Author Topic: Analysis of T5 voices  (Read 496 times)

0 Members and 1 Guest are viewing this topic.

Offline voodoo

Analysis of T5 voices
« on: August 11, 2017, 07:09:47 AM »
Hello,

thanks to PJ I found the wonderful UVF files of many T5 voices (over 1500). I started some grep commands (no XSL so far) and compiled the following Excel chart. For each given UFV I collected:

* MSB/LSB/PC
* Filename
* Category
* Voice Name
* Reverb send
* Chorus send
* DSP Name (if switched on)
* Number of voice elements (layers)
* Internal preset wave numbers (up to 8 per voice)
* Number of key ranges
* Average number of key ranges per layer

In the example in the picture we can confirm the usage of samples for the preset pianos.

Uli
Yamaha PSR-S970 (sold)
Yamaha Genos (ordered)
Nord Electro 5D
 
The following users thanked this post: John Plumridge, Mike W

Offline Cosmorot

Re: Analysis of T5 voices
« Reply #1 on: August 11, 2017, 07:43:00 AM »
What for? There is a ready device sheet. (There's a pair of mine in the casting voice.) Do not pay attention to this)
 

Offline pjd

Re: Analysis of T5 voices
« Reply #2 on: August 11, 2017, 01:42:53 PM »
Hi --

The purpose is to analyze the actual voice content -- not just a list of patch numbers. The UVF files describe how the voices are actually programmed.

For example, the "W" columns in the spreadsheet identify the waveform number used in the voice patch. With a little detective work, it's possible to figure out the layers used in the legacy voice.

The patch "Tutti," for example, is composed of five waveforms: 298, 299, 340, 1049, 1050. With a little backtracking, 298 and 299 are used in the AllegroString voice and 340 is a FrenchHorn.

I did some of this work manually when reconstructing voices that I like on the MOX6 synthesizer.

It's a case of wanting to know how voice patches are put together. I hope this gives context to Uli's work.

Nice work, Uli!

All the best -- pj
 
The following users thanked this post: voodoo