Twitter   Patreon WIGGLING 'LITE' IN GUEST MODE

 Goto page 1, 2  Next [all]
Author how to slow clock rate by a 1-2 % points with minimal setup?
MrTurboparrot
 Hi All, I started this thread on the eurorack section, but am widening it so that I can understand the process more clearly. It's here, FYI: https://www.muffwiggler.com/forum/viewtopic.php?t=202783 Basically, I'd like to do some Steve Reich kind of stuff. I though I could do this with a Pam's, but the ratio of clocks /25 and /24 on two sequences provides too much of a rapid phasing effect. It's a 4% difference. From some research, I gather the timing difference needs to be much lower - more like 1%, or even less. Given that Pam's jumps-up in stages in divisions, I cannot have /99 and /100 (or similar). I tried to do it with maths EOR and EOC - but it means using up both channels, and is also a bit of a compromise because it is difficult to get accurate. I tried to slew the clock rate using maths, but essentially this is only a delay right? Not even sure I did that right in the end. If it helps, my rack is below: https://cdn.modulargrid.net/img/racks/modulargrid_570356.jpg Anyone got any ideas? cheers
authorless
 What do you mean "tried to slew the clock rate with maths"? Slew limiting a CV controlling clock rate would only have an effect if the CV is changing. If you are slew limiting a stream of clock pulses, that is only going to affect the edges of the clock pulses, it isn't going to change their rate it would just give each pulse a bit of attack and/or decay like an envelope. If you are trying to use the Maths as a clock delay, that is going to keep both clocks at the same tempo but one would be delayed.
cretaceousear
 two clocks
MrTurboparrot
 cretaceousear wrote: two clocks

hmmmm. this is actually not a bad idea!

In the meantime, can anyone help with my issue? I'd rather not buy a new clock device, and see if I can find a neat way to do it with what I have.

Thanks for the guidance Authorless. I forgot that about the slew function in Maths.
Sinamsis
 Two LFOs that have a sync input and just dial it in. Ha might be a pain in the ass. But I bet you could do it with Batumi for example.
authorless
 You could have one clock from the Pamela's Workout and use the end of attack or end of cycle output of a channel of maths as a second clock.
hinterlands303
 Yeah two clocks all the way - and as authorless pointed out you already have two (three actually - Pam's, Maths, and Wogglebug). It will definitely take some dialing in though. How about this: Use the EOC output of a cycling envelope of Maths Channel 4 to generate your clock. Temporarily sync Pamela's workout to this clock to determine Maths BPM. Then unplug the maths clock and change the Pamela's BPM to be slightly off Maths tempo, If you only want a 1% difference this will probably work better with tempos over 100 BPMs. Be careful not to bump any of the knobs on Maths channel 4 because you'll throw the whole thing off!
Graham Hinton
 MrTurboparrot wrote: I cannot have /99 and /100 (or similar).

What you are asking for requires more resolution and stability than you will normally find in hardware tempo generators, but there is another way if you are willing to use a DAW.

Use the Sync function in wavtones to create two files at the different tempos you require with up to 5 decimal places of resolution. Repeat each file end to end along its own track in your DAW to make two tempo tone tracks. Route each track to a separate output and use as a x24 clock with your modules. If you need x1 clocks you can either use hardware dividers or divide the tempo setting by 24.

This problem requires a mathematical solution, not a Maths solution.
Carrousel
 Might be tough, but like Sinamsis says: i'd have thought the best way of doing this would be to use square wave outputs from a VCO / LFO as the clock, applying precision CV, possibly from a sequencer, via a slew, to the freq input of the VCO to obtain the tempo shifts that you want Edit: obviously this can't be 'synced' to other stuff.....unless you mult out the square wave and find a way to send that in a meaningful form to all the instruments you're using. Could be great. Edit 2: just seen that your requirements are slightly different from what i understood first time. Still, two VCOs / LFOs will get you there with patience i think
MrTurboparrot
 hinterlands303 wrote: Yeah two clocks all the way - and as authorless pointed out you already have two (three actually - Pam's, Maths, and Wogglebug). It will definitely take some dialing in though. How about this: Use the EOC output of a cycling envelope of Maths Channel 4 to generate your clock. Temporarily sync Pamela's workout to this clock to determine Maths BPM. Then unplug the maths clock and change the Pamela's BPM to be slightly off Maths tempo, If you only want a 1% difference this will probably work better with tempos over 100 BPMs. Be careful not to bump any of the knobs on Maths channel 4 because you'll throw the whole thing off!

This is really helpful - thanks Hinterlands303. This way I can get it quite detailed in terms of actual applied BPM.

Other ideas - i.e. sync, are helpful too - but the accuracy of the BPM approach is great.

Thanks everyone!
cretaceousear
 The other way I thought of but needs more esoteric gear is a master clock running in - I don't know - maybe not megaHertz but lotsaHertz.. and then two fancy dividers that get down to the ticks you want. Probably best accomplished with an Arduino or similar. Me I'd just use two LFOs and enjoy the sloppy drifiting. Still beats me how the Steve Reich guys'n'gals can play those timings...
Rex Coil 7
 cretaceousear wrote: Still beats me how the Steve Reich guys'n'gals can play those timings...
It's known as "Agenesis of the Corpus Callosum" .. the two halves of the human brain did not become connected by the thing in our buckets that does so is known as the Corpus Callosum. The Corpus Callosum is the part of the human brain that connects the two hemispheres of the brain together. This "deficiency" (if you will) permits those particular musicians to be mindful of two different tempos at once. This allows them to create that wonderful evolving sound by playing the same pattern with each hand played at ever so slightly different tempos.

"Agenesis" means "not formed".

Hence the condition is called "The thing that connects the two hemispheres of the human brain did not form" ..... or ..... "Agenesis of the Corpus Callosum".

Playing that dual tempo pattern trick is impossible for humans with a normally developed Corpus Callosum.

If you want it bad enough, you can have your Corpus Callosum taken out ... there's always some surgeon willing to do just about anything in places like Brazil (money speaks every language).

I wouldn't recommend it.

Regarding the mechanics of duplicating the trick through the use of a machine, good luck using some analog module. As someone already mentioned, using a DAW with fine enough tempo resolution is your most likely path to success.

Rex.
cretaceousear
 Thanks RexCoil - very interesting - I've long been interested in the brain so familiar with the corpus callosum's role. They used to cut it in cases of severe epilepsy didn't they.. So not everyone can play Steve Reich parts? There was a nice 3 piece BBC TV doc on minimalism a few months ago - discussed here on MW - and the performers were saying it was tricky playing separate timings. Though that was per person - not one person playing different times over left and right hand.
Sinamsis
 Ha he’s kidding. Agenesis of the corpus collision comes with other midline defects that can be debilitating to say the least. Including possible bilateral under-development of the optic nerves. Haha. I don’t know much Steve Reich but yeah polyrhythms can be very difficult. I can’t play them. But listen to a Messhugah album. I’m pretty sure their drummer has a corpus callosum. And I believe you’re right about certain forms of epilepsy. They also might destroy the temporal lobe in other forms. Ha but it’s not a benign thing.
cretaceousear
 Damn - I believe everything - maybe I'll just opt for a routine trepanning to improve my motor skills.
MRoyce
MrTurboparrot wrote:
 hinterlands303 wrote: Yeah two clocks all the way - and as authorless pointed out you already have two (three actually - Pam's, Maths, and Wogglebug). It will definitely take some dialing in though. How about this: Use the EOC output of a cycling envelope of Maths Channel 4 to generate your clock. Temporarily sync Pamela's workout to this clock to determine Maths BPM. Then unplug the maths clock and change the Pamela's BPM to be slightly off Maths tempo, If you only want a 1% difference this will probably work better with tempos over 100 BPMs. Be careful not to bump any of the knobs on Maths channel 4 because you'll throw the whole thing off!

This is really helpful - thanks Hinterlands303. This way I can get it quite detailed in terms of actual applied BPM.

Other ideas - i.e. sync, are helpful too - but the accuracy of the BPM approach is great.

Thanks everyone!

This will work without even changing the BPM on the Pams due to quantization errors. Just tested this out with an LFO and a Disting, after removing the clock to the disting the two LFOs phase over ~2 minutes until they are back in sync. Definitely gets into Steve Reich phasing territory.
Rex Coil 7
 Sinamsis wrote: Ha he’s kidding.
No, I wasn't.

Extraordinary people are often so because they have differences over the "normals".

For example, note how many totally blind people there are that excel as musicians. Enviably so.

Some of us are gifted with differences, even though most times those same differences are debilitating disabilities. Sometimes, just the right amount of ~different~ can be honed into the extraordinary.

StateVariable
 Yeah, I've heard of people breaking & re-setting their fingers to get round difficult piano parts but messing with your corpus callosum would be a whole 'nuther level... I've done this with a DAW way back in the day too (if cubase even counts as a DAW any more) but I'm really interested in the Arduino idea. It sounds easy. Drifting LFOs is fine but it'd be nice to dial in a ratio and have the parts lock back into sync exactly every few dozen cycles, or at the end of the piece or whatever. Might give that a go sometime.
Rex Coil 7
Knights Who Say Neve
 This should be possible to do with a delay that has voltage control over delay time, and a slow lfo. Try panning the original hard left and copy hard right 100% wet - it will start out in the center, and then split off into stereo as it goes out of phase.
Sinamsis
 Well I happen to be a surgeon, and I have treated patients with agenesis of the corpus callosum. Agenesis means complete absence not underdeveloped or slightly abnormal. Look up septo optic dysplasia. It can be associated with severe hormonal abnormalities that can be fatal when the pituitary fails to develop as well. It is a very rare condition and I assure there’s no ensemble of musicians that have agenesis of the corpus callosum. Regarding surgeons that will do anything for money. I’m an oculoplastic surgeon. I do a fair amount of cosmetic surgery. I can honestly say I have never done anything that would jeopardize my patients overall health or put them through unnecessary risk. I think you might say some people will do anything for money. Some of those people may be surgeons, or doctors in general. But as a subset of the population I can guarantee that we are not outliers, at least not in the direction implied. If anything I’d like the think that a doctor would be less likely to do something unscrupulous compared to other professions. Doctors for the most part make a comfortable living, and you don’t go into medicine to make millions. If that’s your priority there are other fields that make that a much more feasible goal.
batch
 There is a video of the ER101 doing this where you create a sequence on one track then copy the sequence over to another track then reduce duration of one step by 1.
JohnLRice
 A clock/trigger/gate delay maybe?
MrTurboparrot
 Armstrb wrote: There is a video of the ER101 doing this where you create a sequence on one track then copy the sequence over to another track then reduce duration of one step by 1.

yes - i've seen (and don't quite understand) this.

Essentially, the gate time is reduced a fraction on the last step so it rolls over every bar I guess. Not quite Steve Reich, because the shift would need to be continuous.

However, I still cannot see myself being able to do that with my rack (see above for link). I suppose I could insert a pulse into the x-mod on Rene, to get one step to skip forward a touch faster than the others, but then i'm back to relying on PAms divisions...
timoka
MrTurboparrot wrote:
 Armstrb wrote: There is a video of the ER101 doing this where you create a sequence on one track then copy the sequence over to another track then reduce duration of one step by 1.

yes - i've seen (and don't quite understand) this.

Essentially, the gate time is reduced a fraction on the last step so it rolls over every bar I guess. Not quite Steve Reich, because the shift would need to be continuous.

However, I still cannot see myself being able to do that with my rack (see above for link). I suppose I could insert a pulse into the x-mod on Rene, to get one step to skip forward a touch faster than the others, but then i'm back to relying on PAms divisions...

that's right, with the er-101 method you don't have a continuous shift, it shifts a bit after every pattern, if that pattern isn't that long it still feels like a continuous shift. here's an example of this phasing:

[s]http://soundcloud.com/uzala/fm-phase[/s]
it's quite a fast shifting here since i didn't use such a fast clock, the faster the clock the more subtle the shifting(by subtracting one clock pulse from the last step on pattern2, so if you use an oscillator as clock one clock pulse is almost not perceptible, but after time you get phasing).

looking at your setup i would use one clock from pams for sequence one, mult that clock into maths channel one, set it to exponential with minimal attack and decay and then use eor out as your second clock. now attenuate a slow lfo into the rise cv in and the clock should shift slowly...
 Goto page 1, 2  Next [all] Page 1 of 2