Round robin question

Discussion on the SFZ implementation and opcodes available in the ARIA engine.

Moderators: eric_telemaque, davidv

davidv
Site Admin
Posts: 1567
Joined: Tue Mar 02, 2004 7:23 pm
Contact:

Re: Round robin question

Postby davidv » Thu Oct 27, 2011 9:15 am

Assume:

<group>
lorand=0 hirand=0.5 seq_length=2

<region> sample=RED seq_position=1
<region> sample=BLACK seq_position=2

<group>
lorand=0.5 hirand=1.0 seq_length=4

<region> sample=BLUE seq_position=1
<region> sample=YELLOW seq_position=2
<region> sample=ORANGE seq_position=3
<region> sample=PURPLE seq_position=4

1)hit any key .... say random gets evaluated to 0.25
->RED plays

2)hit any key .... say random gets evaluated to 0.75
->BLUE plays

3)hit any key .... say random gets evaluated to 1.0
->YELLOW plays

4)hit any key .... say random gets evaluated to 0.6
->ORANGE plays

5)hit any key .... say random gets evaluated to 0.0
BLACK plays.


Two sequences, one for each random group. but in principle each region could have a different sequence length

I dont know how i could be clearer.
David Viens,
Plogue Art et Technologie Inc. Montreal.
http://www.plogue.com

rockdude
Posts: 21
Joined: Thu Apr 21, 2011 9:12 am

Re: Round robin question

Postby rockdude » Thu Oct 27, 2011 1:09 pm

davidv wrote:I dont know how i could be clearer.

I don't know how you could be clearer either :D ....but I appreciate this explanation very much! When written like this it can't be misinterpreted.

Btw, did you get xfin_xxcc and xfout_xxcc implemented in a good way?

davidv
Site Admin
Posts: 1567
Joined: Tue Mar 02, 2004 7:23 pm
Contact:

Re: Round robin question

Postby davidv » Thu Oct 27, 2011 1:43 pm

rockdude wrote:Btw, did you get xfin_xxcc and xfout_xxcc implemented in a good way?


Yes it should be as per the SFZ 2.0 spec now. Just need to ship the update. Still adding features ....
David Viens,

Plogue Art et Technologie Inc. Montreal.

http://www.plogue.com

rockdude
Posts: 21
Joined: Thu Apr 21, 2011 9:12 am

Re: Round robin question

Postby rockdude » Thu Oct 27, 2011 2:25 pm

davidv wrote:
rockdude wrote:Btw, did you get xfin_xxcc and xfout_xxcc implemented in a good way?


Yes it should be as per the SFZ 2.0 spec now. Just need to ship the update. Still adding features ....

Great, and more features....nice! I would love to see a way to add sfz knob controls in Arias effect section that you can control custom midi cc with so that it's possible to control whatever you want in the sfz code that has an opcode with a midi cc number attached to it.

davidv
Site Admin
Posts: 1567
Joined: Tue Mar 02, 2004 7:23 pm
Contact:

Re: Round robin question

Postby davidv » Thu Oct 27, 2011 2:38 pm

rockdude wrote:Great, and more features....nice! I would love to see a way to add sfz knob controls in Arias effect section that you can control custom midi cc with so that it's possible to control whatever you want in the sfz code that has an opcode with a midi cc number attached to it.


This is in fact a new feature that I've implemented :)
David Viens,

Plogue Art et Technologie Inc. Montreal.

http://www.plogue.com

rockdude
Posts: 21
Joined: Thu Apr 21, 2011 9:12 am

Re: Round robin question

Postby rockdude » Thu Oct 27, 2011 2:43 pm

davidv wrote:
rockdude wrote:Great, and more features....nice! I would love to see a way to add sfz knob controls in Arias effect section that you can control custom midi cc with so that it's possible to control whatever you want in the sfz code that has an opcode with a midi cc number attached to it.


This is in fact a new feature that I've implemented :)

YEEEEEEES!!!!!!!!! 8)

niki
Posts: 15
Joined: Tue Feb 23, 2016 8:12 pm

Re: Round robin question

Postby niki » Thu Feb 25, 2016 3:43 pm

Sorry for bumping old thread, but it may be interesting.. :roll:

I would like to notice that this way of making random round robin is not exactly random with equal chances.
Chances of what is next sample depend on current sample and are not equal.

Code: Select all

<group>
lorand=0 hirand=0.5 seq_length=2

<region> sample=RED seq_position=1
<region> sample=BLACK seq_position=2

<group>
lorand=0.5 hirand=1.0 seq_length=2

<region> sample=BLUE seq_position=1
<region> sample=YELLOW seq_position=2


Let's assume that RED sample was last played sample.
With equal chances random we expect that chances of playing BLACK, BLUE and YELLOW are equal (1/3).
But what we really get here is chance of BLACK is 1/2 and chances of BLUE and YELLOW are 1/4 each.
Chance of playing next sample that is within the same group is always 2 times more than chance of any other particular sample. (chances of groups are equal of course, but chances of samples are not) 8)

pljones
Posts: 100
Joined: Wed Oct 15, 2014 2:05 pm

Re: Round robin question

Postby pljones » Fri Feb 26, 2016 2:19 am

That looks correct.

There's a 50/50 chance of selecting each group.

If the first group (with RED/BLACK) is selected, only one sample can be chosen - Black - as it's the next sequential sample.


Return to “SFZ Programming”

Who is online

Users browsing this forum: No registered users and 2 guests