Forum index > Bug report

none 2/1 microtunings - scala file import problem

noyzelab
have just been loading in some none 2/1 based tunings and i get a problem of notes resetting to further down the keyboard. an example tuning enclosed from the scala library, reproduced below. as you can see it has 21 notes, with the repeating of the tuning appearing at 4/1. from what i can tell from a brief look, it seems to reset to 2/1 after each set of 21 notes.

! bohlen47.scl
!
Heinz Bohlen, mode of 4\47 (1998), www.huygens-fokker.org/bpsite/pythagorean.html
21
!
102.12766
255.31915
357.44681
459.57447
510.63830
612.76596
714.89362
817.02128
970.21277
1072.34043
1327.65957
1429.78723
1582.97872
1685.10638
1787.23404
1889.36170
1940.42553
2042.55319
2144.68085
2297.87234
4/1

------

there are a lot of tunings out there that dont use 2/1 as the repeat. see also some the tunings of wendy carlos alpha tuning excerpted from the scala scale library below, this also has the same problem.


! carlos_alpha.scl
!
Wendy Carlos' Alpha scale with perfect fifth divided in nine
18
!
78.00000
156.00000
234.00000
312.00000
390.00000
468.00000
546.00000
624.00000
702.00000
780.00000
858.00000
936.00000
1014.00000
1092.00000
1170.00000
1248.00000
1326.00000
1404.00000

-----

another useful tuning as a check case is the fixed 1/1 tuning. this also loads incorrectly, but in a different way to the others. if you go into advanced mode it is easy to see that instead of computing the same value for all the notes [which should be the master tuning value], it diverges from 0Hz to a frequency value of 42949672.42Hz. this is the scala file to check :

! fixed-1to1.scl
!
fixed note tuning of 1/1 across whole keyboard
1
!
1/1

----

there is possibly also going to be problems with tunings using repeats at values less than 2/1 if this problem is occurring with the above fixed 1/1 tuning.
DX100
This might explain the problems I'm finding with transposing layers.
KODAMO
Only 2/1 scales are supported as indicated here https://kodamo.org/tutorials.php?id=7
May be improved in the future although it is rarely used
DX100
Oh no!😢😱
Some of the most interesting tunings in the universe are not 2/1, I think this now explains the problems I was having with the patch layer transpose now.
Basically FULL microtuning is not implemented, damn!
Please Kodamo can you implement this otherwise its really half baked and the Essence will not be viewed as an amazing microtuning hardware unit which it is SO close to being.
KODAMO
We'll check how feasible this is.
That's only a limitation for Scala file import, you can still do full microtunings by sending a non-realtime MTS message (full dump - 128 notes) to the EssenceFM then saving the tuning.
noyzelab
ok great 👍 thanks for info, will give it a try. it would certainly be handy to have the .scl import work that way as well if possible. i did a batch convert of the scala library to all the output formats including MTS, available here if anyone is wanting to save time

https://gitlab.com/noyzelab/scala-scale-lib-bulk-conversion

if i'm creating new tunings algorithmically, eg using something like python or mathematica, the scala format is the easiest way to generate & export.
noyzelab
hi
is there any chance the Scala import can be fixed to be inline with full MTS please? i have a lot of files that are .scl only and it would be a massive hassle to convert them all. and as mentioned when generating new tunings [and also when sharing tunings], i suspect most people will make an .scl in the first instance anyway. having this sorted out would massively improve my microtuning workflow on this synth.
thanks
dave
flec8
+1 for this. Would love to have this flexibility. 🙂
Kopplog
+1 to the fullest
Kwerky
Big +1 for full .scl support
DX100
+1 for .scl import being fixed so it mirrors the non-realtime MTS message dump.

Reply