This is based on the genetic data from spideron, the major protein of spider dragline silk. While I've been experimenting with genetic algorithmic music for some time, in the past I've used a bank of Emu Proteus-type samplers - this is my first attempt to use Scope exclusively as the sound source. This one uses only Scope for the sound and ArtWonk for the control, with Behringer MIDI controllers going through ArtWonk. All in real time.
The patch uses 16 Scope synths (Vectron, Poison, Lightwave, and assorted modulars) set up as 8 symmetrical voices, controlled by an ArtWonk algorithm that phases the two halves of each voice in a chase sequence - sort of a musical image of the DNA double helix.
http://algoart.com/download/music/Spideron2.mp3
This was for a live performance by Jamy Sheridan of visuals also produced by ArtWonk. We've been working for some years doing visuals+music where the software coordinates the two in various ways, in real time. Here is a link to Jamy's web where there are images from past pieces:
http://arthink.com
Here is the genetic data used for the piece:
http://whozoo.org/mac/Music/Spidroin.htm
It's from my wife's site. She is a biologist and does protein music also, but she is more interested in highlighting the genetic structures, whereas I am more interested in getting interesting sound. We wrote an article about it a few years back, her approach to the music as a scientist and mine as an artist. Here's the link:
http://www.leonardo.info/isast/articles/lifemusic.html
This was my first time using Scope as a massave multi synth, but it won't be the last. I've had the Scope setup for a while but it's been on a shelf for a couple years. I just put it back into the comuter about a month ago, and now I'm hooked. My next motherboard is gonna be a DFI with 3 PCI slots. Ether that or go all the way with an XITE-1. Anyway, I'm a believer now.
[Edited to update the .mp3 with the DC filter clean up discussed below]
A little genetic music
A little genetic music
Last edited by johndunn on Wed May 14, 2008 5:59 pm, edited 2 times in total.
-
paulrmartin
- Posts: 2445
- Joined: Sun May 20, 2001 4:00 pm
- Location: Montreal, Canada
Three suggestions:
1. Make Spidroin I much slower, I think it will be more effective that way.
2. Replace those awful GM sounds(probably from soundfonts?) with better samples.
3. Work on effects to make the ambiance come out.
This is some of the better results I have heard from genetic music. Spidroin II is quite nice with the orchestration changes.
You can always send me the MIDI files (which you can probably save from your generative program) and see what I can come up with
What gave you the idea of using a spider's genetic code?
1. Make Spidroin I much slower, I think it will be more effective that way.
2. Replace those awful GM sounds(probably from soundfonts?) with better samples.
3. Work on effects to make the ambiance come out.
This is some of the better results I have heard from genetic music. Spidroin II is quite nice with the orchestration changes.
You can always send me the MIDI files (which you can probably save from your generative program) and see what I can come up with
What gave you the idea of using a spider's genetic code?
Are we listening?..
Thanks for the feedback Paul.
1. make it slower...
It was done at a set tempo of 125 bpm because of the interaction with the Jamy's ArtWonk video performance. He had asked and I had agreed to keep the ArtWonk clock rate at a steady 20 milliseconds, which works out to 125 bpm, definitely Allegro.
It also meant no use of Accelerando or Ritardando throughout the piece, but I like to work with an unvarying tempo anyway because 1. it is representative of how DNA is translated in nature, and 2. my pieces are frankly computer music and one thing a computer can do that I like a lot (although I recognize most people don't) is to hold an absolutely steady robotic tempo, and do the rhythm changes in other ways, especially with pitch spikes. Also I plead guilty to a fondness of working in clusters of 16th notes. The trick is to get the transitions right, and I'm always struggling with that.
2. don't use GM sound fonts...
If you are referring to the pieces on my wife's web site, yes they are done with GM sound fonts. The Spideron I posted here was done with 16 Scope synths and modular synths, nary a sampler and certainly not GM. See the OP above for details on the setup.
3. work on effects...
Yes. I do struggle with effects. It's hard to know when enough is too much, especially when dealing with synth sounds, especially when many of them already use delay and chorus effects.
The performance was done in a fairly live hall, with good speakers. So it worked best to not add any more reverb. I did try a bit of PSY-Q but took it back out. The studio version used it on one of the voice-like tracks. I tried adding some EQ and reverb to some of the tracks, but took it out again. When I can't decide, I pull it. Comes from the art school background, less is more. I also used the Optimaster on the studio version, but I admit to being pretty lame with it.
Mostly, I probably should have used less delay in the synths. Individually it sounds great, but when you start stacking them up, it starts getting muddy; I think that is probably what you object to, and I agree it could use improvement.
Regarding a MIDI file, ArtWonk pumps MIDI out in real time, so I usually don't bother to record the MIDI file and didn't for this. But since you asked for it, I went ahead and recorded one today, and uploaded it to my site:
http://algoart.com/download/music/Spideron.mid
...so go for it. It will sound like crap initially because all the program changes are for the 16 Scope synth setup, and none of the presets are the original ones. I pretty much tweak everything and build my own preset lists for each synth. So you are going to have to set each preset to the instrument of your choice.
There are 16 MIDI channels, two identical synths per channel except they are panned at 32/95 to make each synth pair symmetrical left-right. And each synth gets 2 or 3 or more, up to 8, preset changes throughout the piece. Also you need to be sure the volume change messages aren't ignored by whatever sequencer/editor you use. Some of them use Velocity for note volume - I use the velocity for dynamics and Volume for, um, volume. And since the voices are fading in and out all the time, there are a *lot* of volume changes to contend with.
But given that, go ahead and do what you can with the MIDI file. I'd be most interested in your take on it.
The idea of using spideron came from my wife Mary Anne who used it first. It is a very regular, very rhythmic protein, much more so than most. Jamy sent me some of the ArtWonk animations he was working on for the piece, and they seemed sort of darkish and rhythmic so I suggested we base it on Spideron and he agreed it would work for both of us. Thus both the visuals and the sound are based on the same protein data. Probably from the same spider.
1. make it slower...
It was done at a set tempo of 125 bpm because of the interaction with the Jamy's ArtWonk video performance. He had asked and I had agreed to keep the ArtWonk clock rate at a steady 20 milliseconds, which works out to 125 bpm, definitely Allegro.
It also meant no use of Accelerando or Ritardando throughout the piece, but I like to work with an unvarying tempo anyway because 1. it is representative of how DNA is translated in nature, and 2. my pieces are frankly computer music and one thing a computer can do that I like a lot (although I recognize most people don't) is to hold an absolutely steady robotic tempo, and do the rhythm changes in other ways, especially with pitch spikes. Also I plead guilty to a fondness of working in clusters of 16th notes. The trick is to get the transitions right, and I'm always struggling with that.
2. don't use GM sound fonts...
If you are referring to the pieces on my wife's web site, yes they are done with GM sound fonts. The Spideron I posted here was done with 16 Scope synths and modular synths, nary a sampler and certainly not GM. See the OP above for details on the setup.
3. work on effects...
Yes. I do struggle with effects. It's hard to know when enough is too much, especially when dealing with synth sounds, especially when many of them already use delay and chorus effects.
The performance was done in a fairly live hall, with good speakers. So it worked best to not add any more reverb. I did try a bit of PSY-Q but took it back out. The studio version used it on one of the voice-like tracks. I tried adding some EQ and reverb to some of the tracks, but took it out again. When I can't decide, I pull it. Comes from the art school background, less is more. I also used the Optimaster on the studio version, but I admit to being pretty lame with it.
Mostly, I probably should have used less delay in the synths. Individually it sounds great, but when you start stacking them up, it starts getting muddy; I think that is probably what you object to, and I agree it could use improvement.
Regarding a MIDI file, ArtWonk pumps MIDI out in real time, so I usually don't bother to record the MIDI file and didn't for this. But since you asked for it, I went ahead and recorded one today, and uploaded it to my site:
http://algoart.com/download/music/Spideron.mid
...so go for it. It will sound like crap initially because all the program changes are for the 16 Scope synth setup, and none of the presets are the original ones. I pretty much tweak everything and build my own preset lists for each synth. So you are going to have to set each preset to the instrument of your choice.
There are 16 MIDI channels, two identical synths per channel except they are panned at 32/95 to make each synth pair symmetrical left-right. And each synth gets 2 or 3 or more, up to 8, preset changes throughout the piece. Also you need to be sure the volume change messages aren't ignored by whatever sequencer/editor you use. Some of them use Velocity for note volume - I use the velocity for dynamics and Volume for, um, volume. And since the voices are fading in and out all the time, there are a *lot* of volume changes to contend with.
But given that, go ahead and do what you can with the MIDI file. I'd be most interested in your take on it.
The idea of using spideron came from my wife Mary Anne who used it first. It is a very regular, very rhythmic protein, much more so than most. Jamy sent me some of the ArtWonk animations he was working on for the piece, and they seemed sort of darkish and rhythmic so I suggested we base it on Spideron and he agreed it would work for both of us. Thus both the visuals and the sound are based on the same protein data. Probably from the same spider.
"remove the poison"...
At first I thought it was a dx 11 or 100 (it is typical FM sounds you have there). But posion creates offsets in sounds (check the recorded wave samples).... There are better FM generators around for scope, some free, some not free, but what is important is that offset thing (a good synth should present quasi simetrical peaks up and down the zero offset of the wave file representation)....
or use a dc offset as an insert effect in the tracks in the mix.
At first I thought it was a dx 11 or 100 (it is typical FM sounds you have there). But posion creates offsets in sounds (check the recorded wave samples).... There are better FM generators around for scope, some free, some not free, but what is important is that offset thing (a good synth should present quasi simetrical peaks up and down the zero offset of the wave file representation)....
or use a dc offset as an insert effect in the tracks in the mix.
Thanks, Mehdi, for the tip on DC offset on Poison. I didn't know that, obviously.
I put a little Poison in the mix because I'm sort of partial to that old DX7 FM sound. I was a summer student at the Stanford AI lab when John Chowning was trying to get people interested in FM, using the Stanford Sampson box. Pity he had to go to Japan to get it produced. It was the beginning of the end for US made synths.
Which didn't stop me from buying a DX7 when it came out, and then a pair of TX816s. For a total overdose of FM, here is a piece I did using the 2 TX816s (basically 8 DX7's each, sans keyboards), driven by my DOS based MusicBox, a precursor to ArtWonk/MusicWonk. It's called Theta Music, and its based on the EEG theta wave cycle:
http://algoart.com/download/music/Algor ... /THETA.MP3
Here's an old 1989 Keyboard review of MusicBox, with pictures of the insanely cryptic DOS character graphics interface:
http://web.inter.nl.net/hcc/davies/algmbox.html
...so what are the better choices for Scope FM synths?
I put a little Poison in the mix because I'm sort of partial to that old DX7 FM sound. I was a summer student at the Stanford AI lab when John Chowning was trying to get people interested in FM, using the Stanford Sampson box. Pity he had to go to Japan to get it produced. It was the beginning of the end for US made synths.
Which didn't stop me from buying a DX7 when it came out, and then a pair of TX816s. For a total overdose of FM, here is a piece I did using the 2 TX816s (basically 8 DX7's each, sans keyboards), driven by my DOS based MusicBox, a precursor to ArtWonk/MusicWonk. It's called Theta Music, and its based on the EEG theta wave cycle:
http://algoart.com/download/music/Algor ... /THETA.MP3
Here's an old 1989 Keyboard review of MusicBox, with pictures of the insanely cryptic DOS character graphics interface:
http://web.inter.nl.net/hcc/davies/algmbox.html
...so what are the better choices for Scope FM synths?
Hi John,
Well myself I am partial as I use mainly my own FM oscillators in my own devices.
However, The Modular offers ways to make various FMs. Also, there are some freeware that had good reviews in the past (FeeMee, a very old device) but may be there are more recent ones that are as good or better (in the "device" section, or if a generous soul points you to stuff i do not know).
Funny you talk about the TX816.... There are at least 3 persons now who know that synth (me, Xite 4 Live and you). It gave me the idea to make a simple multi-FM synth, but in fact having FM only could be too limitative so it might be a usual spacef synth (I use inserts to allow to change oscillators, filters and sometimes modulators).
Poison is probably the most directly DX-like FM type of synth on Scope, but you must absolutely use a DC offset.
All the best
Mehdi
Well myself I am partial as I use mainly my own FM oscillators in my own devices.
However, The Modular offers ways to make various FMs. Also, there are some freeware that had good reviews in the past (FeeMee, a very old device) but may be there are more recent ones that are as good or better (in the "device" section, or if a generous soul points you to stuff i do not know).
Funny you talk about the TX816.... There are at least 3 persons now who know that synth (me, Xite 4 Live and you). It gave me the idea to make a simple multi-FM synth, but in fact having FM only could be too limitative so it might be a usual spacef synth (I use inserts to allow to change oscillators, filters and sometimes modulators).
Poison is probably the most directly DX-like FM type of synth on Scope, but you must absolutely use a DC offset.
All the best
Mehdi
been there, in summer 1993... there were like 4 or 5 dx and the class (or more, it's blurred memory).... I was amazed to see that there were classroom where the desks were actually synths ! dxII if i remember (not really sure, i remember they were black so...) On the dashboard, there were some diagrams (classic DX ops) and it was scary for me in that time (eventhough i owned a dx 100 since 1984 and ended programming dx stuff really easily, never really understood why people think it was hard to program) .... I was attracted though. I didn't know that 5 years laters I would dive, for the long term, into even more complex construction that probably would have make me panic a lot in 94johndunn wrote: I put a little Poison in the mix because I'm sort of partial to that old DX7 FM sound. I was a summer student at the Stanford AI lab when John Chowning was trying to get people interested in FM, using the Stanford Sampson box. Pity he had to go to Japan to get it produced. It was the beginning of the end for US made synths.
I probably would have taken that class, if I knew , but the money i had left was used in renting cars to travel, SF Area, napa valley and LA with a bunch of friends, and it was a lot of fun too
Mehdi
EDIT: theta mp3: sounds really nice and full. whereas the first track reminded me of 4ops FMs, in that one i can hear the big difference of more complex FM.
This was very helpful information. It turns out that the Modular synths were the biggest contributor to the DC, due no doubt to my sloppy patching. But once I put a filter on all synths that were spilling DC, the whole mix cleaned up and I was back to the sound I had for the performance, that had eluded me afterwards when I wanted to clean it up for the .mp3.spacef wrote:Hi John,
Well myself I am partial as I use mainly my own FM oscillators in my own devices.
However, The Modular offers ways to make various FMs. Also, there are some freeware that had good reviews in the past (FeeMee, a very old device) but may be there are more recent ones that are as good or better (in the "device" section, or if a generous soul points you to stuff i do not know).
Funny you talk about the TX816.... There are at least 3 persons now who know that synth (me, Xite 4 Live and you). It gave me the idea to make a simple multi-FM synth, but in fact having FM only could be too limitative so it might be a usual spacef synth (I use inserts to allow to change oscillators, filters and sometimes modulators).
Poison is probably the most directly DX-like FM type of synth on Scope, but you must absolutely use a DC offset.
All the best
Mehdi
Funny thing, the problem with the performance was that I had to run at the 32KHz sample rate because Scope wouldn't load at 44.1, and there was a SR compatibility issue, so I went through the analog output, which of course unknowingly filtered out the DC.
So now I'm a happy camper, and will be uploading the cleaned up single version later today. A tip of the hat to you!
count me in... the only reason I don't have it is it's price tagspacef wrote:...Funny you talk about the TX816.... There are at least 3 persons now who know that synth (me, Xite 4 Live and you). ...
I have a TX802, TX7 and PLG150DX hosted in a Yamaha groovebox and I love FM
to be honest I was totally blown away recently when I dumped a few (of the better) DX7 presets into the Yamaha thingy to tweak them with realtime dials
excuse the hijack, but it had to be mentioned...
cheers, Tom
