Features request for next firmware versions...
  • Of course what would be ideal, but probably a huge undertaking would be some way of having all of these features available as some sort of plugins or modules.

    So you would choose what to include in the firmware by selecting the features you want up to the limit in terms of size or a rough CPU usage calculation.

    It’s one way to keep everyone happy as each person has different priorities.

    But I imagine it would result in a lot of problems especially in regard to compilation or testing.

  • Regarding plug-ins:

    Everything in Shruthi-land is obviously statically linked, and most variables are static (to benefit from the fact that using explicit addressing is faster on an AVR that using indirect addressing). So the only way of doing that would be to put a compiler or at least a linker, in the loop. We’re talking about a class of products and an architecture most vendors are still coding for in assembly :-/ I think this would be way too far into the over-engineering territory. A very nice technical achievement that would bring value to very few users.

  • That’s why he made it open source!

    It would be cool to have text versions of these “plugins” though, like code patches. So that if you really wanted these features in your firmware, you’d need to figure out how to modify and compile the firmware yourself.

    That said, I haven’t yet…

  • Yeah, well plugin is misleading word. I didn’t mean library or DLL style. But some way of not including a code block.

    But I didn’t realistically expect it to be possible :)

  • This could be possible in a dirty way through pre-processor macros – there are actually already a few things disabled in the main build that can be re-enabled with the right -D flag. But people going down this road would still have to install the toolchain…

  • True, unless there was some web based tool that could roll you a build.

    Something like Jenkins for AVR?

    • lost track
  • Yeah not to chime in uninvited, but I like the idea of perfecting the original idea rather than morphing it into something it was never intended to be. This is primarily for selfish reasons as I have already built four of these things and I am looking forward to building the next cool thing as soon as it rolls off the presses.

    Regarding polyphony, the Shruthis sound wonderful and work surprisingly well when polychained. If anything, it would be cool to have a large case or housing that would make polychaining and working with polychained Shruthis easier to manage, play, and carry around.

    Just my .02…

  • Awww, then you won’t be happy with this latest update (URL as usual).

    It comes with a new osc mixing mode called “duo” which implements a simple duophony mode.

    When the arpeggiator is on : osc 1 = current note ; osc 2 = previous note
    When the arpeggiator is off : osc 1 = current note ; osc 2 = least recently held note

    The least recently held note rule allows to hold a note and play a melody above (or below) it. The sub oscillator follows the previous/least recently held note (to get that droning bass). There are many situations in which this “most + least recent note” concept doesn’t make sense at all, and you will have to learn how to recognize them.

    As expected, coding for duophony in a framework designed for monophony is awful. A lot of things still feel broken to me, and unfortunately there’s no good solution besides rewriting the whole thing as a polysynth, something I won’t do of course.

    So please test that and let me know if you can do something interesting out of it. It sounded very interesting to me but quite alien compared to how a polysynth should feel.

    I won’t implement anything more sophisticated than that. The most/least recent rule is not something I will change. Most of the “bugs” as in weird voicing / weird note stealing are due to that.

    Keep in mind that it’s best to have both oscillators set to the same range ; and mixer balance to the middle position.

    Oh and I fixed the negative modulation on attack thing – hope it doesn’t break the patches that uses it the other way round too much (it might have an effect if the modulation amount causes some clipping of the modulation signal). I haven’t change the LFO mod rate because it would break existing patches too obviously. I can point you to the point where this can be changed.

  • I just want to say that this track is beautiful and I can’t wait to try the next refresh. I simply wanted to lend some support for you when you need to say no and stay true to the original concept. When things start to really break the code and make maintaining the project a hassle, let’s move on and do new things. That makes sense to me, keeps the revenue stream healthy, and keeps my soldering iron warm (ahem).

    This place doesn’t ever disappoint me, Olivier. This place is awesome.

  • I just played with seqmix for an hour and a half using only the test note. Great times! And it’s so nice to be able to tweak the noise volume now. Really amazing stuff, pichenettes. We are far too spoiled here! Can’t wait to check out duo mode. As if the Shruthi-1 wasn’t already in a league of its own, these new features are really making this thing a crazy crazy sexy beast.

  • Just finished a Shruthi XT. Lotsa fun.
    It would be really nice though if the display would switch to the correct page as soon as I turn a knob on the programmer. Would this be possible to implement without too much hassle?

  • >Oh and I fixed the negative modulation on attack thing – hope it doesn’t break the patches that uses it the other way round too much
    Thanks. And Duo mode sounds weird and interesting. :)

  • Wow. I just updated and played with the duophony and mix sequencer. Both are lovely. I am so sad that I have to leave for work, like, now. I just want to stay home all day and play with this. Fantastic work, Olivier.

  • So far i’ve not found these new additions broken, or alien.. Just quirky! Fortunately I like quirky very much!
    Little tricks like this can really set this synth apart from the crowd.
    Thanks for doing it Oilvier :)

  • I was playing with the sequencer and was wondering if it would be useful for the sequencer to have a loop then next mode? Kinda like in abelton live.

    So if you had

    Loop Next
    4 2

    It would play the sequence four Times and then start playing sequencing #2. This way you could build up little songs for shurthi to play while you play the guitar or sing somewhere else/

    Loop Next
    0 x

    Would be to do a oneshot sequence, maybe a sound effect at the end of a song or something.
  • Yeah, regarding what herrprof said, I was just thinking that it would be nice to have an easier/quicker way to change sequences, but I was thinking using MIDI. Maybe having an option to choose whether program change messages update patch, sequence, or both. I guess right now you can have them either both load, by attaching the sequence to the patch.

    Of course then you have to worry about whether the sequences load immediately or only after they touch the last step and then there’s possibly lag due to loading from EEPROM (although you could load only the steps appearing before those currently playing). And then you have to find a spot in the menu for this stuff. Ugh.

  • The lag due to eeprom access is too high.

  • New request… is it possible to add a mod. destination for just the ‘fine’ pitch of oscillator 2? I feel like I want more subtle control than the ‘osc2’ destination gives me.

  • I think it’s a bit late now to add new modulation destinations like that… If I add them at the end of the list, the list looks quite random. If I put the new destination at the right place in the list (somewhere near osc2 maybe), it’ll break existing patches – there’s no indirection between the value stored in memory and the way items are ordered in the UI – and it would be cumbersome to implement this.

  • I don’t think one extra option at the end would be random… just an added bonus :D But that’s ok. I’ll have to make do with an mod value of 1 for subtle chorus like effects. thanks.

  • Yeah I have some issues with some of the ranges, those are probably near the top of the list of things I’d want to modify. Too bad that would break existing patches. It would be ideal if the places in the code to tweak these were really obvious:

    • FM modulation index (“param”) maximum value – want to increase
    • Depth of modulation with LFO rate as a destination – want to increase
    • Depth of modulation using LFO as source – want to make finer at lower values.

    Again probably just another argument for me to figure out compilation out on my own, since it will break existing patches. Although a patch conversion tool might solve that problem.

    • The FM modulation index cannot be increased. Just check the code. Both parameter and the modulator have values in the 0..255 range. This is the normal behavior and range for a DX7-style phase modulation implementation. You expect this to behave like “analog” FM with through-zero oscillators. This is not going to happen just by changing a value in the code :D
    • Use U14ShiftRight6 there ; add a >> 1 there and subtract 64 here
    • ( >> 1) + 64 here.
  • Ohh, so it’s not phase modulation? EDIT: Nevermind. I misunderstood. I just could have sworn I have gotten nastier sidebands out of 2-operator FM.
    Thanks for that code info!!!

  • Please forgive me if this has been covered already.

    I’m trying the new build on a DSP model. When I use the synth in duo mode, I can hear it cycle through the oscillators. The mix setting on the OSC page seems to do a lot more than just mix the two oscillators now. Occasionally there is an odd note with a lot of really strange modulation that I have not dialed in. It’s like random…. just play along single notes, sometimes two notes, and then you get this really warped sound.

    Changing the sub volume has huge awesome unexpected effects…

    The sound of the synth patch seems to be different now. From the start and even after editing to try to get back to the patch I had dialed in pre-duo update. It seems to have neutered the digital effects somewhat… the distortion isn’t as loud nor is it as badass…

    I hope this helps. I’ll test some more later.

  • For the “warped sound”, please post a sound sample!

    > The mix setting on the OSC page seems to do a lot more than just mix the two oscillators now.

    Well, it does, but the fact that one or the other oscillator is off most of the time doesn’t make it clear. Let’s say you set it to 10. If you play only one single note, it will be played by osc 1 with a high volume, and if you play a second note, it’ll be played by osc 2 with a low volume. Setting mix to a value different from 32 makes little sense for the duophonic mode.

    > Changing the sub volume has huge awesome unexpected effects…

    It’s exactly the same behaviour as in the other modes, but keep in mind that the sub “jumps” from osc1 to osc2 depending on which one plays the lowest note.

    I think you have to approach this duophonic mode with a very open mind. By design, it cannot behave like a true duophonic synth – just take it as a playable, unpredictable, harmony thing :)

  • I will post something tomorrow. Any idea why the patches sound different than the last V0.96 preview? That’s what I had installed earlier today… I’m going to downgrade and test my memory.

  • Two recordings of the same patch “before” and “after” the change would be very helpful!

  • I noticed a couple glitches too but I thought they were my imagination and they sounded kinda cool anyway. I need to be more thorough in my testing!
  • What do you mean by glitches? I’m going to burn this on a hundred chip tonight so I’d rather have this fixed, if there is anything to fix :D

  • I would check the CV mode setting. For some reason mine has changed mode before now. It made things misbehave for me.

    It should be 4cv by default. Prgm if you have the programmer case.

  • I had some very brief dropouts and clicks playing in the duophonic mode, mostly right after hitting the 2nd key.
  • Actually, I think something serious may be wrong. I flashed twice because it sounded so odd right off the bat. Like really different and I couldn’t tweak back to where I had been only hours before… I would avoid flashing if you can wait and debug a bit more… I will try to post samples later today.

  • Okay I’m pretty sure this was me. Sorry for the sounding the alarm.

    After puzzling over this again today I realized it’s a problem (that I have had before and forgot about) with MidiPatchBay… sometimes it sends errant midi data and you get every nth note transposed and more. As soon as I noticed it was running and killed it, the Shruthi calmed down and started sounding exactly like it was supposed to.

    Sorry… go ahead and burn your chips ;-)

  • I'm not going to be home for at least another 8 hours but I'll try to replicate it when I'm back in front of my Shruthi.
  • I think there should be an equivalent of the fight club rules for the duophonic mode.

    There’s a very good chance that what you are seeing is:

    • osc 1 and osc 2 set to different waveshapes ; and thus the chord sounding differently depending on which note goes to which oscillator ; or “jumps” due to the same note being reshuffled to a different oscillator the next time it is triggered.
    • double envelope retriggering because the two notes of the chord are not played exactly at the same time.

  • In fact, duophonic sounds awesome!

    Check this

  • @pichenettes Yes, I figured out on my own that sloppy playing can create some really odd stuff in duophonic mode. If you are careful to keep your notes from overlapping it will sound like you expect, but as soon as two notes sound off together it’s completely different. I love it actually.

    What was really throwing me off was the MidiPatchBay throwing in bad midi data randomly because that was happening on the init patch and every other patch even if the synth was just booted and the mixer in “sum” mode.

  • @pichenettes, I noticed that but after I got a sense of the feel of it, it was fine. This was something else, it was like a short dropout or blip after I was already sustaining a couple of notes. It wasn't that obvious though and I actually liked the flavor of it.

    Also I also really like how much variety this feature gives you. I like changing the tuning of the 2nd oscillator so that when you play a second note, the original one goes a little wacky. Lots of opportunity for nice surprises.
  • Off topic but maybe you guys can answer this…

    I have the Oto Biscuit wired up as a send effect. A pair of outputs from my interface are hooked up to the Biscuit’s inputs and the outputs are connected to a pair of interface inputs. I use the interface software to divert other channels to that output pair. Anyway, I’ve done this with other FX that I have, the Biscuit just happens to be wired up to it now.

    What I noticed yesterday is that even if the Biscuit is powered off, if that send was active it was changing my sound! This was contributing to my confusion yesterday which was apparently … profound. :-/

    How can a piece of electronics affect my signal when it’s turned OFF?

  • Is the send still active? Perhaps its passing audio even when turn off? And since the passed audio is a little behind the playback audio phasing is occuring?
  • The OTO uses relays for true bypass. Probably that when it is off, the audio is still going through the cable and the relay. Don’t think this is enough to color sound, though…

  • It was really odd, almost like ring modulation. I’ll try to reproduce.

    Update: Yup, it sounds like ring modulation actually. And it works even if the Biscuit’s PSU is unplugged. I have an analog tape echo pedal similarly hooked up and have never noticed a problem with it. It’s supposed to be true bypass as well. Well anyway, it’s odd.

  • nice titus. i think i`ll have to build a digital filter board, too

  • please forgive my ignorance, but where does one find this mystical update with “duophonic mode”?

  • It’s firmware V0.96 available here

  • oh? oh…yep, it was my ignorance. I already dumped it in. Would help if I had read to change the mix to “duo”. Thanks Titus! The first couple of test notes I pressed just blew me away! What a magical little box this truly is.

  • I'm really enjoying seqmix mode with the pattern sequencer and the test tone. Awesome work.

Howdy, Stranger!

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

In this Discussion