Making Magic Delay And Verb

A while back I was interested in finding more realistic dub twang-delay sounds in VST form, which was discussed in this thread. As far as I am aware, the authentic spring-twang sound is currently impossible to re-create in software, as it relies on having a real spring and something to twang that spring - imagine trying to emulate those physics with maths!

So I set out to try and fake the sound somehow using combinations of VSTs. Predictably, I couldn't exactly get it, but during experimentation I developed some processes that really improve the 'wet' characteristic of delays and reverbs, especially with stiff repetitive input sounds. Here's a summary:

Firstly, I really recommend EchoLIVE as a delay plugin, it's probably the best free plug I have used for getting lush authentic analogue style dub delay. In conjunction with that I nearly always use Moneo for pre/post width and/or channel swapping.

What follows isn't dependent on those two plugs, in fact just using Renoise's native effects you can do wonderful things. To do this you have to use a send channel - in the 'send' chain we are just going to effect the 'wet' characteristic of the sound. So if you're using the native delay, click 'mute source', and in the mpverb mix the dry fader to 0%.

Now try some of these send chains:

- Phaser ---> Delay
- Flanger ---> Delay
- Phaser ---> mpverb
- Flanger ---> mpverb

...And so on. The idea is to modulate the incoming signal in a way that makes the 'wet ambiance' echo out naturally, but uniquely each time there is an input sound. So even a stiff snare drum sample, for example, that is used all the time at the same pitch can have a echo or verb behind it that doesn't sound stiff and machine like. This can make very 'cheap sounding' reverb plug-ins sound quite usable!

Naturally, each composer will prefer a certain sound governed by what they set the parameters for each plug. However I have a few recommendations. My stock standard usage usually ends up look like this (depending on the type of input sounds and the song's mix):

For Delay:
- SEND 1-50% ---> Phaser (500hz-3khz crossover, medium LFO, 50% effect, 0-25% feedback, 0-3% width) ---> LP Filter (Moog No Rez, 1-9khz) ---> Delay (all wet, length+feedback suitable to mix and instrument expression) ---> HP Filter (Moog anywhere between 3-12khz) ---> optional Phaser (lite) ---> Width ---> SEND 1-10% to Verb Channel.

For Reverb:
- SEND 1-50% ---> Phaser (500hz-3khz crossover, medium LFO, 50% effect, 0-50% feedback, 0-3% width) ---> LP Filter (Moog No Rez, 1-9khz) ---> mpVerb (all wet, length+feedback+pre+cpu-quality suitable to mix and instrument expression, cutoff 2-5khz) ---> HP Filter (Moog anywhere between 3-12khz) ---> optional Phaser (lite) ---> Width ---> SEND 1-10% to Delay Channel.

See the example XRNS for how these sound.

Notice that I'm using a pre-LP-filter and a post-HP-filter either side of the 'wet ambiance' effect - this helps to narrow the echo into a sonic-region that sounds more authentic, architectural or distant. There's nothing worse, for example, than an overly bright reverb sounding cheap; or, a delay or verb mashing up a mix with muddy kick drum and bass sound. Ambiance, especially when it's long and deep, tends to speak best between 500hz to 12khz, but many genres, especially dub, get narrower than that again. Your tastes will reveal themselves with experimentation.

There are other optional and creative things you can obviously put in front of the delay/verb to get astoundingly cool spatial sounds. Easy good ones are: Leslie, Pitch modulation, Stretch, Buffer mashups, Reverser, Long delay loops, slow chorus, Formant filters, or anything that has workable modulation in it. You're basically limited by your imagination and the plug-ins you have at hand!

I'm keen to see if anyone else is using this technique or anything similar to it, so post back here if you've tried it out. Otherwise, enjoy putting some 'magic delay and verb' into your songs.

Category: 

2 comments

And dblue glitch on delay sends is a must.
I don`t like the idea of using built in effects. A short visit to dsp coding forums tells me how aproximative effects are implemented in non standalone packages.