I don't have a musical enough ear to have a clue whether this does entirely what it's supposed to. Any ideas? Has this been done before? To be proper I think it needs to work on the original integer ratio data so it computes at the accuracy of the machine running it.
Hallo, padawan12 hat gesagt: // padawan12 wrote:
I don't have a musical enough ear to have a clue whether this does entirely what it's supposed to. Any ideas? Has this been done before? To be proper I think it needs to work on the original integer ratio data so it computes at the accuracy of the machine running it.
I did something similar, but much simpler, with [tunetof], which is a replacement for [mtof] using different intonations. The intonations are specified as a simple table which maps "notes" to frequencies. These tables can be generated from tuning files in Scala's file format using a Python script supplied with [tunetof]. Both are available in CVS at /abstractions/footils/tunetof/
[1] http://www.xs4all.nl/~huygensf/scala/scl_format.html
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
I think any solution that computes it right there is superior to this huge preset table. Well actually I'm curious, because it's truncated to 8 places of decimal afair so I wonder whether anybody with very good musical ears can say how well that works, assuming I got the basic data dump right. What the table offers that a computed solution doesn't is the names, which are in the the CLM catalogue, so you have a handle on the scales instead of trying to remember a sequence of coefficients.
btw what are the proper replacements for unsymbol, list2symbol and tosymbol - some people aren't finding them, what should I use instead? cheers, a.
On Wed, 25 Oct 2006 10:04:45 +0200 Frank Barknecht fbar@footils.org wrote:
Hallo, padawan12 hat gesagt: // padawan12 wrote:
I don't have a musical enough ear to have a clue whether this does entirely what it's supposed to. Any ideas? Has this been done before? To be proper I think it needs to work on the original integer ratio data so it computes at the accuracy of the machine running it.
I did something similar, but much simpler, with [tunetof], which is a replacement for [mtof] using different intonations. The intonations are specified as a simple table which maps "notes" to frequencies. These tables can be generated from tuning files in Scala's file format using a Python script supplied with [tunetof]. Both are available in CVS at /abstractions/footils/tunetof/
[1] http://www.xs4all.nl/~huygensf/scala/scl_format.html
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hallo, padawan12 hat gesagt: // padawan12 wrote:
btw what are the proper replacements for unsymbol, list2symbol and tosymbol - some people aren't finding them, what should I use instead?
unsymbol: [list trim] (but in your patch you can just delete it)
list2symbol: list-l2s.pd from [list]-abs is an implementation using only core objects.
tosymbol: [symbol] or also use list-l2s.pd, but then again, I just removed it from your patch as well and it seemed to work.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
Cheers Frank. I removed the redundant parts and replaced the list2symbol with embeded core only stuff, the attached version should be more portable. Also I removed the loader because of the OSC errors it was throwing up for sime people.
Looks like 3 of those decimal places got lost in moving the data over through OSC as floats, or at least Pd is only showing 5 of them now :/
a.
On Wed, 25 Oct 2006 23:00:55 +0200 Frank Barknecht fbar@footils.org wrote:
Hallo, padawan12 hat gesagt: // padawan12 wrote:
btw what are the proper replacements for unsymbol, list2symbol and tosymbol - some people aren't finding them, what should I use instead?
unsymbol: [list trim] (but in your patch you can just delete it)
list2symbol: list-l2s.pd from [list]-abs is an implementation using only core objects.
tosymbol: [symbol] or also use list-l2s.pd, but then again, I just removed it from your patch as well and it seemed to work.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Here's the more portable version.
On Wed, 25 Oct 2006 23:00:55 +0200 Frank Barknecht fbar@footils.org wrote:
Hallo, padawan12 hat gesagt: // padawan12 wrote:
btw what are the proper replacements for unsymbol, list2symbol and tosymbol - some people aren't finding them, what should I use instead?
unsymbol: [list trim] (but in your patch you can just delete it)
list2symbol: list-l2s.pd from [list]-abs is an implementation using only core objects.
tosymbol: [symbol] or also use list-l2s.pd, but then again, I just removed it from your patch as well and it seemed to work.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
huge preset table. Well actually I'm curious, because it's truncated to 8 places of decimal afair so I wonder whether anybody with very good musical ears can say how well that works.
I'm no authority, but I work with tuning and scales, and 8 decimals is good enough for me any day :)
Everything is context-sensitive, of course; simpler timbres and sustained vertical sonorities need to be a lot more accurate than other things. For me, 8 places is "good enough."
Chris Bryan
Just checking out SCALA, it's the standard scales format I see. So, does [tunetof] embed thousands of scales in a table like my thing or do you load them on demand through the py script? Which is better in your opinion? I was happy to see that the 1000 or so from CLM fit happily into < 64k even when packed very clumbsily. The only rough edge so far is that scales are variable length, some are just 4 or 5 notes and others run up over many octaves.
On Wed, 25 Oct 2006 10:04:45 +0200 Frank Barknecht fbar@footils.org wrote:
Hallo, padawan12 hat gesagt: // padawan12 wrote:
I don't have a musical enough ear to have a clue whether this does entirely what it's supposed to. Any ideas? Has this been done before? To be proper I think it needs to work on the original integer ratio data so it computes at the accuracy of the machine running it.
I did something similar, but much simpler, with [tunetof], which is a replacement for [mtof] using different intonations. The intonations are specified as a simple table which maps "notes" to frequencies. These tables can be generated from tuning files in Scala's file format using a Python script supplied with [tunetof]. Both are available in CVS at /abstractions/footils/tunetof/
[1] http://www.xs4all.nl/~huygensf/scala/scl_format.html
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hallo, padawan12 hat gesagt: // padawan12 wrote:
Just checking out SCALA, it's the standard scales format I see. So, does [tunetof] embed thousands of scales in a table like my thing or do you load them on demand through the py script?
Neither: The Python script is used in advanced to convert Scala-files to [table] files, keeping their original name (with .tab appended).
Then you need to load the table-file manually through a message or a creation argument of tunetof.
Which is better in your opinion? I was happy to see that the 1000 or so from CLM fit happily into < 64k even when packed very clumbsily.
Yes, that's a very cool way to pack a lot of settings into a patch.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
-
c.m.bryan -
Frank Barknecht -
padawan12