Need help creating definition.txt for Drumkit

G
gmeader posted Oct 6 '17, 07:31:

I'm trying to create a definition.txt file for this drumkit:
https://musical-artifacts.com/artifacts/4

I see that there is a way to define choke groups (%mutegroup), but there's no example provided in the documentation and I can't figure out how to make it work. Could someone provide an example, perhaps using the filenames in the drumkit (above).

Also I don't know what to do with the multiple numbered files for each percussion instrument. Are they velocities or are they to be used with %seq randomization.

I'm trying to use this documentation to figure it out. http://samplerbox.readthedocs.io/en/latest/Sample_Sets/

G
Glenn posted Oct 18 '17, 20:38:

crickets...

H
HansEhv posted Oct 18 '17, 22:56:

Hi,
The question is somewhat out of scope: sfz is only partly supported and that's only in Alex's build. So make sure to use that one. Also I suppose using both the defintion.txt and the <name>.sfz will be confusing - if possible at all.
I'd say, take a look at http://www.sfzformat.com/legacy/ so you can understand what you've downloaded, ie what the multiple numbered files are.
Next you can decide on what to use for the samplerbox sampleset - and how.
Questions arising then will be more likely in scope of the samplerbox.
Hope this helps.

G
Glenn posted Oct 20 '17, 01:41:

Thanks Hans,
You've given me good clues to figure it out.
The drumkit is downloaded in a 387MB archive file: salamanderDrumkit.tar.bz2
That file extracts to a directory named salamanderDrumkit that contains several .sfz files and a directory named OH that contains 537 .wav files. I am not trying to utilize the .sfz files at all. I just copied the .wav files into my /media/10 drumkit folder and now am attempting to create a definition.txt file that will allow SamplerBox to use these wav files.

Looking in the ALL.sfz file does provide some clues. For example, below is what's there for the crash1 instrument:

There are 3 sets (groups) of files.

  1. 6 files with crash1_OHFF in the name are for velocity 60 thru 127 (midinote 55)
  2. 5 files with crash1_OHP in the name are for velocity 1 thru 59 (midinote 55)
  3. a file crash1Choke_OH_F_1.wav for choke (midinote 54) tagged as group 2
    It seems like the sets of 6 and 5 files are meant to be chosen randomly (according to some probability from 0 to 1?)
    The off_by=2 opcode means that if midinote 54 (tagged as group 2) is played, then any midinote 55 sounds are choked off.
    There is a different volume opcode for each group

So I am wondering what is the proper way to create SamplerBox definition.txt entries for this example.
(I realize additional definition.txt entries need to be created for all the instruments)

<group> volume=-12 amp_veltrack=100 ampeg_release=0.6 key=55 loop_mode=one_shot lovel=60 hivel=127 off_mode=normal off_by=2 // crash1 ////6 Samples Random!
<region> sample=OH\crash1_OH_FF_1.wav lorand=0 hirand=0.18
<region> sample=OH\crash1_OH_FF_2.wav lorand=0.18 hirand=0.26
<region> sample=OH\crash1_OH_FF_3.wav lorand=0.26 hirand=0.45
<region> sample=OH\crash1_OH_FF_4.wav lorand=0.45 hirand=0.68
<region> sample=OH\crash1_OH_FF_5.wav lorand=0.68 hirand=0.85
<region> sample=OH\crash1_OH_FF_6.wav lorand=0.85 hirand=1

<group> volume=-29 amp_veltrack=100 loop_mode=one_shot key=54 group=2 // crash1Choke /////
<region> sample=OH\crash1Choke_OH_F_1.wav

<group> volume=-19 amp_veltrack=95 ampeg_release=0.6 key=55 loop_mode=one_shot lovel=1 hivel=59 off_mode=normal off_by=2 // crash1 ////5 Samples Random!
<region> sample=OH\crash1_OH_P_1.wav lorand=0 hirand=0.2
<region> sample=OH\crash1_OH_P_2.wav lorand=0.2 hirand=0.4
<region> sample=OH\crash1_OH_P_3.wav lorand=0.4 hirand=0.6
<region> sample=OH\crash1_OH_P_4.wav lorand=0.6 hirand=0.8
<region> sample=OH\crash1_OH_P_5.wav lorand=0.8 hirand=1

...

  (not published)
  I want to post as guest
 

Post