|
|
Post by feijai on Oct 19, 2023 22:13:57 GMT
The following ditty was done with a SINGLE SEQ16, no other sequencers, driving a few other common modules. It require a fair bit of careful tweaking. Note how close the pitch knobs are to one another. The puzzle: how was it done?
I'm a relative novice with regard to modular, so I'm sure everyone will nail this fast.
[Forgive me, where I am I have no access to an audio interface or even microphone port at all. So I had to record this on my phone. Youtube seems to hate embedding this video -- just click directly on Youtube to see it.]
|
|
|
|
Post by pt3r on Oct 20, 2023 7:41:12 GMT
Hard to tell from just looking at seq 16 but my guess it’s all about the trigger on step 1 and the cv cable going off screen and another cable going into the quantizer. So my guess is that a S&h is triggered on step one which samples from an lfo going along with seq16 and the sampled cv gets added to the cv coming from the seq 16 before going into the quantizer.
|
|
|
|
Post by feijai on Oct 20, 2023 8:28:28 GMT
Hard to tell from just looking at seq 16 but my guess it’s all about the trigger on step 1 and the cv cable going off screen and another cable going into the quantizer. So my guess is that a S&h is triggered on step one which samples from an lfo going along with seq16 and the sampled cv gets added to the cv coming from the seq 16 before going into the quantizer. You're right that the cable at step 1 is critical, and that the SEQ16 doesn't go directly into the quantizer. Good guess aside: my S&H is not used, and one LFO is used but only to drive the SEQ16. I'll add that there's no other SEQ16, SEQ8, or TRIQ164 involved. |
|
|
|
Post by pt3r on Oct 20, 2023 9:28:31 GMT
Do you happen to run a switchmatrix of some kind?
|
|
|
|
Post by feijai on Oct 20, 2023 10:13:15 GMT
|
|
|
|
Post by tIB on Oct 20, 2023 10:29:57 GMT
If there's five voltages into 20 pitches you're combining the sequencer output with something else - sample and hold would be the obvious choice, but if you're not using that I'll go with MMDIV in binary mode into a 4att mix.
|
|
|
|
Post by pt3r on Oct 20, 2023 11:55:31 GMT
If there's five voltages into 20 pitches you're combining the sequencer output with something else - sample and hold would be the obvious choice, but if you're not using that I'll go with MMDIV in binary mode into a 4att mix. I think we have a winner. This seems a very plausible explanation, reading up on the MM-DIV. (I don’t have one myself) |
|
|
|
Post by tIB on Oct 20, 2023 12:16:55 GMT
If there's five voltages into 20 pitches you're combining the sequencer output with something else - sample and hold would be the obvious choice, but if you're not using that I'll go with MMDIV in binary mode into a 4att mix. I think we have a winner. This seems a very plausible explanation, reading up on the MM-DIV. (I don’t have one myself) Anything that outputs 3 fixed voltages with the 4att/mix could do it - you could probably logic it as well. Nice patching btw. |
|
|
|
Post by funbun on Oct 20, 2023 12:41:39 GMT
Yeah, I was about to say you're offsetting the signal via another CV source, summing them into a CV mixer, then quantizing the output. This is the main way I run my solo voice in my jazz fusion-type patches. The only difference is I'm using the rbss as the main sequencer, and a chaos generator and a S&H are added to the rbss via a precision adder. The sum of that goes to the quantizer. Basically it's CV patching.
|
|
|
|
Post by feijai on Oct 20, 2023 15:08:18 GMT
I think we have a winner. This seems a very plausible explanation, reading up on the MM-DIV. (I don’t have one myself) Anything that outputs 3 fixed voltages with the 4att/mix could do it - you could probably logic it as well. Nice patching btw.
Yes, a winner. LFO drives SEQ16, which on step 1 triggers an MM-DIV in Binary mode. I take the first two outputs (bits 0 and 1) and feed them, plus the SEQ16 CV, into a 4ATT/MIX on DC (CV) mode, and output the mix into a QUANTIZER, and then into the CV of the VCO. When the bits are 00 on the MM-DIV, it's one scale, when it's 01 we add a constant so it's second scale, then when it's 10 we add a different constant so it's a third scale, and when it's 11 we add both constants to make the fourth scale transposition.
There are some gotchas. The biggest one is that mixing three CV signals doesn't really sum them -- that'd be too easy :-(. The 4ATT/MIX also seems to divide the CV by a constant, probably by 3, and also I think compresses it (so it's not v/oct any more). In the 4ATT/MIX, I set the output of the SEQ16 to maximum, then have to tweak the other two: but to get the pitch back up again I feed a FOURTH value in: +5V, which increases the division (to 4?) but raises everybody back up to a reasonable pitch. It requires a lot of tweaking to get SEQ16's CV I to be the right four base pitches. After that, adjusting the remaining CV knobs on the SEQ16 will get the rest of the notes right: but because of the compression, the knobs are now VERY sensitive and close to one another as you can see.
The rest of it is a filter, envelope, and delay.
BTW, I am pretty sure this can't be straightforwardly done with one S&H (and sawtooth say), because S&H would sample the four CVs *linearly*, so the pitch would go up by the same amount every 16 steps. But in this example, we transpose to a fifth, then to a minor seventh, then to a major 12th. To do this, bits 00 are the base scale, bits 01 is the fifth, bits 10 are the seventh, and of course bits 11 are a twelfth (seventh + fifth). You need two separate CV sources. Maybe two S&H offset from another?...
Am saddened you guys figured this out so fast. Took me a while to figure out if I could pull it off with the few modules I had.
|
|
cpruby
Junior Member
Posts: 65
|
Post by cpruby on Oct 21, 2023 7:06:30 GMT
Also don't forget that the S&H will decay, so you will have a declining pitch with it. The CVSHIFTER module would be the way to lock in a more stable CV source.
I am really impressed with the thought to do this. I usually have a mind of CV's are CV's and gates are gates, but you're right taking the separate outputs of the MMDIV and throwing that into an attenuator and then mixing it together (I think the CVADD-HQ would be easier than just the 4ATT/MIX) is a great way to trigger harmonic progressions.
|
|
|
|
Post by keurslagerkurt on Oct 21, 2023 12:11:10 GMT
Also don't forget that the S&H will decay, so you will have a declining pitch with it. The CVSHIFTER module would be the way to lock in a more stable CV source. I am really impressed with the thought to do this. I usually have a mind of CV's are CV's and gates are gates, but you're right taking the separate outputs of the MMDIV and throwing that into an attenuator and then mixing it together (I think the CVADD-HQ would be easier than just the 4ATT/MIX) is a great way to trigger harmonic progressions. That's also why you could take multiple gates from the MMDIV (or elsewhere) and throw them at different inputs of a passive divider (like Great Divide, or XMix, or a breadboard with a few resistors) and you get a 'melodic' sequence out! |
|
