youuuuuuu are a damn SAINT and HERO. this is wonderful news! thank you for sharing it!
it was quite a simple exercise after rod’s modeling wizardry but thank you sir derks.
half the cool of the modes code is in the buffer clicks and playback modifiers for karma though. as I’ve hinted I’ll be releasing a m4l-mlr-thing anaphora for grids and no-grids soon and I’m planning on adding in all that with the saturators after hearing the demo patch.
to my ears this stuff is easily as good as a pricey plugin. as someone with 0 knowledge of hardware modeling I’m pretty impressed and super happy to be able to grab it up and incorporate it into my own open-source projects.
a few days late: nick, thank you so much for the kind feedback and for engaging so wholeheartedly with the album <3
‘pingy sine tones’ is the name of the game, for sure. they’re all from the oscillating 3 Sisters, which is FM’ing the filtering 3 Sisters, all gated with DPLPGs.
short of a custom m4l device for looping and digital artifacts (the latter @andrew has thankfully collected!), the album was made with about 62hp (minus power):
- cold mac
- make noise function
- 2x 3 Sisters
- 2x DPLPG
- 2x w/
“his name was squeeze” is the only track that adds the two w/'s into the mix. iirc, one was feeding into the other. the ‘blown horn’ sounds are the cross-FM’ing 6 Sisters again, played manually. lots of knob-riding, which is why a small system was key.
thank you for asking and plz ask any others!
this is the thing that still makes me go “ modular”
is there a difference between resonant 3 sisters and a sine wave?
I just realized that none of the things you listed are supposed to make sound which is funny to me
resonant three sisters can be “pinged” if you patch a gate into the input // will still pass sound through even in self-oscillation. Even in the absence of the above it’s also not likely to be purely a sine wave, which is actually a good thing.
lol, and THAT’s the thing that makes me go “ modular”
re: oscillating 3 Sisters…
the last few months have been a journey to find an alternate sine wave oscillator. ended up back at 3 Sisters. to my ears, it is exactly the right vibe — and the sines it generates are perfect for achieving complex timbres through cross-patching with another 3 Sisters (one sine into the regular inputs, another into the FM jack). the pair opens up some really deep yet controllable modulation.
so i guess my go-to “complex oscillator” is two filters. they’re magic, though, and riding the FM knob makes for a really expressive instrument.
fun fact: i dunno what topology is used in 3sis, but it happens that the 2-pole state variable filter does indeed produce a perfect sinewave as its impulse response (with Q=inf.) in fact it produces a sine and a cosine.
in that configuration, the difference equation of the discretized version would be:
// init: x = 0 y = amp // update sample: x = x + c * y y = y - c * x
c is the frequency coefficient
2pi * freq / SR, and
y produce cos/sin.
as far as i know, the main reason this isn’t more widely used as a low-cost oscillator, is that the amplitude of the sinewave varies nonlinearly with frequency (low freq = higher amp.) its a great trick for quadrature LFOs though.
(of course its true, in analog world nothing is perfect, including impulse functions, thank goodness)
thank you for this! I just reached the part of my project where I have to think about whether I wanted to just copy-paste Peter McCulloch’s
gen~ ladder filter or look into something else… this adds more fuel to the “learn something” side of the equation! (although, like, the filter isn’t supposed to be the focus of the thing, it just felt like it would be rude not to include one, given it’s got plenty of capability for harsh, harmonic-rich tones)
I similarly don’t know the topology of 3 sisters, but reading between the lines in the manual suggests it would be easiest to make such a thing with three state variable filters…
whoa. okok - filters are currently blowing my mind. my personal creative view of analog/modular synthesis is changing kind of rapidly.
'mind being blown'
SO - like, a week ago, I talked to @jwm about this album because it sounded extremely lovely to me and I could not for the life of tell how it was made:
IT TURNS OUT - yep - also mostly filters (from what I gleamed of his description). specifically, resonant bandpass filters turning noise into sine-ish drones and percussive bits.
I tried the same technique out in Ableton by setting parallel Auto Filter’s to bandpass/OSR/high resonance, then manually setting Freq to notes and feeding gated lake superior into them (+ my reverse delay). it worked magic:
(the way this works is actually quite comparable to the way that the body of a string instrument colors sound which is so super-nice next to my acoustic instrument background)
MEANWHILE - fm synthesis has unexpectedly worked its way back into my music (after selling my big ol’ yamaha DX27 to @Jonathan_Riley (hope it’s treating u well jon) for the same interface-related reasons that a lot of people do) in the form of one organ preset on a ebay toy keyboard I had laying around (specifically a yamaha pss-170). It sounds amazing to my ears and it’s sort of become an instrument all on its own, (especially sampled and slowed down) ex:
the problem: it’s a preset
I WANT CONTROL - so last weekend I tried recreating preset 13 in Ableton Operator and got pretty close. it sounds like 4 sines split between two octaves, with one of the sines FM-ing one of the other sines. messing with fine-tuning and amplitudes reveals some nice n playable results.
WHAT IT SOUNDS LIKE THIS MEANS - is that both drifty-filtered-noise drones and bloopy-fm keyboard scribbles can be made with the same system. a synthesis of the synthesis things that I like.
I found this thing (one sister?). from what I know right now four of them with some cv friends seems like it can do both techniques. maybe more. or I might just explore the idea in max. it’ll probably be both. thanks every1.
The Ableton filter is great. I love the ms2 mode with 12db slope. If you pull the drive up it sounds really really nice.
A really interesting filter design to check out would be the xaoc belgrad. It has tons of ways to adapt the poles in different configurations
Also analog filters are really useful in feedback patches, where you mult out from the filter output and feed it back into the input with some attenuation. This allows you to control the resonance in some other way even on modules that don’t have a cv input for resonance. I’m curious to know how the Ableton filter would react in a feedback patch
be warned: DX7 / Operator-style FM is linear (adds to the frequency instead of multiplies) and through-zero (setting the frequency negative inverts the waveform). I think the FM input on Three Sisters is linear, but it’s unlikely that it would be through-zero.
also, Max is actually pretty good for playing with this, they even have builtin objects to play with these ideas!
ETA – this reminds me to see if I can get anything useful from adding an input option to my Max synth project
also, 3Sis (and other filters capable of 1v/oct) is not temperature-compensated. so, one will work well and track musically, but they just aren’t as stable as a dedicated sine wave osc. so unless you want to keep stuff more-or-less monophonic (3Sis’s Span can help you fudge this), keeping a bank of oscillating filters in tune with each other is going to get frustrating quick.
this does make me wonder if there’s a comparable module, though, whose sine mimics but is temperature-controlled. i just have never been able to get the same timbral modulations from triangle core sines (really, DPO and STO) as i do 3Sis’s.
interesting interesting. this is the whole analog vs. digital thinking thing. as a player of instruments I am kind of used to multiple pitches going out of tune, but not every few minutes/hours. in what kind of timescale does 3sis lose the tune? I suppose it also helps to have a fine-tune knob.
and yea, the original iteration of the idea was max-only, as this is the-way-i-usually-do-things, but I’ve been starting to feel the direct-vs-indirect music interface thing, and the analog is imperfect thing. I love doing looping and delays in an indirect-interface way (where the link between interface and sound is more arbitrary/set by me), but my curiosity of modular direct-interfaces is growing.
I guess i just have to evaluate the validity of an analog thing which is as flexible as I want/need it to be.
sorry, I edited my previous comment to clarify – once warmed up, it doesn’t drift much at all. the overall tracking is a potential trouble spot when playing with others, since it isn’t really designed to do that as well as a dedicated osc.
I have a sequence running right now that spans two and a half octaves on both a DPO and 3Sis set to the same base frequency and there’s a noticeable detune (sharp) on 3Sis’s side once more than an octave up. using the Low output, as that is trimmed for 1v/8. again, super musical and expressive on its own – against other voices, though, it might not track as well as you’d like.
Mannequins Three Sisters
hmm, ok that does make more sense. things being in tune with other things is generally something I’m interested in so that is good to know before potentially diving into something.
I guess one alternative is using a hardware digital filter (4ms smr is what @jwm was using), though that complicates my oh-so-holy original intent for attempting hardware, and makes software yet again seem more desirable (which I’m pretty down with).
another solution that comes to mind I guess is using a control source (computer) that’s more flexible can be ‘tuned’ to what the filters feel like doing at the moment.
yeah things being in tune is generally something i am interested in as well, and, at the time, the smr was handy in that way. the comments made re: keeping banks of oscillating analog filters in tune is pretty on point and really, a dedicated sine oscillator with v/oct tracking and fine tune is where you wanna be (and yeah, theres a lot to be said about the purity of a filter’s sine wave vs and tri core osc’s sine, but…) but even then, especially with analog oscillators yeah, you’re gonna have drift…then you have the issue of the calibration of your v/oct source, power supplies, etc etc.
i know there is a lot of love around here for sisters and mannequins in general, but theres definitely some other (digital and otherwise) options out there that will get you into this territory, its just a matter of how you use/interface with them (ie my mentioned abuse of formant filters.) i’m sure a lot of people would look at my patches and go, yo dude, you know there is like, 2000 better ways to do this right? but it works for me the way i built it and the way i like to put my hands on things and patch i guess, and i just never enjoying mousing around in ableton that way (though i applaud those who can/do for sure)
also, i have no idea if i added anything to the small part of this conversation i’m connected to! i don’t forum very well, ha
in closing, here is a pretty great vid of a really expensive sequencer pinging a really expensive filter(s)
intellijel is saying it tracks accurately over 4+ octaves but my knowledge of these things is obvs not so vast
sorry to overwhelm your thread with gearspeak dan, just currently very curious about these things !
couldn’t ask for better conversation! the core of ‘less concepts’ (the album) was tackling these very questions and investigations. I haven’t arrived at any definite + unwavering answers, but I’ve learned a lot through the process and am really excited by what you’re exploring.
re: uvcf, 3Sis is designed to track 4 octaves as well – I haven’t futzed with re-calibrating, so will likely do that tomorrow to try and nail this down. if I can dial in both of mine to track along that range, I’ll report back.
you for sure did. sidenote: Barque Of Phosphor is gorgeous. so, if for nothing else, I’m really glad you stopped by