Hi all,

I'd like to share my new composition called O Ritual Da Sombra for solo tenor recorder and live electronics. Unfortunately I am myself a very bad recorder player, so I am still looking for someone to perform it live and record it. Meanwhile, what I have is a demo of the piece using a sampled recorder.

Cheers,
Gilberto

Ok now it works! Maybe the pc wasn't in a good mood yesterday hehe The piece is really great, I like how it starts a single melody to slowly turn more polyphonic. Lets hope you find some flute player to be able to record it with the real thing!

Thanks for your kind words, Daniel. I also hope to find a player very soon, and then I'd repost the link here.

I reuploaded the compostion as I made a small change on the algorithm that generates its score.

Actually it sounds better to me. Thanks for sharing.

@Daniel-Daowz Thanks! The thing is almost all my instrumental pieces are generative, and so how they sound depends on which "version" of a piece you are listening to (in theory there are infinite versions of every single piece, each score is absolutely unique).

Oh ok, I get it. So even with the new algorithm you created for this you every interpretation will be different right? In my opinion that is one of the strongest points of composing with a computer.

Yes, that's correct! The only change from this new algorithm compared to the old one is that I forbid consecutive notes that are too difficult/impossible to play on a recorder (the lowest C, C#, D and D# are quite temperamental). Everything else (structurally speaking) is the same as before.

In my opinion that is one of the strongest points of composing with a computer.

I agree, I am really into generative music.

@gsagostinho I'm curious as to why you are concerned with the playability of an algorithmically computer-generated part. In a performance with a real flutist, would she have a fixed part to play? Would you randomly generate a part for her based on these algorithims? Until you get a real flutist though, it really shouldn't matter, unless you want to deceive listeners into thinking this is a real flutist - listeners who are also expert enough to know that the recorder has trouble in that range.

@s.elliot.perez Hi,

I'm curious as to why you are concerned with the playability of an algorithmically computer-generated part.

Well, I want my music to be playable. After all, my main occupation in life is to produce music and so I consider myself as a composer

In a performance with a real flutist, would she have a fixed part to play? Would you randomly generate a part for her based on these algorithims?

Yes, an interpreter would have a fixed part, but this part would be generated by the computer program. Instead of sending him/her a score, I can as easily send the computer program and allow the recorder player to generate a version for him/herself. Therefore it must produce playable scores.

Until you get a real flutist though, it really shouldn't matter, unless you want to deceive listeners into thinking this is a real flutist - listeners who are also expert enough to know that the recorder has trouble in that range.

That is not correct. You are mistaking a demo of a work with a work generated by computers. What I posted is simply a demo, something very common among composers: you are simply showing how the work will sound, not trying to deceive anyone. Regardless of this, it's obviously very important that my computer program takes into consideration the limitations of the instrument since these scores it generates are meant to be performed - regardless if I already have a performer in mind or not.

Cheers,
GIlberto

@gsagostinho That's fair enough, but have you any interest in works that don't rely on a performer? I've gotten tired of trying to have some of my own scores performed, so I've been collecting the needed samples in order to self-produce them. But whether there will be a live version in the future or not, I'm not opposed to having something in the computer version that can't be replicated in the live performance. It would mean that there's something special about both versions instead of the one being seen as a mere demo.

Well, I do also create purely electronic music, if that's what you asking -- although I see myself much more as an instrumental composer (the absolute majority of my pieces don't use any sort of electronics). Dealing with performers can be quite exhausting, I know that quite well (particularly since a lot of them seem to be quite uneducated on contemporary music, at least around my region here) but when things go well the result is generally very satisfying to me. Also, there is always an element of surprise when you first hear someone playing your music in a rehearsal, and sometimes they bring some very interesting elements you hadn't thought about it before.

I personally don't have much interest in creating a work that sounds as if it could be played, but which is not destined to be played. I think that if I go the route of composing something to be played automatically by the computer, then I am interested in exploring the almost limitless sounds it can produce and not try to imitate a violin or a piano. But that's just my own opinion.

But this subject is quite interesting. For instance, do you know the work of Conlon Nancarrow? He has several works for piano player which can't be played by any human being, and the results are utterly fascinating:

Cheers!
Gilberto

@gsagostinho Yes, that's what I was getting at: that if one realizes something on the computer, one might as well take advantages of the possibilities offered by the computer. On the other hand, your demos are certainly useful for competitions or for giving the interpreter a better idea of what he has to do.

I also agree that humans contribute something special to piece's performance. I'm just kind of tired of waiting/hunting for players at the moment. At least PD allows us to somewhat recapture the spontaneous element through its generative abilities.

And yes! I have heard some of these player-piano pieces from Nancarrow. I must confess, watching the video of the piano makes it a bit more hilarious. It makes Alkan's pieces look "easy" (i.e. playable).

For me, demos are a way of sharing what I am doing right now without having to wait for a performance that may well never come. They are also very useful for myself so I can check how the work is sounding, and this is really a standard right now: I don't know of a single composer that does not produce demos beforehand.

For me, the problem with creating music with a computer but using regular sounds (violin, piano, etc.) is an aesthetic one: the computer is simply trying to imitate something, and without a very strong concept I think these compositions almost always turn to be very poor or not interesting: they are not as good as real life performance, nor they are as interesting as an electronic work that really embraces the electroacoustic world.

And you find Nancarrow hilarious? I don' think that is actually his intention, I am pretty sure he meant those compositions in the most serious way. They have some amazingly precise accelerandos e ritenutos, as well as other aspects of music, and so it's not only the sheer amount of notes that makes it impossible to be played by a human being.

Still on this subject, these days I bought a CD of Xenakis keyboard works performed by a MIDI piano (a real piano which can play itself according to a MIDI track, so a modern version of Nancarrow's instrument). The programmer (or should I say pianist) made the rhythms as precise as possible, something that no human is capable of doing either. The result is really nice and gives a very nice light to these works that I adore: https://neos-music.com/?language=english&page=output.php%3Ftemplate%3Denglish-album-details.php%26content%3DAlben/10707.php

Cheers,
Gilberto

PS: regardless of anything, Alkan is still unplayable

@gsagostinho It seems Nancarrow transcribed a piece by Cecil Taylor? I could be wrong.... (Tee Hee).
David.

@whale-av I wouldn't know about that (actually, I didn't even know Cecil Taylor until you mentioned him). While I do listen to jazz once in a while, this is absolutely not my area at all

@gsagostinho Try this!
I was lucky enough to mix his Trio once.... a long time ago. It's as close as you are likely to get to "impossible:

It's as close as you are likely to get to "impossible"

Well, it depends on what you mean by that. When one is improvising, there is a lot of liberty concerning what is right and wrong, particularly when comes to rhythm. So yeah, someone can improve a super ultra complex rhythm which the same person would never be able to read from a score and reproduce perfectly (think of Xenakis, Ferneyhough, etc.). So there are degrees of impossibility, as well as types of it (e.g. a minimal work lasting for 10h might be literally impossible to play for most pianists).

Cheers!

@gsagostinho said:

(e.g. a minimal work lasting for 10h might be literally impossible to play for most pianists).

... well there's Terry Riley also, who's having his 80th birthday soon

@seb-harmonik.ar Oh, Riley's music is really cool. I programmed his In C for my thesis about algorithmic instrumental music. Pity that in that time I was working with Max

I know several who don't produce demos, myself included. Often, the techniques involved are hard to get samples for, the notation is handwritten and contains concepts which are difficult to convert to a computer language.

I would argue that Nancarrow's piece is the same as making a piano piece on the computer. It's the human element that makes live performances exciting, so a MIDI-piano performance should be just as poor or uninteresting as a computer-made composition that incorporates instrumental sounds.

The hilarious thing about Nancarrow's player piano is the technical premise, and watching those keys move up and down so rapidly on their own is very amusing to me. That doesn't mean the composition itself is or is not serious. I rather like the ending of that piano study.

Mr. Taylor is quite impressive. And Alkan is playable! People play his music... sometimes.