Another one? So how is this one different than all the languages that do this (like Chuck or Overtone) AND the libs for C++/all other languages that generate audio as well? The original THX sound was generated in C.
I would be inclined to use a language that had more signal processing built in, ie Mathematica:
That wasn't what was implied. Usually if a project isnt named "Johns treehouse" and instead "X: The Ytinator" you might care if there is a bit of hyperbole or there really is actual scene person who decided to improve upon the current landscape given their breadth of knowledge. As oppose to John who doesnt know much about woodworking but likes creating bullshit with "made by John" carved into it.
If i understand what they are doing then it's not really equivalent as this seems to be a DSL for transforming textual patterns into music, rather than a general purpose music language. While you can dive under the covers it's not required. Getting to this point in Overtone would be some work, so here you can just get started playing with it. Chuck i have less experience with enough to think the same thing applies.
The pedant in me already observes that any programming language is "a DSL for transforming textual patterns into..." but i hope my sense in representing the difference is clear enough.
That's exactly what this is - it's a small DSL that evaluates to create triggers at moments in time. These triggers are then evaluated by an endpoint (like the sampler included with tidal) to make sound.
I'm in a related field (I do a lot of signal processing in Pure Data), and I also keep an eye on the live coding scene. You're absolutely right here in that Tidal isn't about signal processing, it's about a flexible, creative, and expressive way to issue musical commands as text. It isn't so much an audio programming language, as it is a musical programming language.
Other languages such as Chuck, Supercollider, Pd, or Max are much better suited to the signal processing side of things (though there are certainly plenty of pained souls who have done plenty of musical work in these languages).
Edit: I can safely say that if you're planning on getting into live coding, Tidal is the way to go. I wouldn't recommend anything else. There be dragons.
Yes it's for exploring patterns. It's not unique in this (cf supercollider patterns, foxdot, isobar, gibber patterns, ixilang etc) but is fairly unique in its purist focus on pattern.
Yeah... most musicians (i.e. "creatives") aren't coders. They prefer high-level GUIs (that strangely always look like expensive vintage analogue gear). And most coders aren't musicians.
If you get to see someone perform live coded music as I did at EMF 2014 (Yaxu (Alex McLean) it's a blast, watch in real time as the code is edited, glitches out then drops back in, if the coding is projected you can predict where the music is going :) I could not find the EMF show I saw, but this gives a feel https://www.youtube.com/watch?v=3HXcb5_RuNg
For beginners(assuming some programming knwoldge), what is the easiest way to improvise electronic music: live-coding ? or or a keyboard based software studio application, like fruity loops ? or are there any others ?
I don't think there's a right answer, but I think it might come down to budget. If you want to pay $1000 for an off-the-shelf drum machine, sequencer, or synth, modern musical hardware is pretty easy to use and you can go far. However, live-coding software is free software and open source. As for ease-of-use, I think it really depends on the person. If you have some programming experience, then a live-coding musical environment might be fairly easy. It's a different set of challenges.
A physical device such as a "groovebox" might be interesting for you, they include sequencing, drum machine and synth in one box, much like teenage engineering's OP-1 (which is expensive and modern). One I'm looking at getting is the Roland MC-505. Some people will say they sound this or that, but listen to some demos people have put on youtube and decide for yourself. Think they go for less than £300, and you can make a whole album on one.
If you have an iPad, GarageBand for IOS is an absolutely incredible start for that kind of thing. Or Korg Gadget, or UVI Beathawk. They act like the physical ones but have more features and cost way less--you've already paid for the hardware.
I've given Sonic PI multiple good tries. I really have. TidalCycles is much more of a natural fit for me (which is an emphasis on rhythm and time manipulation rather than chords/melodies/synths). To each their own. Sonic PI resonates with a lot of people.
Related question: Are there any nice videos demoing this kind of music artistry in a different genre? I'm thinking of stuff like https://www.youtube.com/watch?v=yY1FSsUV-8c or maybe something with vocal sampling?
Tidal (they added cycles when the Tidal music service came afterwards) is great fun. I was hoping to get Hylogen working with it.
Hylogen is an EDSL for livecoding shaders in Haskell [1]
My weapon of choice is Extempore by Andrew Sorensen [1]. Dual languages in one system available for blistering speed in livecoded graphics and sound/music.
Andrew Sorensen's demo makes me wonder if classical music could be "decompiled" as a series of recursive functions, giving us a clear conceptual explanation of what is happening. Then, we could reverse the process to generate similar music, or explore variations.
The ideas "Cyber-physical programming" around extempore seem interesting!
Livecoded visuals/sounds is something i have been doing with a livecoding tool that i develop where a graphical canvas is actually both, the audio spectrum and the visuals, all driven by a single GLSL fragment shader : https://www.fsynth.com
Please articulate, Civilly, which guideline I Broke that other users don't break everyday and I will gladly change my ways, otherwise don't articulate my perceived shortcomings at all. If you embue me with the power to link to peoples comments I will find all shenanigans and post them for you to comment this message to.
this reminds me of Bret Victor's talk, "stop drawing dead fish", which takes a similar concept of live-performance programming and applies it to the world of animated storytelling:
No, I think it has always been "a language for live-coding pattern". I might agree that it is a bit unclear. Music is the the most common use case, but it can do a lot more: https://www.youtube.com/watch?v=TMrv_wEFdqE
Tidal has a MIDI add on which makes playing with MIDI instruments quite fun! I get patterns of notes playing and then fiddle on the knobs of the synths. So much fun :-)
I have been using Tidal for my live show (and more recently composition) for a few years and have played at some pretty big clubs in LDN, LA, etc. It's been a revelation and I will not be going back to DJing or 'controllers' any time soon.
I'm interested in how you're using it for composition. There are some things I feel I'm missing. I'm not so much interested in creating noise with it; that's never been my thing. But I'd love to see some examples of how you're using it for composing, if you could share.
(FWIW I have watched and enjoyed all Bret Victor's videos, and am looking forward to his big reveal. I've also experimented with visual front-ends to tidalcycles, so agree with you to some extent.. But text is really good for describing pattern.)
I know this sort of goes against the live coding thing, it is it possible to do a hybrid sort of thing where you use frp to manage inputs from sequencers?
59 comments
[ 2.3 ms ] story [ 117 ms ] threadI would be inclined to use a language that had more signal processing built in, ie Mathematica:
https://reference.wolfram.com/language/guide/SoundAndSonific...
https://reference.wolfram.com/language/guide/SignalProcessin...
The pedant in me already observes that any programming language is "a DSL for transforming textual patterns into..." but i hope my sense in representing the difference is clear enough.
Other languages such as Chuck, Supercollider, Pd, or Max are much better suited to the signal processing side of things (though there are certainly plenty of pained souls who have done plenty of musical work in these languages).
Edit: I can safely say that if you're planning on getting into live coding, Tidal is the way to go. I wouldn't recommend anything else. There be dragons.
(I have yet to run into anyone else that's heard of the thing I linked, and I think it's really awesome.)
I think you're being downvoted for negativity, but you make a fair point.
The THX logo is a classic of sound design made entirely with code.
Given all the newer langs, it should be easy to do better.
Oddly, this doesn't happen. It's very rare that anyone makes non-toy music with these langs, never mind creating an acknowleged classic.
There may be reasons for this.
Sonic Pi: http://sonic-pi.net/ On GH: https://github.com/samaaron/sonic-pi Aerodaynamic: https://www.youtube.com/watch?v=cydH_JAgSfg
Hylogen is an EDSL for livecoding shaders in Haskell [1]
My weapon of choice is Extempore by Andrew Sorensen [1]. Dual languages in one system available for blistering speed in livecoded graphics and sound/music.
Livecoded visuals/sounds is something i have been doing with a livecoding tool that i develop where a graphical canvas is actually both, the audio spectrum and the visuals, all driven by a single GLSL fragment shader : https://www.fsynth.com
https://news.ycombinator.com/newsguidelines.html
Woah.
Please articulate, Civilly, which guideline I Broke that other users don't break everyday and I will gladly change my ways, otherwise don't articulate my perceived shortcomings at all. If you embue me with the power to link to peoples comments I will find all shenanigans and post them for you to comment this message to.
https://vimeo.com/64895205
http://overtone.github.io/
If anyone is in London I will be giving a talk about using Tidal soon at this free event: http://www.musichackspace.org/events/music-hackspace-present...
I will hopefully find time to write something public at some point!
https://vimeo.com/36579366
Can you use text to move a slider?