Jump to content


Photo

Behavior of Y00 in FX column and Yx in note columns


  • Please log in to reply
9 replies to this topic

#1 It-Alien

It-Alien

    Probably More God or Borg Than Human Member

  • Moderator
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 9593 posts
  • Gender:Male
  • Location:Avellino, Italy
  • Interests:trackers, basketball, chess, tabletennis, programming, writing, drumming, burping

Posted 27 December 2013 - 19:16

Attached File  Y00 ignores Yx.xrns   31.23KB   109 downloads

everytime a 0Y00 (mutually exclusive note triggering) is put on a command column, the chosen note is triggered regardless of its own probability.

in the attached example, I would expect the Y2 in the second note couple to affect the probability of the notes being triggered; the player should thikn this way:

- ok, there is a 0Y00, let me choose a note to play.
- here it is
- oh wait, it has a Y2; let me do a toss to see if I should really trigger it


instead, it just thinks this way:

- ok, there is a 0Y00, let me choose a note to play.
- here it is, let's play it!


so, in the end, the current version always plays a note of the second couple, while it should play one of the two notes, either only 1 time out of 8
  • dblue and radian like this
my specs:
cpu:Intel Core2 Quad Q9550 (4x2.83Ghz, FSB 1333Mhz)
os:Win7 Home Premium 64bit - OSX 10.9.1 64bit
audio: M-Audio Audiophile 24/96 PCI


Unschemed, hypertechnical tracked music. Sounding different since 1993.

my latest song: Lavender Town Syndrome (11'33", Psychedelic Beast)


where to find more music by me: soundcloud , jamendo , last.fm , napodano.com , facebook

RTFM.gif

("Lavender Town Syndrome, 11'33" Psychedelic Beast)

#2 f+d+k

f+d+k

    Big Super GrandMasta Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 716 posts
  • Gender:Not Telling
  • Location:On Land
  • Interests:고추장

Posted 27 December 2013 - 20:26

I think it does this:
+ Ok, there is a 0Y00, let me choose a note to play.
+ Add all the individual Yn's and normalize this sum to probability 1.
+ Do a toss to see which one to play with respect to their normalized probabilities.

E.g.
If there is only one note with Y2 (or one with Y2 and several others with Y0), it gets played 100%.
If there are two with say Y2 each, they both get played 50% each.
If there is one with YF and one with Y8, the first one gets played 2/3, the other one 1/3 of the times.

What I think is weird is if they all have Y0, nothing is played. I'd expect equal chances.

#3 Syflom

Syflom

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 1456 posts
  • Gender:Male
  • Location:Australia

Posted 27 December 2013 - 20:42

Confirmed with beta 2.
The Yx commands probability seems to not be respected if 0Y00 is used.

It acts as though YF (100% probability?) is used.

Edit: thanks for the description dblue.

Edited by Syflom, 27 December 2013 - 20:52.

In the beginning was the Word
 My XRNX tools: Loop Block Size, Write BPM, Sustain Instrument
Renoise my suggestions: sample pause command, continually paste within selection, advanced edit resize
OS: Windows and OS X


#4 dblue

dblue

    Dodgy Geezer

  • Admins
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 5586 posts
  • Gender:Male
  • Location:Berlin
  • Interests:Code. Music. Graphics.

Posted 27 December 2013 - 20:49

I think it does this: (...)

You're exactly right.

In "mutually exclusive" mode (when applying a 0Y00 on top) the behaviour changes a bit, so that one of the notes will always play.

The Yx command you assign to each note will then define its weight, ie. how likely it is to be played compared to the others.

For example:
C-4 Y2  D-4 Y1  0Y00

C-4 or D-4 will always play, but C-4 will play roughly twice as often as D-4.


The problem here is the dual behaviour of the Yx command in this particular scenario. A Yx command on a note cannot really do both desired things here. It cannot define both how likely the note is to be chosen by mututally exclusive, AND then also define how likely that note will actually play if it does get chosen. Or in other words, it could do this, but it's probably not going to behave in the way you expect, since the Yx commands are stacking and doing two different things at once.
  • Syflom and Daed like this

#5 It-Alien

It-Alien

    Probably More God or Borg Than Human Member

  • Moderator
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 9593 posts
  • Gender:Male
  • Location:Avellino, Italy
  • Interests:trackers, basketball, chess, tabletennis, programming, writing, drumming, burping

Posted 27 December 2013 - 21:40

Ok, I'm fine with this behaviour. As long as I know how to deal with it, it is ok. For me it is a nobug then
  • f+d+k likes this
my specs:
cpu:Intel Core2 Quad Q9550 (4x2.83Ghz, FSB 1333Mhz)
os:Win7 Home Premium 64bit - OSX 10.9.1 64bit
audio: M-Audio Audiophile 24/96 PCI


Unschemed, hypertechnical tracked music. Sounding different since 1993.

my latest song: Lavender Town Syndrome (11'33", Psychedelic Beast)


where to find more music by me: soundcloud , jamendo , last.fm , napodano.com , facebook

RTFM.gif

("Lavender Town Syndrome, 11'33" Psychedelic Beast)

#6 dblue

dblue

    Dodgy Geezer

  • Admins
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 5586 posts
  • Gender:Male
  • Location:Berlin
  • Interests:Code. Music. Graphics.

Posted 27 December 2013 - 21:52

Ok, I'm fine with this behaviour.

I would actually like the other behaviour to work somehow as well, because they're both useful. Just not sure how to combine everything without turning it into a complete mess. For example, putting Yx in volume column and Yx in panning column could *technically* work as to define both the mutually exclusive probability and the note's own trigger probability, but I think people would find this unnecessarily confusing (and I know taktik hates such fuzzy/hacky stuff).

#7 It-Alien

It-Alien

    Probably More God or Borg Than Human Member

  • Moderator
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 9593 posts
  • Gender:Male
  • Location:Avellino, Italy
  • Interests:trackers, basketball, chess, tabletennis, programming, writing, drumming, burping

Posted 27 December 2013 - 22:33

I would not like that as well: I usually put most of commands in the volume command, but if it is already filled, then I have to be able to put the very same command on the panning column; if the two version would work differently, I wouldn't be able to apply the desired behaviour.

Maybe YFF could be used for one and Y00 for the other?
my specs:
cpu:Intel Core2 Quad Q9550 (4x2.83Ghz, FSB 1333Mhz)
os:Win7 Home Premium 64bit - OSX 10.9.1 64bit
audio: M-Audio Audiophile 24/96 PCI


Unschemed, hypertechnical tracked music. Sounding different since 1993.

my latest song: Lavender Town Syndrome (11'33", Psychedelic Beast)


where to find more music by me: soundcloud , jamendo , last.fm , napodano.com , facebook

RTFM.gif

("Lavender Town Syndrome, 11'33" Psychedelic Beast)

#8 00.1

00.1

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 1347 posts
  • Gender:Male

Posted 28 December 2013 - 01:01

Maybe YFF could be used for one and Y00 for the other?

The whole time I thought that Y00 had some probability weight and not strictly a mode to coincide with vol/pan Yx.

I thought increasing it to say, Y70 or something, I'd get a higher probability that the notes in line would play together.

I can't think of anything right now, but both behaviors would be a nice conclusion for the final release.

Windows 10
Toshiba Satellite L875D
AMD A8-4500M APU with Radon™ HD Graphics 1.90 Ghz
6.00 GB RAM
64bit Operating System, x64-based processor


#9 spacecult

spacecult

    Super Advanced Member

  • Normal Members
  • PipPipPipPip
  • 152 posts

Posted 28 December 2013 - 01:44

I would not like that as well: I usually put most of commands in the volume command, but if it is already filled, then I have to be able to put the very same command on the panning column; if the two version would work differently, I wouldn't be able to apply the desired behaviour.

Maybe YFF could be used for one and Y00 for the other?


Could you just put 'dummy' notes (some zero volume instrument) in another column, giving the same effect as 'no play' when randomly selected by Y commands?

#10 00.1

00.1

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 1347 posts
  • Gender:Male

Posted 03 January 2014 - 13:37

Could you just put 'dummy' notes (some zero volume instrument) in another column, giving the same effect as 'no play' when randomly selected by Y commands?

Just wondering if this is a valid technique, the 'dummy' note, anyone ?

I tried the method once, and it seemed ok to go on with this way.

Windows 10
Toshiba Satellite L875D
AMD A8-4500M APU with Radon™ HD Graphics 1.90 Ghz
6.00 GB RAM
64bit Operating System, x64-based processor