Swells, sweeps, transitions, reverse reverbs, delays............?

[Stamped Records]

I'm seeking some discussion on this topic. It's a difficult one to research because of the lack of consistent terminology that crosses the boundaries of genre and style.

How do you design your swells?

1. Source Audio - percussive, chordal, pluck, bass?
2. Reverse the source before, after, or before and after reverb/delay is applied?
3. Choice of reverb?
4. Reverb settings?
5. Post production on the designed swell?
6. Swells grouped for post production?
7. Resampling swells?
8. White noise?
9. Layering?

Edit: in musical terms, what I'm talking about is "transition effects", but "swells" sounds better in refering to music and it is more of a pointer or guide for sound design than it's "transition effect" counterpart IMHO. The word swell kinda says dynamic, wheras the term transition effect doesn't really give you a point of reference for what it is.

[Kwurqx]

Too many variables....

[Stamped Records]

Too many variables....

I'm guessing there aren't that many people who get practised and versed in the successful production of swells. It would seem to be the end point of the ultimate preset dodger and there can't be too many of them around, unless in the movies or something, right?

How many people are producing professional tracks from intitialised presets on ALL of their equipment?

[sqigls]

i don't release a lot personally, although i have a lot of personal projects.
but i produce a lot

Swells and transitions are actually one of the few tricks I don't often spend much time with.
but that's my new years resolution.
exercise, less smoking, and more swells.

[Stamped Records]

i don't release a lot personally, although i have a lot of personal projects.
but i produce a lot

Swells and transitions are actually one of the few tricks I don't often spend much time with.
but that's my new years resolution.
exercise, less smoking, and more swells.

Cool man, I'd say I'm pretty much in exactly the same boat although I do plan to start releasing and that's part of my new year's resolution.

What kind of music do you make?

[sqigls]

i don't release a lot personally, although i have a lot of personal projects.
but i produce a lot

Swells and transitions are actually one of the few tricks I don't often spend much time with.
but that's my new years resolution.
exercise, less smoking, and more swells.

Cool man, I'd say I'm pretty much in exactly the same boat although I do plan to start releasing and that's part of my new year's resolution.

What kind of music do you make?

some of my 'electronic' projects revolve around PsyDub... Electronic dub reggae... Crunk... Minimal psychedelic progressive tribal techno... goatranch... breaks meets ambience meets house... dub to bass glitch music... deep techno from dub to psy-ish... minimal bass music...

[disquieter]

I like this topic. I'll just toss out a few ideas that I've had about swells in the past. I'd also like to see what other people are doing.

Personally I like reverse delays more than reverse reverbs, as I think it's difficult to do anything new with reverse reverbs. Reverse delays aren't washed out so you maintain more rhythm. You can do odd timings and have stranger poly rhythms before the drop. You can play with the more degraded sounds after they've run through a filtered, saturated feedback loop for twenty mins, resample it and see what you can do with it for a swell/transition.

I think if you are going to do a more washed out reverb, you could play with a stepper multi stage lfo and fade it in or out as you get closer to the drop. Granular synths are great for that sort of thing too, especially the ones that specialize in drone-ier sounds.

[bearstakemanhattan]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

1. I usually use almost anything tonal, and in general I like to use the first tonal element that appears in the section into which I am trying to transition. For example, If I'm moving form an intro into a verse in a track with vocals, I'll probably use the first note of the vocals. If it's a buildup section in a festival EDM song I'd be more likely to use the first supersaw chord, and so on.
2. The short answer is before and after. The long answer is that I use heavy reverb beforehand, then I bounce to audio and reverse, and lastly I might add an additional reverb tail (more within reason) in order to make the end of the swell not end so abruptly, although sometimes I want that abrupt ending.
3. Feel free to use any type of reverb you enjoy in terms of brand. I personally use Valhalla Room for this kind of sound design-y stuff. In terms of reverb type I'd stay away from spring verb and maybe convolution verb if you want that more class swelling sound, but the latter two are totally game if you're feeling adventurous.
4. A really long tail/decay setting (maybe in excess of like 10 sec?). If your reverb plugin has control over the ER vs LR balance then I'd set that to mostly late reflections
5. After I finish bouncing and reversing, I might add some normal reverb like I mentioned. I also chop off the very end of the sample (the crescendo), for a more concise ending because a 100% wet reverb softens the attack of my source. I also shorten the overall length further to fit it into my track, and I add a fade to the beginning if I want more of drastic swelling effect. That's what I do at a base level but you can take that and run with it.
6. I usually group swells into an FX buss with my other transition elements and atonal stuff (white noise sweeps, random noises, and such)
7. I mean I had never thought of resampling a swell but it's definitely an idea. Maybe give it a try if you're bored?
8. I wouldn't really try to create my own white noise swells because there are so many sample packs with premade ones.
9. Layering a bunch of swells is how EDM buildups work. Also yes, layering can get you a really unique-sounding riser.

Lastly, here (https://hyperbitsmusic.com/5-creative-composition-techniques-using-reverses/) is a great resource for applying those reverses. I personally found those tips to be quite helpful.

Hope this helps!

[disquieter]

3. Feel free to use any type of reverb you enjoy in terms of brand. I personally use Valhalla Room for this kind of sound design-y stuff. In terms of reverb type I'd stay away from spring verb and maybe convolution verb if you want that more class swelling sound, but the latter two are totally game if you're feeling adventurous.

I'm curious why you tend to stay away from convolution verbs?

[DJ Warmonger]

I always design riser for every track from scratch, possibly a number at a time. Usually just pick rising envelope or filter with some LFO, depending on what the synth allows. Certainly you can get more movement with Serum or Harmor that with simple synth like LuSH-101.

Another great technique I discovered recently is inverted kick with reverb - Etasonic style. Does have strong impact without cluttering the mix.

Thanks for the tips, BTW, may try more adventurous tricks in future projects

[Stamped Records]

You can play with the more degraded sounds after they've run through a filtered, saturated feedback loop for twenty mins, resample it and see what you can do with it for a swell/transition.

I think if you are going to do a more washed out reverb, you could play with a stepper multi stage lfo and fade it in or out as you get closer to the drop. Granular synths are great for that sort of thing too, especially the ones that specialize in drone-ier sounds.

Interesting, I'd love to hear some elaboration on each of these paragraphs - a signal flow example for my mind if you will.

1. I usually use almost anything tonal, and in general I like to use the first tonal element that appears in the section into which I am trying to transition. For example, If I'm moving form an intro into a verse in a track with vocals, I'll probably use the first note of the vocals. If it's a buildup section in a festival EDM song I'd be more likely to use the first supersaw chord, and so on.
2. The short answer is before and after. The long answer is that I use heavy reverb beforehand, then I bounce to audio and reverse, and lastly I might add an additional reverb tail (more within reason) in order to make the end of the swell not end so abruptly, although sometimes I want that abrupt ending.
3. Feel free to use any type of reverb you enjoy in terms of brand. I personally use Valhalla Room for this kind of sound design-y stuff. In terms of reverb type I'd stay away from spring verb and maybe convolution verb if you want that more class swelling sound, but the latter two are totally game if you're feeling adventurous.
4. A really long tail/decay setting (maybe in excess of like 10 sec?). If your reverb plugin has control over the ER vs LR balance then I'd set that to mostly late reflections
5. After I finish bouncing and reversing, I might add some normal reverb like I mentioned. I also chop off the very end of the sample (the crescendo), for a more concise ending because a 100% wet reverb softens the attack of my source. I also shorten the overall length further to fit it into my track, and I add a fade to the beginning if I want more of drastic swelling effect. That's what I do at a base level but you can take that and run with it.
6. I usually group swells into an FX buss with my other transition elements and atonal stuff (white noise sweeps, random noises, and such)
7. I mean I had never thought of resampling a swell but it's definitely an idea. Maybe give it a try if you're bored?
8. I wouldn't really try to create my own white noise swells because there are so many sample packs with premade ones.
9. Layering a bunch of swells is how EDM buildups work. Also yes, layering can get you a really unique-sounding riser.

Lastly, [url=<span class="skimlinks-unlinked">https://hyperbitsmusic.com/5-creative-c ... rses</span>/]here[/url] is a great resource for applying those reverses. I personally found those tips to be quite helpful.

Hope this helps!

Thanks man, appreciate your time, good stuff.

It seems we're all going down the road of creating the effects from elements in the track, it would definitely seem to work more favourably using tones from the track, but I see no reason I wouldn't exploit the oddity of using something that's not in the track as a bass for an effect in future, even if it's only one "special" effect per track.

What I do now, just to attempt to prevent my mind and my project from becoming overloaded with reverb settings. I use a single reverb setting for all of the swells that I create. I just try to get an interesting reverb setting, while imagining it in reverse, therefore, the detail in the early reflection is what I give most attention to, the peak of the riser when reversed. I use Bitwigs native reverb mostly but I'm startying to use a single Aether per project for the specific purpose of transition effects. I then use exactly the same reverb as the "big" send reverb for my track.

I do tend to run the effects through a bus and add some form of modulation to group them to the ear and give some movement to them. Compression depends on the sound but usually it's just hard compression, no fancy settings.

I'm finding that I often need to roll off the highs on them - the cutoff point varies from 1khz to 5khz but not usually much higher. Does anybody else find the need to roll off the highs? Perhaps this is just a matter of taste, but reverb does have a lot of sparkle.

I also send the effects, either individually or via the bus, to all the sends in my track as needs be, chorus, multiple reverbs, distortion, phaser, flanger, whatever sounds good.

[Kwurqx]

3. Feel free to use any type of reverb you enjoy in terms of brand. I personally use Valhalla Room for this kind of sound design-y stuff. In terms of reverb type I'd stay away from spring verb and maybe convolution verb if you want that more class swelling sound, but the latter two are totally game if you're feeling adventurous.

I'm curious why you tend to stay away from convolution verbs?

Not sure about other peoples motivations (since they're not provided). But.....I can have a guess.

Generally Convolution reverbs are meant to replicate the actual acoustical/reverberation properties of an actual space (or machine).
https://en.wikipedia.org/wiki/Convolution_reverb

Not very "tweakeable" in a strict sense. Think of it as "sampling". It's very useful if you want an exact replica/reproduction. But it has many limitations since you would need individual impulse responses for any change in variables (like change of room, delay time etcetera).

Many Convoltution reverbs START with an impluse response and convolution, but offer many tweakable variables (stepping away from actual "convolution"/"sampling").

A bit like how you can use samples as oscillators in a sampler/synth. But you would need wave tables to emulate dynamics/mudulation that occur when variables change. And you can use that as input for further processing like pitch and amplitude envelopes and filtering/EQ.

So, from a creative point of view algorithmic reverbs offer more flexibility, especially in creating "unnatural" (huge and modulating) reverbant spaces.

Personally for this type of reverbs I'm an absolute fanboy for Valdemar Erlingsson's FREE Cloudseed project. This thing can create very natural and open sounding reverbs of any size. As well as totally unnatural chaos of any size. Ultrawide with clear centers is so desired.
http://valdemarorn.github.io/AudioProjects.html



But there are many more options in Reverbs.....
https://caulixtla.com/JP-08/vst-reverbs.html

[disquieter]

I gotcha. I use convolutions a bit and I’ve enjoyed using samples as impulses (though often it doesn’t get the results you’re hoping for.

Recently I bought Lexicon PCM verbs which are gorgeous and definitely give me the sort of tweak ability that I’ve wanted. The main issue is that you can’t demo what it sounds like reversed until you.. well, reverse it.

To get around this I made a small Reaktor project where you can listen to a reversed signal 1 bar after you’ve made your tweaks. Though it’s kinda hard to control.

[donkey tugger]

Interesting topic.

I''m probably coming at it from a different angle than most (indie rokk for the most part) but the principles are the same, if not the sounds used - usually guitar, ebow or vocals .

No point in repeating all the above good tips. What I'd add is that I generally spend a fair bit of time automating modulation/reverb/volume effects/stereo width over the course of a swell - e.g the level of vibrato and tremelo will rise and width increase as you get to the crescendo..or fall indeed. Sometimes I resample, and the resample again

[disquieter]

Interesting topic.

I''m probably coming at it from a different angle than most (indie rokk for the most part) but the principles are the same, if not the sounds used - usually guitar, ebow or vocals .

No point in repeating all the above good tips. What I'd add is that I generally spend a fair bit of time automating modulation/reverb/volume effects/stereo width over the course of a swell - e.g the level of vibrato and tremelo will rise and width increase as you get to the crescendo..or fall indeed. Sometimes I resample, and the resample again

I had a listen to yer stuff. You're pretty talented man. I'm curious how you got the sound @2:00 mins in your track "A Nervous System". Sounds like an ebow-ed guitar but the tone has a thin almost viola sound to it? I've tried to get a more glitchy version of that. Kinda like the sound of the sputtering guitar in beginning of the Sigur Ros song "Saeglopur". The one that happens just as the massive piano hits. The closest I've come to that was with waveshaper distortion.