Algorithmic Future


    Sep 04 2014 | 10:41 am
    Hi,
    I would like you to join a discussion on a topic of algorithms in electronic music production. We know that beat is essential and how do we get this beat is also essential. I made a research work, which I would like to share with you about how algorithmically produce a rhythm and sound synthesis. This will help from one side, novice users to get quickly appropriate results and from another side, experts could gain more knowledge on the rhythm as a whole. There are only two deterministic algorithms as I encounter in the world for rhythm generation (without any randomness or chains I mean), which are Euclidean algorithm for rhythm generation only and Compositor Software algorithm, which is not only generates rhythm, but also synthesizes a sound. This both algorithms have advantages and disadvantages. From my point of view, Euclidean algorithm gives results that are more versatile, but without an implied sound synthesis and Compositor is a unification of both rhythm and synthesis.
    Euclidean nature is additive, Compositor is bottom-up physical modeling instrument.
    What to choose is up to you. Unfortunately, most of the Euclidean algorithm Max For Live devices disappeared from the market by the will of time and you can do your own search for those, which still there. Compositor Max For Live from another side can be painlessly updated, because it is written in its entirety on Gen~.

    • Sep 04 2014 | 7:44 pm
      It is a very interesting discussion Euclid 300 BC was cleverly plagiarized by software companies. A brilliant mind, which illuminated the humanity for centuries was unable to get an email or a patent.
    • Sep 04 2014 | 8:25 pm
      I got curious and checked the Compositor demo videos and soundcloud samples on your site. To me they all sound largely the same. Not the best way to promote an algorithmic generator I guess. And this is a matter of taste but both sounds and rhythms are entirely unappealing to me.
    • Sep 05 2014 | 1:32 am
      I believe that you have to have some sort of moment to moment personal interface in whatever music that you create for it to be relate-able. Now i DO realise there are some incredibly expressive musicians such as Gyorgy Legeti and Iannis Xenakis that compose mainy with algorithms. However, they still seem to prefer to use orchestra musicians over computerized performers to balance it out. Its kind of humorous that the phrase occurring to me is, "where's the humanity?"
    • Sep 05 2014 | 8:16 am
      Compositor is about the challenge to make an algorithm, which not only make rhythm but also generate the sound. Physical modelling used is subtle but can be controlled in Pro version to create Doppler shifts. In Lite version it is controlled automatically to make frequency shifts very subtle too. So the whole instrument is not about radical changes in sound but rather subtle ones. I don't remeber the name but here is the quote for you: "Techno is all about subtle changes". So in Compositor you will not be fooled by any control it really DO a subtle changes if it's only a 50Hz or so in sound - you can hear it.
      As an artist under "Boosty" alias with appearance on Magda's Fabric 49 compilation I produced a techno based on a conception of Live performance of instrumental sections applied to techno music production. With Compositor it is controversary, because I only need to mention that Frequency Modulation, which algorithm is based on gives a fundamental material for subtractive selection of harmonics. I only need to follow them with my mind to hear the sequence it's that easy, but later they processed by a much larger chain of post-processing effects.
      So the whole thread is to compare Euclid varieties with Compositor strict application. Yet, I stress out if someone could find an algorithm to make Euclid rhythms not only to be a beat generator, but a sound synthesis it will be a challenge. But now we are on the step to continue a discussion and your opinions are welcome.
    • Sep 05 2014 | 10:42 am
      "Pretentiou$ and Paran®id"
    • Sep 05 2014 | 12:27 pm
      There are only two deterministic algorithms as I encounter in the world for rhythm generation
      there are thousands.
    • Sep 05 2014 | 12:41 pm
      "Pretentious and Paranoid."
      "That's just like, your opinion man." - Jeff Lebowski
    • Sep 05 2014 | 12:48 pm
      such as Gyorgy Legeti and Iannis Xenakis that compose mainy with algorithms
      not to forget mozart, beethoven, schubert, bach, diabelli, reich, the bee gees, jeff mills, just to name a few where it is maybe a bit more obvious then for the other 5 million we could mention here ... a musical composition always contains mathematical algorithms, and using them conciously is a widespread method of composing musical events.
    • Sep 05 2014 | 12:50 pm
      It's difficult to have a conversation about something you're 'sharing' when I first have to part with $40 for the software or $30 for your .pdf explaining it.
      I can't see that anyone can really contribute anything useful here going on what info you have made available for free.
    • Sep 05 2014 | 12:53 pm
      Roman Thilenius
      He's taking about Physical Modeling. Mozart and the Bee Gees never had to worry about that. In fact not many musician fiddle with it at all. However, it is growing in popularity. Also I thought it was safe to assume that a discussion of this topic would pertains to more relevant artists. I chose these musicians because it is just what I associated with this topic. They are the artists I was listening to while I was becoming aware of the use of Algorithms used in music or even what an algorithm is in the first place. Give me a break.
    • Sep 05 2014 | 1:20 pm
      it does not make much sense to limit a discussion about algorithms with what they are calling "euclidian rythm", because "euclidian rythm" is sooo basic and sooo natural, that its use is more found in conjunction with improvisions or copied music than with actual composition.
      to be concrete, that guy euclid has not invented that; it is one of the most used basic structures in west african music, and those who were using it (probably before euclid) were not dealing with math at all.
      and the arrangement of some of the polyrythms in bee gees songs or the cadences of beethoven have far more in common with the computerbased "algorithmic composition" you are talking of than with the euclid bullshit from 2300 years ago.
    • Sep 05 2014 | 1:31 pm
      And I'm sure we all know this(meaning the objective parts of what you just said). And I wasn't limiting. I was showing appreciation toward two specific and very talented artists. So do you ever apologize, or do you just poor cold piss all over everything?
    • Sep 05 2014 | 1:59 pm
      Get used to the max forums. People are seriously interested in the topics here. Nobody here has a lot of time, and you can feel lucky about getting cold piss in this discussion. I mean, nobody knows why, but you didn't get asswiped! I'm always thankful for a community that would tell me that I'm an idiot if I'd come and say something meaningless with such a pretentious attitude.
    • Sep 05 2014 | 2:04 pm
      Pretentious? Why should I be called pretentious for trying to write as clearly as possible. Not to mention, I think that if you look back at my original post you'll see that its an argument for both sides. What could possibly be pretentious about that? The only time I talk that way is when I look up to the person/people I am talking to. I was surprised when you all took it the wrong way
      Roman - "bullshit"
      Abowman1293 - "cold piss"
      WOYTEG - "asswiped"
      And as long as we're all nit picking, Roman - "Musical composition always contains mathematical algorithms" Roman - "those who were using it (probably before euclid) were not dealing with math at all."
      So far none of us look great
      Now if anyone has anything else they need to say to me about this you should email me privately.
    • Sep 05 2014 | 4:25 pm
      I think what we talking about is much different from my point of view. I mentioned two algorithms for sequential generation of events, without an implied randomness or stochastic chains, which one can consider also as musical algorithm. I wouldn't also bother to name a composition based music as it composed purely by nature of will of a composer, not by a math formula. I'm also aware that Euclid algorithm has nothing to do with rhythm in a first place, but it is a research of Godfried Toussaint, which leds to this implementation. Said that I will enter the discussion not only polite but in chronological order.
    • Sep 05 2014 | 4:35 pm
      I am curious to know to what extent these algorithms are to be used in composition or whether or not they are mainly used for the purposes of synthesizing sound rather than composition.
    • Sep 05 2014 | 4:51 pm
      One Flew Over the Cuckoo's Nest ?
    • Sep 05 2014 | 5:03 pm
      Euclid algorithm can generate one bar patterns suitable for sequential music like electronic dance music, where people dance to a repeating pattern. Euclid algorithm implies that you will trigger some midi instruments with it's output. Compositor from the other hand uses FM to determine sequential pattern and FM side bands to produce the sound. Mainly Compositor's output is a one bar cycle of FM output, rhythmically is more like ratamaque rudiment. Composition wise this could be used with variations, such as rhythmic permutations, which makes rhythm more versatile. My idea is not so far from polyrithmic structures. But personally if I do the Euclid synth one day I will make it not as sequecer, but as it should be a ratio of two numbers and setting those there will be a rhythmic output. The main reason I stick with Compositor is that both rhythm and sound synthesis defined by one function, which is not additive, but subtracitve for an order of things.
    • Sep 05 2014 | 5:04 pm
      Yeah I suppose so. All I want is a discussion. You guys were the ones who attacked me.
    • Sep 05 2014 | 5:18 pm
      I do not know if you've thought something like a few days ago when you suspected all Cycling74 wanted to steal your code, just because slow to answer a query. Now you pretend that someone discuss your claims based on your complete ignorance. Suddenly you say. there are only two usable algorithms for a particular purpose, and one of them is yours. If you have nothing to do, and want to participate must pay .. I think now you thought you "crafty, and want to put a kiosk of lemonade in a forum.
      Pretentious, paranoid and greedy algorithm running in your brain there.
    • Sep 05 2014 | 5:20 pm
      Who me?
    • Sep 05 2014 | 5:24 pm
      I really don't think this guy comes across so bad. I mean it sounds like he is just trying to create something and is asking for some input.
    • Sep 05 2014 | 5:29 pm
      Ruslan, okay I think I am starting to get the picture now. Using euclidean algorithmic relationships in audio synthesis is certainly an interesting idea.
      "Compositor from the other hand uses FM to determine sequential pattern and FM side bands to produce the sound. Mainly Compositor’s output is a one bar cycle of FM output, rhythmically is more like ratamaque rudiment."
      So are you saying you are using the euclidean algorithm to determine the frequency relationship of the FM synthesis (carrier and modulator freq)?
    • Sep 05 2014 | 5:30 pm
      Many times a great idea is just a tiny detail away
    • Sep 05 2014 | 5:36 pm
      WOYTEG – "asswiped"
      :) Apologies if I sounded rude. Asswiped is of course refering to Raja and my surprise that he hesitates to contribute one of his rant/troll/asswipe/serious opinion mixtures that tend to be misunderstood as offending. Whatever, sorry.
    • Sep 05 2014 | 5:40 pm
      No big deal. Just my god! One post and people start calling you names and insulting your intelligence.
    • Sep 05 2014 | 5:48 pm
      @ ANTHONY PALOMBA - Yes, you are right. I tried to control the multiplier, which is Fc to Fm ratio with the use of Euclid least common divisor algorithm. I ran into some problems, because in my case I use not integer Fc to Fm ratios, such as 1.54 for example and here Euclid algorithm is not applicable. I can use Euclid algorithm to control ratios such as 1:2 or 1:3 for example. And I also tried to find maximum integer relationships for float values such as 1.54 for example. There should be a field for research at least I ran into big numbers this way and they are not what I'm looked for. The idea of Compositor is fully automatic performance and if I would estimate the Fc to Fm ratio with Euclid algorithm I can find a proper automation algorithm. As of now I control multiplier with stochastic algorithm using Beta distribution and exponential one.
    • Sep 06 2014 | 1:15 am
      :P excactly what I wanted to say
    • Sep 06 2014 | 8:09 am
      @RAJA The topic originally is not to discuss techno, trance or any genre of EDM, with the use of passion and spirit as you say one can use Euclid algorithm or Compositor in any circumstances. I made techno demos as I set up Compositor mainly with 4/4 straight beats, which came as artistic nature of how I got them in a first place. There is no sequencer of any sort it's beatings of frequencies, which defines such a rhythm. I don't see any criticism in this thread apart of ignorance to the whole idea of the topic. Maybe we got it in wrong direction. I would like to discuss also a problem why so many Euclid sequencers use polyphonic idea instead of just spreading a one line of rhythm on different lines of a staff. It is originally discussed by David Garibaldi in his linear drumming conception applied for funk music. So I personally don't see any use in any of Euclid generators out there because of this limitation.
    • Sep 06 2014 | 10:33 am
      This thread seems to be a symptom for what happens in communication when a small group becomes bigger (due to M4L?) and the confirmed & agreed common ideology is challenged. Interesting to see how former respectful friendly behavior goes savage ... Once there was a healthy tendency to let meaningless dissent just hang out to dry itself until silenced; now everybody, even the big ones, beats the drum / the piggy. The need for sacrifice is increasing ... Is this because of a higher frustration level?
      And related to the question of the thread: I find the real question when thinking algorithmic music is humanity. Assuming that every human invention is aimed to have a functional purpose, I like to think that one big job for music is to reflect back our unconscious and reassure us about ourselves as emotional beings. All the things 'parking' in music are based on feelings; this spreads from the most basic 'tension of tritone being released in the octave' to joy/sadness of a song. It even goes for 'celestial' experiences (yes, I do believe that 'god' and alike are a projection, rooted in a human necessity to believe in something other then themselves because they cannot bear being alone in some situations; in this case [playing/listening] music helps, as much as praying).
      It used to be a composers job to do assure that line. Producing something, a tension/release product that he felt in his body, based on his own mind/soul condition, getting it out of his body through the, and sharing it with the listeners or dancers.
      Then machines were made to emulate humans, mainly for physical work but soon we had mechanical ducks, faked chess players, clumsy robots, but also planes and cars. And then it went electrical; first the coils, then tubes, transistors, chips and now we do not know exactly what they are at ... It has to do with big data and ai for sure. Of course you know all that. But what I really find interesting is the shift of 'the human position' in all that. What started as an emulation of humanity has reversed. The machines started to determine our body (see Chaplins 'Modern Times' for that) and since shortly they determine our way of thinking and feeling.
      I see a shift in the composers attitude; the process of creation is shifting. The posture of 'I have an urge of expression' and look out for means to get that into the world is drifting to 'Oh look, there is a new app / vst / patch, lets see what IT can do'.
      In other words: people are submissive to possibilities that seem to be given to them by the machines. They are enjoying 'pressing buttons' to see what comes out. The freedom is cropped down to the experience of being a 'user'.
      There is a harsh parallel in the behavior towards the economic markets.
      More and more it seems that we all are becoming servants to a virtuality we once created. Even music follows that path into the attraction economy.
      I always thought Max/MSP to be a sideway of that movement (and funny to know that one of Miller Puckettes' needs was to get out of the box with a spreadsheet he was working on).
      So in other words: The question of where Algorithmic Music is going should be: 'Where do we want it to go' and push it there. Gheee. The creators are here, not? And, instead of butchering each other, should be working on beautiful stuff; and create platforms for present them. And help out each other while doing so.
      sorry; I just had one of my rare religious moments ... now back to programming.
    • Sep 06 2014 | 10:57 am
      @CTRLZJONES Personally I agree with this conception. And tritone is a nice primer. But hey diminished harmonies is a style of many minimal techno tracks and it's easy to emulate this. So here is an answer. We do move in algorithmic future, but someone still want to apply their knowledge in pressing buttons. Yet, they must realise if the product is so minimalist, then it could be comprehended and emulated. And I think it is not a thankfull deed to struggle for techno or other electronic music styles to be human created. Instead, I can speak for myself only, after creating my program I lost motivation to produce techno music in conventional way. Thats why algorithm even if it sound weird or sometimes strict in application could be thought of a first steps in motivic music creation. My way of estimation the style is that I prefer to automate regular tasks. We can use scripts, we can use algorithms. But the unification of this approach seem to have one standalone application, which aims to fulfill the niche.
    • Sep 06 2014 | 4:44 pm
      i have not read that pdf for three times now, but it still doesnt make sense to me.
    • Sep 06 2014 | 4:57 pm
      raja wrote:
      it’s goa trance i hate
      i always knew there is something really wrong with you, but i am shocked that it is so serious.
      loool..
    • Sep 07 2014 | 12:08 am
      about 10 minutes into this talk i thought was/still is an interesting take on the algorithmic future: http://vimeo.com/30955464
      i've always found using other people's explicit algorithms doesn't really work for me and makes things harder in a bad way. feels like someone saying "because i said so" when you ask why. a lot of why i got into doing electronic stuff was so i could figure things out before deciding if i wanted to use it or not.
      the euclidian stuff i've heard seems kinda inoffensive, which is worse than bad to me. haha. i guess i like to attempt to steal from what i already find good or interesting as a whole because failing or getting sidetracked is good too and i'm really skilled in those areas! play to your strengths and all that.
      about "fundamental" beats, i read that birds and people are the only animals that can intuitively sense a rhythm. dunno if that's true. i mean, i'm pretty sure it's true for people, haha. i'd never really thought that other animals couldn't, i assumed more they just didn't like it! i'd guess why we can has something to do with time perception? maybe the way we remember things? the circadian rhythm is an interesting oscillation/beat feedback network though. it's kinda like how the acoustics of a space will influence the sounds you use there and the sounds you want to use will influence the acoustic spaces created to use them in. or beats are kinda like sampling rate, you do them fast enough it's seems like a continuous oscillation or you grab enough discrete points and it becomes a beat. or like particle/wave duality. lol
    • Sep 07 2014 | 7:53 am
      I think grasshoppers is a nice primer of rhythmically reproducing the sound. When I tried to approach grasshopers sound I was thinking of it as AM, but later I decided that it could have a wider spectrum. I don't think the rhythm grasshoppers do is connected to any mechanical reproduction, but it may be a continuous chain of events or just a random probability. First of all to be considered as discrete "machine" the process should be sequential. That what I'm talking in the thread. With Compositor I looked only in one direction of one drum rudiment named earlier. But there are more rudiments here and if we will think of each rudiment as a finished machine we will get a complete future fulfilled with thinking machines. Those rudiments I adwise to use as an additive formulas of beats or better to say a Fourier theorem can shed a light on additive nature of a rhythm and sound. However additive is not a law, such as Frequency Modulation for example and if we will take all the technics I better stick with non-linear transformations to approach this rudiments and the way we got them.
    • Sep 09 2014 | 11:20 pm
      Due to the high amount of requests I decided to upload a new demo album of Compositor Pro v2 software replacing the older one. It contains different tracks in downtempo and techno style. You can listen to it here.