Author |
Message |
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Fri Dec 16, 2005 11:30 pm Post subject:
continuosly variable oscillator waveforms |
|
|
i may be asking a painfully obvious question..
would anyone have any insights on how to programme an oscillator to 'morph' from a sine to a square, ramp etc over time? _________________ Spiral Recordings |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Sat Dec 17, 2005 1:51 am Post subject:
|
|
|
The Moog Voyager has continuously variable triangle->saw->square->pulse controls on its oscillators. Not a common feature but I’ve seen how they do it thanks to data posted on a Voyager site. The triangle to saw section is a simple cross fade, saw to square combines cross fade and AM between the variable width rectangle and the sawtooth, the final section is your basic PWM from square to narrow pulse.
The technique used to morph only effects the in between waveforms, cross fading is probably the least interesting. The G2 shape oscillators have a true triangle to saw option, where the ratio of rise and fall times changes with the modulation signal. This is very hard to do on analogue oscillators but on the bright side, the effect sounds a lot like low pass filtered PWM, so no big deal.
You can use variable saturation (overdrive/clipping) to morph a triangle or saw into a square-ish wave. |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Sat Dec 17, 2005 2:15 am Post subject:
|
|
|
g2ian wrote: |
You can use variable saturation (overdrive/clipping) to morph a triangle or saw into a square-ish wave. |
cool..thanks for your interesting response
and i will try that simple solution you mentioned above _________________ Spiral Recordings |
|
Back to top
|
|
|
aquanaut
Joined: Apr 25, 2004 Posts: 313 Location: Montreal
Audio files: 4
|
Posted: Sat Dec 17, 2005 6:57 am Post subject:
|
|
|
With Sounddiver you can import or draw all 64 waves of a microwave XT wavetable. If you put a sine as wave 1 and a square at 64 you get a morph for all remaining waves. |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Sat Dec 17, 2005 9:12 am Post subject:
|
|
|
nice ! _________________ Spiral Recordings |
|
Back to top
|
|
|
elektro80
Site Admin
Joined: Mar 25, 2003 Posts: 21959 Location: Norway
Audio files: 14
|
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Sat Dec 17, 2005 9:25 am Post subject:
|
|
|
yeah right on !... the 'sequence' mp3 on that site demonstrates this well _________________ Spiral Recordings |
|
Back to top
|
|
|
elektro80
Site Admin
Joined: Mar 25, 2003 Posts: 21959 Location: Norway
Audio files: 14
|
Posted: Sat Dec 17, 2005 9:26 am Post subject:
|
|
|
The Plan B Model 15 has CV controlled morph too. Cute.
http://www.ear-group.com/model_15.html _________________ A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"
MySpace
SoundCloud
Flickr |
|
Back to top
|
|
|
Mohoyoho
Joined: Dec 03, 2003 Posts: 1632 Location: Tennessee
Audio files: 8
|
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Sat Dec 17, 2005 6:51 pm Post subject:
|
|
|
Mohoyoho wrote: | This might be obvious and not what you are looking for, but in an instrument like the Voyager or some soft synths, would not using an envelope generator to modulate the wave form do the trick? or a slow LFO? |
funny , yes i did think about that, but like you, thought it might be too simple..but maybe it's just the trick...will try that out as well _________________ Spiral Recordings |
|
Back to top
|
|
|
seraph
Editor
Joined: Jun 21, 2003 Posts: 12398 Location: Firenze, Italy
Audio files: 33
G2 patch files: 2
|
Posted: Sun Dec 18, 2005 2:41 am Post subject:
|
|
|
aquanaut wrote: | With Sounddiver you can import or draw all 64 waves of a microwave XT wavetable. If you put a sine as wave 1 and a square at 64 you get a morph for all remaining waves. |
right, but not on OSX because that f***ing application has never worked under OSX |
|
Back to top
|
|
|
aquanaut
Joined: Apr 25, 2004 Posts: 313 Location: Montreal
Audio files: 4
|
|
Back to top
|
|
|
seraph
Editor
Joined: Jun 21, 2003 Posts: 12398 Location: Firenze, Italy
Audio files: 33
G2 patch files: 2
|
Posted: Sun Dec 18, 2005 10:26 am Post subject:
|
|
|
Quote: | allows OMS MIDI communication under Classic |
are you kidding me no Classic, no way _________________ homepage - blog - forum - youtube
Quote: | Don't die with your music still in you - Wayne Dyer |
Last edited by seraph on Thu Oct 12, 2006 1:08 am; edited 1 time in total |
|
Back to top
|
|
|
Kassen
Janitor
Joined: Jul 06, 2004 Posts: 7678 Location: The Hague, NL
G2 patch files: 3
|
Posted: Sun Dec 18, 2005 1:05 pm Post subject:
|
|
|
Í don't understand any of this discussion. What are you programing in? What way should they morph? You see; you can make anything into a square by clipping it but that's not the smoothest way of doing it. You can probably crossfade in many systems but that does depend on either sample acurate math or sample accurate sample playback.
as long as you just want to go from one shape to another and don't particularly care about all the details then I'd take a stereo sampler, put a sample in it with the two waves, one panned hard left, the other hard right, get both outputs to the centre position of my mix and take a pan controler to the sampler. It's cheap, easy, dependable and it's going to glitch if that pan isn't smoothed before being applied. More advanced samplers might allow for controler or envelope based crossfading between two or more samples. Multi samples should give you sample acurate playback for free but that depends on you finetuning your samples (which here would be more acurately called wavetables).
The world is your oister here; you can make it as complex or simple as you want it. A second hand esi-32 will do it for you for peanuts. Matlab will also do it, matlab will potentially do it in a much nicer and more interesting way at the cost of your social life ;¬) _________________ Kassen |
|
Back to top
|
|
|
Mohoyoho
Joined: Dec 03, 2003 Posts: 1632 Location: Tennessee
Audio files: 8
|
Posted: Sun Dec 18, 2005 1:29 pm Post subject:
|
|
|
paul e. wrote: | Mohoyoho wrote: | This might be obvious and not what you are looking for, but in an instrument like the Voyager or some soft synths, would not using an envelope generator to modulate the wave form do the trick? or a slow LFO? |
funny , yes i did think about that, but like you, thought it might be too simple..but maybe it's just the trick...will try that out as well |
That's what I do. The Voyager has the variable waves from sine to square. I love modulating it with one of the mod busses or with one of the envelope generators. There's a CV input in the back of the V'ger that give you a direct input so you can use an outside CV as well. The busses are already hardwired to modulate the waves. Unfortunately the input in the back modulates waves on all three osc. at the same time and not individual osc. My Q lets you modulate the different oscs' waves. |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Sun Dec 18, 2005 10:30 pm Post subject:
|
|
|
Kassen wrote: | put a sample in it with the two waves, one panned hard left, the other hard right, get both outputs to the centre position of my mix and take a pan controler to the sampler.
|
it's a good idea in itself, but does not provide a 'continuosly variable' wave form shape..it's just cross-fading between just two distinct waveform states
---
i discovered the Korg Prophecy has a 'waveshaping' parameter that will alter the waveform shape over time...am looking into that as well |
|
Back to top
|
|
|
Kassen
Janitor
Joined: Jul 06, 2004 Posts: 7678 Location: The Hague, NL
G2 patch files: 3
|
Posted: Sun Dec 18, 2005 11:21 pm Post subject:
|
|
|
Yes, I know that, in fact I noted exactly that in the half-phrase just before the bit you quoted.
This would be much easier if I knew exactly what it is that you want to program and what you want to program it in.
Did you considder that if phase is no issue then additive synthesis might actually be a viable option here? You might not actually get a square or saw but you will get the sound of those and implementing the morphs would be quite straighforward. You could add a check that compares individual harmonics's eventual frequency to the nequist and mute them if they end up over it; that would do quite nicely as a anti-aliassing strategy. _________________ Kassen |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Mon Dec 19, 2005 1:00 am Post subject:
|
|
|
Kassen wrote: | Did you considder that if phase is no issue then additive synthesis might actually be a viable option here? You might not actually get a square or saw but you will get the sound of those and implementing the morphs would be quite straighforward. You could add a check that compares individual harmonics's eventual frequency to the nequist and mute them if they end up over it; that would do quite nicely as a anti-aliassing strategy. |
interesting..i do have access to a basic additive synth..[the kawai k3]
but the harmonic wave editing parameters seem pretty esoteric on the 10 character LED displayhehe _________________ Spiral Recordings |
|
Back to top
|
|
|
Kassen
Janitor
Joined: Jul 06, 2004 Posts: 7678 Location: The Hague, NL
G2 patch files: 3
|
Posted: Mon Dec 19, 2005 2:46 am Post subject:
|
|
|
Yes, that sounds like a chore and I supose that once you have one implementation you can do it all over again next time you need a different morph.
To realy be able to get any arbitrary wave-morph that you may want you are going to need something modular, practically speaking (though I maintain that a decent sampler with multisamples might get you by). The problem is also that unless you can set your morph to some NRPN it's going to glitch unless parameters are smoothed internally. This is one of those wonderfull subjects that manifactureres see no need to document in manuals. It would be a good idea if samplers and synths would have a option to delay the application of CC's untill the next zero crossing and for all I know some do. I wouldn't bet on it though.
Next up; most softsynths are going to choke on a dozen or more arbitrily defined oscilators for a good number of voices. That's not a fundamental (no pun intended)problem; you could render offline.
Basically I think your choices come down to making do with what manifacturers have already implemented (which will typically be a fairly limited fairly concervative range of options from a theoretical point of view but quite effective musically) or improvise something and just swallow the glithces or cook up your own. If all else fails then my proposed strategy implemented in Csound *will* get you what you want with all hamonics up to the Nequist, no glitches and no aliassing at all for any number of voices. It won't be that much fun and in all likely hood it won't be realtime either but it *is* going to work.
I don't think that's worth it since clever use of filtering and various kinds of waveshaping can get you quite close to the sound you'd get that way with much less trouble but then again; I have no idea how badly you need morphed waveshapes. _________________ Kassen |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Mon Dec 19, 2005 6:47 pm Post subject:
|
|
|
Kassen wrote: |
but then again; I have no idea how badly you need morphed waveshapes. |
so bad i can taste it..hehe no just kidding..just a fascinating idea to me right now...also thinking about an alternative method of sequencing an album of trax, by applying a similiar concept of variable waveform shapes..i.e. analyzing the 'shape' and therefore the order of how the trax appear on the album , based on either tempo or some subjective 'feel/groove' parameter as the 'envelope'
and i appreciate your input/insights |
|
Back to top
|
|
|
Kassen
Janitor
Joined: Jul 06, 2004 Posts: 7678 Location: The Hague, NL
G2 patch files: 3
|
Posted: Tue Dec 20, 2005 3:55 am Post subject:
|
|
|
Ah, ok.
Well, if it's such a conceptual thing I would indeed tackle it on the waveshape level instead of a sound/harmonics one. In that case I think a modern (soft?) sampler with multi samples might be your best bet. Try and find one that can use a parameter like modulation to X-fade between samples. That's not so rare or exotic but mostly those get used to X-fade based on velocity and typically people use presets (think of those multi-gig pianos). It's of cource vitally important that you cut all of your samples to have their zero crossings at the same point.
It's a realy interesting problem; you'd think there would be mor ready made solutions because it's not such a outlansish plan at all. _________________ Kassen |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Tue Dec 20, 2005 8:56 am Post subject:
|
|
|
Kassen wrote: |
It's a realy interesting problem; you'd think there would be mor ready made solutions because it's not such a outlansish plan at all. |
yes indeed, i agree... _________________ Spiral Recordings |
|
Back to top
|
|
|
ndkent
Joined: Jan 03, 2006 Posts: 66 Location: new york
|
Posted: Mon Jan 30, 2006 1:34 pm Post subject:
|
|
|
Serge built a VC waveshaping circuit into their NTO (their higher end VCO). It apparently starts with saw, shapes that into a sine and then into a variable pulse. Enough modulation voltage will take through all of them or attenuated modulation will keep you within a limited range.
I suspect Modcan built their 12a Morph / Peak & Trough based in some part on Serge's designs. It proves that the circuit is a genuine waveshaper rather than crossfading waves since with modcan module you supply your own saw wave externally (...or any other wave which of course makes me curious what other waves will do if processed, I have an NTO which is hardwired to a saw and has linear FM, but no external wave input like Modcan).
Finally Serge makes a Waveshaper that inputs an external saw and VC shapes that into a sine. Their "entry level" osc, the PCO doesn't have a sine output so this module might have been a unique way to produce one. At the moment no one has a similar or copied module.
nicholas d. kent
http://technopop.info/ndkent/ |
|
Back to top
|
|
|
paul e.
Joined: Sep 22, 2003 Posts: 1567 Location: toronto, canada
Audio files: 2
|
Posted: Mon Jan 30, 2006 9:47 pm Post subject:
|
|
|
ndkent wrote: | Serge built a VC waveshaping circuit into their NTO (their higher end VCO). It apparently starts with saw, shapes that into a sine and then into a variable pulse. Enough modulation voltage will take through all of them or attenuated modulation will keep you within a limited range.
|
hey, now that sounds fun ... _________________ Spiral Recordings |
|
Back to top
|
|
|
morbius
Joined: Feb 12, 2006 Posts: 95 Location: Great Smoky Mountains - USA
|
Posted: Sun Feb 12, 2006 3:25 pm Post subject:
|
|
|
Well.... Ok... let's assume that one wants or needs to patch a modular system to 'morph' from one waveshape to the next.
Output all four wave outputs, each to it's own VCA. Use a sequencer (set to four-stages) to control when each VCA opens. Ta da.
~Morbius~
www.unisynth.com |
|
Back to top
|
|
|
|