Autor Tema: Some Pt3 info  (Leído 10031 veces)

0 Usuarios y 1 Visitante están viendo este tema.

Dioniso

  • Visitante
Some Pt3 info
« en: 04 de Septiembre de 2006, 12:12:55 pm »
Hello.

Long ago I tried to write a kind of doc about PT3 modules. I stopped it but due to a certain person who asked for it ... (this goes especially for Nerlaska) I write here what I have. I can't remember exactly what I wrote on some papers so when you see something like (?), it means I can't remember what I wrote  :o :o :o

It would be very nice if someone could complete the info here given. It's very easy; just write some basic on Vortex Tracker II and start watching with an hex-viewer the changes ... at least, that's what I did  ;D

Have fun.

===========================================================================

PT3 format:

*Info inside the module*

$00: "ProTracker 3."
$0D: Version (always "5" for Vortex Tracker II).
$0E: " compilation of "
$1E: Name of the module.
$3E: " by "
$42: Author of the module.
$63: Frequency table (from 1 to 4).
$64: Speed ----------------------------------------------------------- Delay:    1 byte
$65: Number of patterns.
$66: Pattern loop --------------------------------------------------- LPosPtr:  1 byte
$67: Position of patterns inside the module ---------------- PatsPtrs: 2 bytes
$69: Position of samples inside the module ----------------- SamPtrs:  2 bytes
     (2 bytes per address * 32 samples = 64 bytes)
$A9: Position of ornaments inside the module ------------- OrnPtrs:  2 bytes
     (2 bytes per address * 16 ornaments = 32 bytes)
$C9: It points to where the patterns order ist ------------- CrPsPtr:  2 bytes
     If you have a look to where this address points you can see, for instance,
     $00,$03,$06,$03,$FF, which means:

pattern 0 ($00), pattern 1 ($03), pattern 2 ($06), pattern 1 ($03), END ($FF).
As you can see, you have to multiply for 3 every pattern number.

------------------------------------------------------------------------------

Now, in group of 6 bytes, come the address of every channel of every pattern.
So the first 6 bytes will point to the address of channel A, B and C of
pattern 0 (3 channels * 2 bytes = 6 bytes). The next 6 bytes to the 3 channels
of pattern 1, etc ...

------------------------------------------------------------------------------

In $69 (SamPtrs) we have the address where the definition of the samples
start. So $69 and $70 will contain the address of the definition of the first
sample in the module. $71 and $72 of the second, and so on.

The definition of the samples are like this:

byte 1: Loop
byte 2: Length of the sample
byte 3: 0 0 0 0 0 0 0 0
           ¦_¦ ¦______¦ ¦___ 0: Envelope   1: No envelope
            ¦         ¦
            ¦         ¦________ Noise: 0 - 31
            ¦
            ¦______________ 10: Volume down   11: Volume up

byte 4: 0 0 0 0 0 0 0 0
           ¦_¦ ¦_¦  ¦____¦
            ¦    ¦       ¦
            ¦    ¦       ¦______ Volume
            ¦    ¦
            ¦    ¦___________ Noise slide (?)
            ¦
            ¦______________ Frequency slide (?)

byte 5: Low byte.
byte 6: High byte.
These two bytes (5 and 6) contain the frequency slide. In case of negative
slide (higher pitch) the frequency you want to rest is rested to $FFFF and
stored in these 2 bytes.

Bytes 3,4,5 and 6 form a line of the sample (raster). So a sample will have
so many groups of these bytes as lines has the sample.

------------------------------------------------------------------------------

Definition of ornaments are much easier:

Byte 1: Loop
Byte 2: Length
Byte 3: Note (0-95)

The same as for samples. Byte 3 contains the definition of one line so there
will be as much bytes containing the new note as lines of definition has the
ornament.

------------------------------------------------------------------------------

Now it becomes more complex :)

The address of the channels of every pattern come in group of 6, as we said.
If we go to the address pointed for channel A (the first 2 bytes of those 6)
we'll have definition of the channel A in the tracker. The bytes are coded so
we have to read the higher 4 bites:

$4x: Ornament ($40-$4F)
$50-$AF: Note to play. There are 96 (12 notes * 8 octaves). The byte before
         this one tells the length in lines inside the tracker. For instance:
         $05,$50 will play C1 (Do in octave 1) during 5 lines of tracker. If
         the Speed (Delay) is 6, the player will play this note during 30
         (6 rasters * 5 lines in tracker) rasters.
$Bx: Envelope. We have to add 1 to know which value should we give to register
     13. If there's envelope (a value different to $B1) the next 2 bytes are
     register 12 and register 11 of the PSG. An example: $B9, $00, $50 means
     R13=$08, R12=$00, R11=$50.
$Cx: Volume - $C0-$CF
$Dx: Sample - $D0-$EF
$0X: Effects/Commands ... The values they handle are set after note value
     ($50-$AF).

*Effects*

$01: Sound frequency decreasing in that channel.
     First byte indicates the delay used to add the new frequency.
     Next 2 bytes will indicate the frequency to add. Example:
     $02,$23,$00 will add $23 to the final frequency in that raster
     and another $23 every 2 rasters.

$01: Sound frequency increasing in that channel.
     It's the same as above but the value is rested to $FFFF. Example:
     $01,$DD,$FF will rest $23 to the final frequency in every raster.

$02: Tone portamento.
     First byte is delay again. (?)

$03: Starts playing sample from a particular position/line definition.
     Byte indicates from which position.

$04: Starts playing ornament from a particular position/line definition.
     Byte indicates from which position.

$06: Periodical sound off/on in that channel.
     Two bytes are used here. The first one tells how many rasters will be
     played and the second how many will be not. Example. $03,$04 means
     that that channel will be played 3 rasters, 4 not played, 3 played, ...

$08: Envelope frequency increasing.
     Two bytes will indicate the frequency to add to registers R12 and R11.
     Example: $34,$00 will add $34 to R11.

$09: Set playing speed (new Delay).
     The byte after the note will tell the new delay.

$0A: Envelope frequency decreasing.
     Same as increasing but the value is rested to $FFFF.
     Example: $CC,$FF will rest $34 to R11.

------------------------------------------------------------------------------

*Example 1*

So if we have this in the module: $D2,$CF,$B1,$05,$50 we will have this in the
tracker:

C-1   2..F   ....
---     ....    ....
---     ....    ....
---     ....    ....
---     ....    ....

$D2: Sample 2    ($Dx samples)
$CF: Volume 15   ($Cx volume)
$B1: No envelope ($Bx envelope)
$05,$50: Play note $50 (C1) during five tracker lines.

*Example 2*

If we have this in the module: $D1,$CE,$09,$B1,$08,$50,$07

$D1: Sample 1.
$CE: Volume 14.
$09: New delay will be set. See after note is set.
$B1: No envelope.
$08,$50: Play note $50 (C1) during eight tracker lines.
$07: New Delay.

===========================================================================

pitpan

  • Karoshi Forum's Guru
  • *******
  • Mensajes: 1812
Re: Some Pt3 info
« Respuesta #1 en: 05 de Septiembre de 2006, 08:58:22 pm »
That's useful information. Thanks for sharing it with us, Dioniso.  :)

Huey

  • Karoshi Newbie
  • *
  • Mensajes: 10
Re: Some Pt3 info
« Respuesta #2 en: 20 de Agosto de 2010, 06:36:34 pm »
Sorry for bumping this old thread. But I just had to thank you Doniso for your explanation.

It is just what I needed. B-)

Dioniso

  • Visitante
Re: Some Pt3 info
« Respuesta #3 en: 20 de Agosto de 2010, 09:37:29 pm »
I just had to thank you Doniso for your explanation.

It is just what I needed. B-)

You are very welcome  :)

Huey

  • Karoshi Newbie
  • *
  • Mensajes: 10
Re: Some Pt3 info
« Respuesta #4 en: 07 de Marzo de 2011, 01:14:07 pm »
Is it true that the global noise is at (0x20 - 0x3f) of channel A data?

And how is the global envelope stored? 0x00 - 0x1f?
« Última modificación: 07 de Marzo de 2011, 03:22:38 pm por Huey »

Huey

  • Karoshi Newbie
  • *
  • Mensajes: 10
Re: Some Pt3 info
« Respuesta #5 en: 08 de Marzo de 2011, 12:53:14 am »
Duh. :-[

I see now. Sorry. Forget my questions. ::qbiqsray::