Minimal Techno Production Techniques

[sparschaeler]

Hi All

I thought let's share some production techniques. My inspiration comes from a number of youtube vids about Ableton being able to do this and that - I love to do just the same with buzz, but much easier and faster.

OK here goes technique number 1 - good for sample based rhythm patterns:
1. Open a new file
2. Fill your entire wavetable with oneshot samples
3. Create an Utrk, randomize all notes, waves and velocities
4. Enter the effect 0B (note cut) and as an argument 0010 (or a different value, to your liking)
5. Create a Jeskola Groove 1, trigger it with 30-40 1/16 swing
6. Play
7. Repeat 3 until you get something you like
8. Record the loop
9. Delete all wave files (use ctrl+w to get to the old wavetable view - there you can select all waves at once)
A. Load your loop in Utrk, add some FX
B. Cut the loop apart and create single samples. Use them to create variations of your loop and to spice it up

the attached file is an example of how that can sound

[thOke]

cool

Buzz-0030.zip

(194.68 KiB) Downloaded 200 times

[sparschaeler]

Technique Nr 2:
Uses Samplegrids multiout to route random samples to peeradsr-controlled ch.amps - much nicer envelope than just cutting the samples, sounds smoother. plus, random level can be applied to each sgridout/ch.amp - makes a track extremely lively.

1. Load a Samplegrid Byte, a Pattern XP and at least 2 SampleGridOuts, as many cheapo amps as Outs (make sure inertia is set to 0), a peerADSR and a PeerLFO. don't forget groove1 and let it swing.
2. Load all wave slots in the wavetable
3. Create a Pattern XP pattern with the triggers and the wave selectors from Samplegrid
4. randomize all and make sure it doesn't get too messy with all the triggers (not too many at the same time)
5. assign each channel in sgrid a group
6. connect the outs to the sgrid and assign them to the groups in sgrid
7. connect each out to a ch.amp.
8. connect the amps to the master
9. in the peeradsr, set att, sus, rel to zero and the decay very short (0.5 tick), create a track per ch.amp.
A. place a trigger for every channel and every tick
B. assign each track to a ch.amp MAX volume
C. Link the PeerLFO to the peeradsr such that there is a track per track, all hinging on the same lfo with a random curve.
D. add fx, beat, etc. and keep randomizing the samples in pxp

[snowglobe]

Technique Nr 2:
Uses Samplegrids multiout to route random samples to peeradsr-controlled ch.amps - much nicer envelope than just cutting the samples, sounds smoother. plus, random level can be applied to each sgridout/ch.amp - makes a track extremely lively.

Nice, and just a quick plug for IX's work -- SG3 uses native multi-out, so no need for separate out machines.

[sparschaeler]

thanks, didn't know that. just installed and having fun

[sparschaeler]

Technique 3: How to create unique samples

1. Load a sampler with a random sample, preferrably a tiny chop of a voice or something percussive
2. add an x-verb
3. play back the sample and real-time manipulate the x-verb's delay length
4. record these noises
5. load the file into the wavetable and chop it into pieces. don't forget to normalize them.

Technique 4: Clicks+Glitches
use a simple sawtooth osc and play it back VERY low. filter and envelope the resulting crackle and add it to e.g. a bassdrum or snare. gives a nice lowfi feeling.
edit: manipulate the pitch env, try different kinds of filter. hipass is particularly nice.
edit2: combine the simple sawtooth with other waveforms to generate a less regular pattern

[sparschaeler]

Technique 5: load a sample with different sounding parts in a sampler (e.g. a hihat with very tonal and a very noisy part). Then play it back at different pitches - not too many instances simultaneously, the sound must still be distinguishable.

Then add a joachim chopper and randomize it. done. record the output if you want and add it to your sample collection.
also works well with vocals.

[syphilissauna]

OK here goes technique number 1 - good for sample based rhythm patterns:
1. Open a new file
2. Fill your entire wavetable with oneshot samples
3. Create an Utrk, randomize all notes, waves and velocities
4. Enter the effect 0B (note cut) and as an argument 0010 (or a different value, to your liking)
5. Create a Jeskola Groove 1, trigger it with 30-40 1/16 swing
6. Play
7. Repeat 3 until you get something you like
8. Record the loop
9. Delete all wave files (use ctrl+w to get to the old wavetable view - there you can select all waves at once)
A. Load your loop in Utrk, add some FX
B. Cut the loop apart and create single samples. Use them to create variations of your loop and to spice it up

wow thanks for reminding me of this trick ! as a time saver i use the sample pool functionality ..

[sleepzzz]

Some cool tips in here. I'll definitely try these, since I'm becoming more interesting in making more abstract sounds and doing more Minimal Techno these days. Thank you!


Backwards Verb Recipe:

1. Sample/record a short generator note or two or use a sample you have already.
2. Reverse the sample in the wavetable.
3. Put the reversed sample in a sampler and add a (preferably long) reverb of your choice.
4. Play and record this again into its own WAV and put back into the wavetable.
5. Reverse the newly recorded sample!

Now you have a backwards-building reverb that ends with the original gen/sample playing forward. You can experiment with the length of the reverb and the thickness/thinness to make big build ups or thin subtle rises. A short tail might be good for weird drum effects. Don't forget to use predelay for certain sounds!
Also, if you can't figure out how to match the reverb and the note to the proper place back into the track, just cut the sample into two, one for the reverb and the other for original sound and mash them how you want in the tracker.

[sparschaeler]

nice one! I also started using looped fractions of differently coloured reverb tails to create lively background noise

[mute]

This isn't really about minimal, but I thought I'd chime in with a reverse/reverse reverb tip.

In many buffered (FFT, like Joachim Neptune) or delay based pitchshifters (like http://www.plosive.net/vst/#pitchback) you can reverse your audio by using large buffer sizes/windows or negative pitch values. In FFT effects, the output will be delayed by the # of samples in the buffer so you would also need to combine it with a wet only delay on the dry signal to place the dry sound after the FFT/Delay/Pitch reverse. Feed this fx setup with a reverb or dry signal and you can do realtime reverse effects that typically require post editing or non-realtime destructive editing (samples, etc). It takes some practice, and experimenting with different effects to get the sound you want, but it is there. I use it alot. There are also alot of normal delays that can reverse the delayed signal, but usually they are short taps instead of a long delay... but some, I think for example Unweildly Delay by Fuzzpilz, can achieve this. BlockFX is good for this as well, but it can be pretty complicated so I won't go into it.