an AVR.

  • It would be possible to have a vocoder with a single audio input, if the carrier waveform was hardwired to be the osc output from the control board. Doesn't solve the processor problem though, obviously...

  • @toneburst: this doesn’t work. The audio input is mixed with the oscillators signal and sent into the AVR for processing. The AVR only sees one signal, the mix of the oscillators and the audio input.

  • patch the sequencer in to transpose the arpegiator.

  • Or maybe the other way round, direct the midi flow of up to x notes into a sequencer bank, played by the arpegiator.

  • @pichenettes Ah, OK. I hadn’t looked at the circuit layout closely enough to notice that.


  • Could we have some extra indication with the SVF board which filter you’re currently editing. I know the filter screens are slightly different, but that’s easy to miss if you’re in a hurry, and concentrating on the Cutoff and Resonance values.
    Maybe the LED above switch 2 could flash when you’re editing the second filter.


  • Also, and I know this has been discussed, and dismissed in the past, but can I put in another request for MIDI CC control of the 2nd filter in the SVF? Please….

    Looking at the chart in the manual, there are plenty of unused CCs, and it seems crazy to be able to tweak the first filter from my controller keyboard, but not be able to touch the second. I know the parameters are probably accessible via NRPNs (are they?), but (and I know i keep harping on about this, but it’s still an issue), the ever-popular and otherwise excellent Ableton Live doesn’t support NRPNs….

    Sorry to whinge on about this again. I guess I’m just a bit of a fanatic about being able to control, and sequence everything via MIDI. I know I’m not alone in this though.


  • For now, you can route CC16-CC19 to any parameter in the modulation matrix, including CV outs 1/2, which are those used by the SVF to control the second filter.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion