MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Using one sequencer for multiple oscillators
MUFF WIGGLER Forum Index -> Eurorack Modules Goto page 1, 2  Next [all]
Author Using one sequencer for multiple oscillators
kanstraktar
Mylarmelodies has a whole series on creating a case for live improvisation, based on Steevio’s ideas, one of which is using only one sequence of notes (that may come from multiple sequencers added together using precision adders) and distributing it to all the oscillators in your system. The point here is that if you tune the oscillators differently, they won’t be playing the same notes, instead they will create a chord based on the original sequence. For example, you could have your bass oscillator tuned to C, a second one 5 semitones above and a third one 7 semitones above. So far makes perfect sense and especially listening to Steevio’s improvisations, it sounds really good.

The problem for me is putting it in practice: in my system it sounds wrong. I don’t know too much music theory but the problem I see is that each oscillator is playing a different key. Based on the example above, a sequence in a major key would have one oscillator playing in C major, second one in F major and third one in G major. It’s true that when they sound together they form a chord, but they are triggered differently so they create three different melodies in three different keys. Whenever I try it it just sounds too chaotic.

The only way I see it possible is to tune all oscillators to the same note, transpose the sequence and pass it to a quantizer for each voice, this way they could create chords and play the same scale.

Any other advice, pointer?

Thanks


Edit:

The videos I'm refering to:
https://www.youtube.com/watch?v=x6hJa2lRRgM
https://www.youtube.com/watch?v=QnYDR61a8TA
My system:
https://www.modulargrid.net/e/racks/view/973366
Parnelli
Are you running your pitch CV through a buffered mult? I ask because if you run it through a regular mult or use stackables the pitch cv becomes loaded down a bit by each oscillator it feeds and changes. If you run it through a buffered mult each output will provide each oscillator exactly the same CV as the initial input value.
Moerder
heyho, could you give more info about your setup? like which sequencers and oscillators you are using etc. perhaps even a link to the video you were watching, for reference.

I'd say it depends on the actual modules to some extent, because some oscillators might not have an input or parameter to adjust, and the sequencer as well. for example, a Voltage Block has 8 channels that can each be set to a specific scale, division, length.. if you made one sequence on one channel and then multed this to your 3 oscillators (that have been tuned to the same note) then you'd just have all 3 playing the exact same thing, which is probably not particularly exciting.

so you could look into music theory to find out which notes/scales and so on go with each other or don't, to set up specific seq/scales for each osc. a quantizer would certainly help in making sure that all 3 osc notes 'go with each other' in the way I think you intend. but perhaps you can also just try simple things like a suboctave or one above, different gates, reverse or otherwise alter the 'playback' of your sequence to the osc and so on.

still not 100% sure what you mean and can't be bothered to look through the yütüb archive at the moment. also don't have 3 dedicated oscillators but trying to make the most and more from what I have so will check back here later smile all the best!
damase
maybe this trick is meant to sit on each step of the sequence for a long time or to only advance to the next step as a ‘song change’...

by doing this you have just a major triad chord, yes, but you can fire off the envelopes for each oscillator in varying ways, like starting to drone on the 1st, then gradually bring in the 5th ect. ect. then eventually go to the next step for a new chord/song

its still pretty limited honestly, but i cant see how it would sound good to feed a fast moving sequence into a pre tuned chord without any way to change it... well it would be wayyy more limiting at least
kanstraktar
Parnelli wrote:
Are you running your pitch CV through a buffered mult?


Yes
kanstraktar
Moerder wrote:
heyho, could you give more info about your setup? like which sequencers and oscillators you are using etc. perhaps even a link to the video you were watching, for reference.

I'd say it depends on the actual modules to some extent, because some oscillators might not have an input or parameter to adjust, and the sequencer as well. for example, a Voltage Block has 8 channels that can each be set to a specific scale, division, length.. if you made one sequence on one channel and then multed this to your 3 oscillators (that have been tuned to the same note) then you'd just have all 3 playing the exact same thing, which is probably not particularly exciting.

so you could look into music theory to find out which notes/scales and so on go with each other or don't, to set up specific seq/scales for each osc. a quantizer would certainly help in making sure that all 3 osc notes 'go with each other' in the way I think you intend. but perhaps you can also just try simple things like a suboctave or one above, different gates, reverse or otherwise alter the 'playback' of your sequence to the osc and so on.

still not 100% sure what you mean and can't be bothered to look through the yütüb archive at the moment. also don't have 3 dedicated oscillators but trying to make the most and more from what I have so will check back here later smile all the best!


I updated the original post with the links to the videos and my current system. I use Metropolis as my main sequencer, while the Deopfer is going into its Aux in, modifying further the sequence. Then into a buffered mult and then: first out goes to the first oscillator, while the second and third go into sample and hold and then to the oscillators. I then use the triggers from the Metropolis to control the amplitude of the first VCOs, while the others are being sampled from other triggers - this way they don't change note with the sequence until they receive a new trigger.

Hope it's more clear now.
kanstraktar
damase wrote:
maybe this trick is meant to sit on each step of the sequence for a long time or to only advance to the next step as a ‘song change’...

by doing this you have just a major triad chord, yes, but you can fire off the envelopes for each oscillator in varying ways, like starting to drone on the 1st, then gradually bring in the 5th ect. ect. then eventually go to the next step for a new chord/song

its still pretty limited honestly, but i cant see how it would sound good to feed a fast moving sequence into a pre tuned chord without any way to change it... well it would be wayyy more limiting at least


Check first link in my updated post for Steevio's explanation. It actually gives him more freedom by combining varying sequences and trigger patterns. And IMHO it sounds really good. Here's some xamples:
https://www.youtube.com/watch?v=CY8lR7Hkjyw
https://www.youtube.com/watch?v=6NRoRT3fHG0
nostalghia
The Instruo Harmonaig may be the ideal solution for you:
Harmonaig on Modular Grid

Discussion thread here:First look at the Instruo Harmonàig harmony quantizer

Video:


Disclaimer-I don't own one, but I think one way to get a useful result might be taking gate out of the Harmonig into an envelope generator, the out of the EG into CV in of a VCA. Audio out of your 3-4 VCOs into a mixer, mixer out into input of the VCA. This way one EG modulates volume of all VCOs at same time. If you want different attack/decay etc times for each VCO, you'll need an EG and VCA channel for each one. Mult the gate to trigger each EG.
kanstraktar
Thanks but I think I'd prefer a micro ornament and crime for its smaller size. Although, I'd rather not buy another module grin
Parnelli
Heck that's the way I do it and it works fine for me, I'll use the Z8000 or a random source cv, buff mult it and send it to the oscillators. I usually don't even tune them to scale pitch just so long as they are in tune root, 3rd, 5th, or whatever I'm doing.

I also have a MISO I run them through for pitch offset should I go that route. I wonder what type of osc. you are using? I use 2 Pittsburgh Waveforms, a Dixie 2+ and a Rubicon, and they all seem to play well together. I also have a SE Quadnic with 4 oscillators that responds well to this.
kanstraktar
Parnelli wrote:
Heck that's the way I do it and it works fine for me, I'll use the Z8000 or a random source cv, buff mult it and send it to the oscillators. I usually don't even tune them to scale pitch just so long as they are in tune root, 3rd, 5th, or whatever I'm doing.

I also have a MISO I run them through for pitch offset should I go that route. I wonder what type of osc. you are using? I use 2 Pittsburgh Waveforms, a Dixie 2+ and a Rubicon, and they all seem to play well together. I also have a SE Quadnic with 4 oscillators that responds well to this.


Here it is: https://www.modulargrid.net/e/racks/view/973366

Not sure it has to do with the oscillators, just that it sounds weird to my ears, as each oscillator is playing a different scale.
insoul8
Do you have an audio recording of you attempting this type of setup? It sounds to me like you are going at it with the right approach.
damase
kanstraktar wrote:
damase wrote:
maybe this trick is meant to sit on each step of the sequence for a long time or to only advance to the next step as a ‘song change’...

by doing this you have just a major triad chord, yes, but you can fire off the envelopes for each oscillator in varying ways, like starting to drone on the 1st, then gradually bring in the 5th ect. ect. then eventually go to the next step for a new chord/song

its still pretty limited honestly, but i cant see how it would sound good to feed a fast moving sequence into a pre tuned chord without any way to change it... well it would be wayyy more limiting at least


Check first link in my updated post for Steevio's explanation. It actually gives him more freedom by combining varying sequences and trigger patterns. And IMHO it sounds really good. Here's some xamples:
https://www.youtube.com/watch?v=CY8lR7Hkjyw
https://www.youtube.com/watch?v=6NRoRT3fHG0


i was incorrect... to say it wouldnt sound good is the wrong thing to say...

but its a very specific sound, transposing chords like that... house music has good roots of this sound because of samplers transposing chord samples. and i love house music w00t

if youre having trouble achieving that then i would think maybe you need a buffered mult to keep the tuning tighter, possibly calibrate your oscillators so that they track closer to each other
kanstraktar
insoul8 wrote:
Do you have an audio recording of you attempting this type of setup? It sounds to me like you are going at it with the right approach.


No recordings attempting this, as it sounds crap.
kanstraktar
damase wrote:
if youre having trouble achieving that then i would think maybe you need a buffered mult to keep the tuning tighter, possibly calibrate your oscillators so that they track closer to each other


I use a buffered mult but it may be a tracking issue.
damase
also +1 the Harmonaig for your chord controller... its so nice for that purpose
insoul8
kanstraktar wrote:
damase wrote:
if youre having trouble achieving that then i would think maybe you need a buffered mult to keep the tuning tighter, possibly calibrate your oscillators so that they track closer to each other


I use a buffer mult but it may be a tracking issue.


Do you have a tuner? If it really sounds crap and very discordant, it might be worth slowing the sequence down and actually seeing what notes you are getting on each oscillator throughout and see if it what you were expecting.
BaloErets
I'll bring good news and bad news;

Yes, a lot of the problem is music theory. That is the bad news. It's one thing to tune 2 oscillators to different intervals (as the example of one Osc being tuned to G while the other is tuned to C), but it's another thing that the interval makes sense with each voltage that is being sent to them. With only 2 notes being played, each interval will make musical sense, but not all of them will make sense together. So yes it does take some theory to make it sound harmonious.

The good news is that there's another way to approach this. Tune your 2 oscillators to the same pitch. Send your sequence to a buffered mult. Send one of these straight to your Oscillators 1v/oct. Take a 2nd output and send it to and attenuator/offset, and then send that output to a quantizer. Send the quantized output to the 1v/oct of your 2nd oscillator. Now set the scale of your quantizer to the same scale as your sequencer.

You can use attenuation and offset to produce different notes for the 2nd oscillator, but they will be musically relevant, because they are using the same scale as the sequencer.
kanstraktar
BaloErets wrote:
The good news is that there's another way to approach this. Tune your 2 oscillators to the same pitch. Send your sequence to a buffered mult. Send one of these straight to your Oscillators 1v/oct. Take a 2nd output and send it to and attenuator/offset, and then send that output to a quantizer. Send the quantized output to the 1v/oct of your 2nd oscillator. Now set the scale of your quantizer to the same scale as your sequencer.

You can use attenuation and offset to produce different notes for the 2nd oscillator, but they will be musically relevant, because they are using the same scale as the sequencer.


This is what I'm currently doing, using Disting's quantizer (instead of S&H) and transposing the sequence before sending it. The problem I was trying to avoid was buying a new quantizer for the third oscillator grin
batch
For a small setup I’d recommend a uO_C for this. It’s quantizer is perfect for having everything working in key. Have your CV from Metropolis going into CV1, have all your quantizer channels use this as the input but then have different triggers for each channel so it works like a sample and hold. And you can set different transpose settings per channel. It really works well.

And if you want to save some space, swap your Metropolis for a Mimetic Digitalis which has 4 CV channels in 8HP. So combined you can have 4 CV channels with quantizers in 16 HP. Super powerful setup and very playable.

If you can have a bigger setup, the Sinfonion can’t be beat.
BaloErets
kanstraktar wrote:
BaloErets wrote:
The good news is that there's another way to approach this. Tune your 2 oscillators to the same pitch. Send your sequence to a buffered mult. Send one of these straight to your Oscillators 1v/oct. Take a 2nd output and send it to and attenuator/offset, and then send that output to a quantizer. Send the quantized output to the 1v/oct of your 2nd oscillator. Now set the scale of your quantizer to the same scale as your sequencer.

You can use attenuation and offset to produce different notes for the 2nd oscillator, but they will be musically relevant, because they are using the same scale as the sequencer.


This is what I'm currently doing, using Disting's quantizer (instead of S&H) and transposing the sequence before sending it. The problem I was trying to avoid was buying a new quantizer for the third oscillator grin


Then I'll +1 that an O_c would be a great asset in your system, and it will bring a bucketload of other useful functions to the table, no matter if you're in the original firmware or the Hemisphere firmware. You can't loose thumbs up
BaloErets
batch wrote:
And if you want to save some space, swap your Metropolis for a Mimetic Digitalis which has 4 CV channels in 8HP. So combined you can have 4 CV channels with quantizers in 16 HP. Super powerful setup and very playable.


Personally I wouldn't suggest this, but only if you really like what the Metropolis does. I regret selling mine, which I sold thinking I could find the same functionality in another sequencer that was smaller. As of now, I've only found software solutions (Reaktor and VCV).
kanstraktar
Thanks for your suggestions. Am thinking also of sending the original sequence and the transposed one to Kinks Logic, then Min & Max outputs to a Switch and its output to S&H and see what comes out of it. It will be in the same scale and given the triggers most of the time won't happen at the same time, it might not play the same note as either of the other two. Also having this third one an octave above the original. Will give it a try.
BaloErets
kanstraktar wrote:
Thanks for your suggestions. Am thinking also of sending the original sequence and the transposed one to Kinks Logic, then Min & Max outputs to a Switch and its output to S&H and see what comes out of it. It will be in the same scale and given the triggers most of the time won't happen at the same time, it might not play the same note as either of the other two. Also having this third one an octave above the original. Will give it a try.


I use both the Plog and the Sequential Switch Matrix for creating sequences from sequences. You are thinking along the good path thumbs up
cg_funk
I have the Toppobrillo Quantimator for all this....
I really like the module, it's quirky, but you get an easy 3 outputs from 1 input.

Depending on the mode, these three outputs can be either a chord or analogue shift register. There's a 'hold' input that if it's high it'll either rotate the shift register or change the chord inversion.

I've tried other ways of sequencing, and it's all too involved for me. I'd rather have 3 notes that are derived from 1 in a basic way than have to program every note of every chord by hand.
MUFF WIGGLER Forum Index -> Eurorack Modules Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group