Please create an account to participate in the Slashdot moderation system

 


Forgot your password?
Close
typodupeerror
×
Image

Chess Games Translated To Music 87

Posted by samzenpus from the sound-of-strategy dept.
An anonymous reader writes "Blogger Jonathan W. Stokes used algebra to map famous chess games onto a piano, and then outputted the results as MP3s. The tunes created are surprisingly listenable."
This discussion has been archived. No new comments can be posted.

Chess Games Translated To Music More

Chess Games Translated To Music

Comments Filter:

  • YMMV (Score:1)

    by Zephyn ( 415698 )

    I suspect most of my games would wind up sounding like the piano part to "Louie Louie".

    • I was thinking the next step would be a music2chess. The awkward situation of Louie Louie beating a chess grand master.

      Or even worse, Viswanathan Anand being checkmated by a drummer.

  • Why is slashdot linked to twitter? Oh wait.....
  • I like the idea and everything, but it seems disingenuous to "set blues chords in the left hand to justify the constant tonal shifts from B to b flat in this chess game. The chords modulate from C Major to F Major and finally end in B Flat Major." Not really impressive IMO if the algorithmic compositions can't stand on their own.

  • Very cool! But what about the opposite? (Score:1)

    by Anonymous Coward

    I wonder if there is a song out there that could be mapped to a chess game. This would be much more difficult (if not impossible) to find one AND have the moves be legal / in the correct order.

    It'd make for a heck of a flash movie though.

  • And now, for your listening pleasure... (Score:5, Funny)

    by Zephyn ( 415698 ) on Wednesday February 23, 2011 @04:17PM (#35293784)

    Victor Borge and Bobby Fischer will perform "In the Hall of the Mountain King's Gambit Accepted."

    • Followed in order by "A Variation on an Opening by Lasker in G file".

    • Re: (Score:3)

      by cruff ( 171569 )

      Let us speculate:
      Would Victor use the pieces he captured to modify the sound of his piano by letting them bounce around on the strings?
      Or would Victor just provide musical accompaniment and commentary while Bobby played another person, perhaps an opera singer?

  • I've thought of this basic idea years ago, but never got anything working. I think a better algorithm can be made by using the "tension" of a game defined by whether something has an "only move" sacrifice or whether it is a simple positional game. Other factors include how threatening or placid a move is.

  • make it [wikimedia.org] stop!!!

  • I've heard lots of jazz that was far, far worse.

  • Amiga (Score:2)

    by Chysn ( 898420 )
    I did a chess game-to-music algorithm for a university class back in 1993. The algorithm itself was written into an AREXX script for Amiga. AREXX was a neat scripting language for the Amiga; many software packages included an AREXX port, which exposed an API to the language, so that AREXX programs could control the software. So I had the script read chess notation from my word processor, parse it, calculate the music, and write the notation to my notation software, where I could print it, play it, change

  • Listenable compared to what? (Score:5, Insightful)

    by jfengel ( 409917 ) on Wednesday February 23, 2011 @04:35PM (#35293954) Homepage Journal

    Seems to me the first thing you'd want is a control game. Convert a game played by two amateurs to music. Or two computers making random legal moves. Do they sound any less listenable?

    Or have you just rediscovered basic music theory that random notes in the same key end up sounding like music?

    It's pretty clear from listening that it's not following any time pattern; it's got no beat.

  • These, particularly the second sample (The Opera Game) sound like something Richard D James [wikipedia.org] would do [sun.com]. A certain random, almost generative quality.

  • Is it the chess game that sounds good? (Score:3)

    by Normal Dan ( 1053064 ) on Wednesday February 23, 2011 @04:37PM (#35293974)
    Or is it the program that interpreted the chess game into music? That is, would a bunch of random numbers being fed into the program produce just as good of a sound? Sometimes I wonder about these things.

    • Re: (Score:2)

      by jovius ( 974690 )
      You can't really miss if you choose a harmonic framework and make timing interesting enough.

      What's interesting here is the method to generate the notes and the relative pitch variance. The former is the choice of the composer (or producer in this case) and the latter comes from the game, from the players.
  • This exercise is a waste. We could do a similar exercise with sectors of the lawn your dog chose to piss in and the results would be just as random and useless.

    Sometimes I wonder how this stuff makes it onto the main page.

    • Re: (Score:3)

      by obarel ( 670863 )

      I've actually found a strong correlation between the sectors of the lawn my dog chooses to piss in and the stock exchange. And I don't even have a dog.

    • Yes, he possibly could have used Go; but the algorithm would be much too complex. Every stone is a stone somewhere, and placing one stone to one point to key mapping is not useful. Its threats and influence are rather difficult to calculate, difficult enough for humans to ascertain even. Indeed, the translation of Go to music seems frivolous and illogical; the translation of chess to music seems more likely, but only with a strict structured algorithm intent on the purpose of getting good results in the

  • What's being done is starting with a chess game, throwing out most of the information (the positions of non-moving pieces, which piece is moving, and one of the two dimensions of movement), converting (deterministically) what little is left into a sequence of notes, deciding (creatively/non-deterministically) what rhythm to put them in, and deciding (creatively/non-deterministically) how to harmonize them. It's only a mapping between maybe 10% of the game and maybe 20% of the music.

    It's a mildly interestin

    • Re: (Score:2)

      by Anonymous Coward

      What's being done is starting with a chess game, throwing out most of the information (the positions of non-moving pieces, which piece is moving, and one of the two dimensions of movement), converting (deterministically) what little is left into a sequence of notes, deciding (creatively/non-deterministically) what rhythm to put them in, and deciding (creatively/non-deterministically) how to harmonize them. It's only a mapping between maybe 10% of the game and maybe 20% of the music.

      It's a mildly interesting way to "seed" the creative process, but it's neither an impressive intellectual accomplishment (from a musical or mathematical perspective) nor a testament to hidden order in the universe. Most people seem to be misinterpreting it as one (or paradoxically both) of those.

      This is what I hate about fellow nerds. As soon as someone in the community accomplishes something, they have to shit on whatever they accomplish and bitch about everything they do wrong and explain how they could do it much better without actually producing actual results. All this guy wanted to do was create something fun out of something most people thing is boring and I think he succeeded beautifully. If this leads to increased appreciation for what we nerds do then we all win. Put up or shut up. Y

    • RTFA again. The pieces moving aren't being thrown out, they're being used to determine the note value, and thus the rhythm the notes are being put in is being determined by the chess game. The rest of your points are valid.
      • Ah, yes, I missed that. The point remains that this is not (and is not even intended to be) algorithmic composition by any stretch of the imagination.

    • The only good point you've made is the elimination of one of the dimensions of movement. However, the combination of piece value and target file is much more significant than a 10% compression of the game. For example, at any given point in the game, there's very few different moves possible from those conditions. The non-moving pieces are non-moving; even in chess notation nobody cares about those.

      Additionally, the goal of this exercise is to deterministically convert chess to music. So I think you're miss

      • It sounds like you know more about chess than I do, so I'll take your word for it that more than 10% of the game is represented. But speaking as both a mathematician and a musician, I was being generous in describing melody alone as 20% of musical content. Listen to a piano arrangement of Pachelbel's canon, then a song called Heeding The Call by power-metal band Hammerfall and tell me I'm wrong. The non-deterministic additions Stokes is making to the deterministic chess output match, or arguably slightly

        • Re: (Score:2)

          by tibit ( 1762298 )

          OK, call me very, very thick today, but I don't understand what you really wanted to say. Do you mean that Pachelbel's Canon [in D?] and Heeding The Call share a melody? When you are comparing Stokes's non-deterministic additions to the difference between PC and HtC, do you mean they are both big?

          • Yes, Heeding The Call takes its entire melody, and even most of its chord progression, directly from Canon in D--and yet they sound so different that a non-musician wouldn't notice that they had anything in common at all. That's how small a part melody plays in music. And that tiny fragment of common ground--melody--is the ONLY part of the chess music that is generated algorithmically from the game descriptions. The rest--the vast majority--comes from Stokes' imagination.

            There's certainly nothing wrong w

            • Re: (Score:2)

              by tibit ( 1762298 )

              Heeding The Call takes its entire melody, and even most of its chord progression, directly from Canon in D--and yet they sound so different that a non-musician wouldn't notice that they had anything in common at all.

              Very interesting. Obviously I'm not a musician ;)

            • Re: (Score:2)

              by tibit ( 1762298 )

              I find it interesting that when one googles for pachelbel canon and heeding the call together, only your post comes up, or else I gave up browsing too early. I know that it may be that no one else had explicitly mentioned it, but I still find it a bit strange that it's not a widely known piece of trivia given popularity of both the canon and Hammerfall. I've tried to find the similarities, and lo and behold, one can start from the now obvious [youtube.com] and go from there.

            • Re: (Score:2)

              by tibit ( 1762298 )

              This rant [youtube.com] gives even more examples, even if it misses Heeding The Call. And thanks to you I've spent 3 hours on youtube instead of working, great ;)

              • Wow, that was hilarious, thanks for posting it. I feel for the guy--orchestral trombone parts (I'm a trombonist) are all like that.

  • "This content has been blocked in accordance with (my company) Webfilter Policies.
    URL: jonathanwstokes.com/2011/02/14/chess-music/
    Category: Pornography"

    Hmmm....

  • The important question is how does the sofa get round that bend in the stairs?

    • Re: (Score:2)

      by Macgrrl ( 762836 )

      Richard MacDuff called and wants you the hell out of his apartment. And take that damned monk with you when you go. You can leave the horse.

  • The idea of using fairly random or non-musical elements in composing music is definitely not a new idea: John Cage [wikipedia.org] famously created a piece using the I Ching as the source of randomness. The thing is, how it sounds depends largely on how you set the parameters you randomize. For instance, if you allow pitch to change because somebody played Qd2, but have all the notes at the same volume, the most noticeable effect will be the relatively constant volume.

    And yes, I am a music geek who's even composed a few th

  • You could probably easily adapt this script that plays random-ish music to get the input from this type of mapping to chess moves, as well. :) (plays directly to /dev/dsp)
    http://kmkeen.com/awk-music/ [kmkeen.com]
  • "Outputted" - seriously? More coffee for the poster methinks. Or a spell check enabled browser.
  • What they need to do next is to map a companies financial reports to music. That might make some money if the government doesn't buy it out.
  • Next thing you know some lawyer will be running around getting people to reveal what they aren't telling him by playing an elaborate game of mental chess.

  • Another cool chess/music project (Score:5, Interesting)

    by jbum ( 121617 ) on Wednesday February 23, 2011 @09:23PM (#35296250)

    Back in 2009, I did something very similar with one of the 1997 Kasparov vs. Deep Blue games.

    One difference is that I used a chess engine, and made the search tree audible, so you can hear the chess computer "thinking". Here's my original blog article: http://www.krazydad.com/blog/2009/05/musical-chess/ [krazydad.com] and here's video from the concert: http://www.youtube.com/watch?v=42G6P0b72Gk [youtube.com]

  • i was already working on 'audiobrain' for pChess - mapping move scores to pitch, but jon stokes system for mapping square values to note and octave makes too much sense - using Bflat for B, and Bnatural for the H column is just genius, and provides the missing link - i sense a new feature coming to pChess.. :-}

    pChess (open source chess application for OS X) [earthlink.net]

  • I've been listening to these so long... (Score:3)

    by srobert ( 4099 ) on Thursday February 24, 2011 @10:54AM (#35300596)

    I've been listening to these so long, I don't even hear the music anymore. All I see is pawn, knight, redhead.

Slashdot Top Deals

Technology is dominated by those who manage what they do not understand.

Close