Appendix 3

●About binary file format

Here we describe the format of the sending file. Basically, we will include a complete version of the format submitted by YAMAHA with any missing, missing, or incorrect parts corrected. The rough structure of the SATURN SCSP format is as follows.

Item	size
Mixer start offset from the beginning	2 bytes
Velocity start offset from start	2 bytes
PEG start offset from start	2 bytes
PLFO start offset from start	2 bytes
Offset of voice data number 0 from the beginning	2*a byte
:
Offset of voice data number a from the beginning
Mixer Data 0	18*b bytes
:
Mixer Data b
VelocityData 0	10*c bytes
:
VelocityData c
PEGData 0	10*d bytes
:
PEGData d
PLFOData 0	4*e bytes
:
PLFOData e
VoiceData 0	a{F(x) is 4*a+Σ(F(i)*32) for each voice Returns the number of included layers i=0}
:
VoiceData a
WaveData 0	f ΣW(i) W(x) returns the size of the waveform i=0
:
WaveData f

●Details

Next, we will analyze each item in detail.

◆Mixer part format

Mixer has 1 byte per channel, 7-5 bits are assigned Effect Send Level, and the rest is assigned Effect PAN.
The size will be 18 bytes.

	7	6	5	4	3	2	1	0
channel 0	EffectSendLev				EffectPAN				1 byte
:	:				:				:
channel 17	EffectSendLev				EffectPAN				1 byte

◆Velocity part format

The Velocity format requires that in addition to the parameters set on the tone editor, another parameter must be calculated for each point.
Please refer to Sound Tool Guide Vol1 for information on this parameter.
The size will be 10 bytes.

Slope encoded value 0	1 byte
Velocity Point 0	1 byte
Velocity Level 0	1 byte
Slope encoded value 1	1 byte
VelocityPoint 1	1 byte
Velocity Level 1	1 byte
Slope encoded value 2	1 byte
Velocity Point 2	1 byte
Velocity Level 2	1 byte
Slope encoded value 3	1 byte

◆PEG part format

Since operability on the tone editor is prioritized, the value to be edited and the value to be sent are assigned different values.
The level of the PEG to be edited is the table number.

long LevelTable[128] = { 
 	     0, 125, 250, 375, 500, 625, 750, 875,  
 	  1000, 1125, 1250, 1375, 1500, 1625, 1750, 1875,  
 	  2000, 2125, 2250, 2375, 2500, 2625, 2750, 2875,  
 	  3000, 3125, 3250, 3375, 3500, 3625, 3750, 3875,  
 	  4000, 4250, 4500, 4750, 5000, 5250, 5500, 5750,  
 	  6000, 6250, 6500, 6750, 7000, 7250, 7500, 7750,  
 	  8000, 8250, 8500, 8750, 9000, 9250, 9500, 9750,  
 	 10000, 10250, 10500, 10750, 11000, 11250, 11500, 11750,  
 	 12000, 12500, 13000, 13500, 14000, 14500, 15000, 15500,  
 	 16000, 16500, 17000, 17500, 18000, 18500, 19000, 19500,  
 	 20000, 21000, 22000, 23000, 24000, 25000, 26000, 27000,  
 	 28000, 29000, 30000, 31000, 32000, 33000, 34000, 35000,  
 	 36000, 37000, 38000, 39000, 40000, 41000, 42000, 43000,  
 	 44000, 45000, 46000, 47000, 48000, 49000, 50000, 51000,  
 	 52000, 54000, 56000, 58000, 60000, 62000, 64000, 66000,  
 	 68000, 72000, 76000, 80000, 84000, 88000, 92000, 95990};

From this table and PEGTime

	PEGRate=LevelTable/PEGTime;

Calculate and find PEGRate. From this, find the table number and use it as the transmission data.
The PEG rate table is written in C code as follows.

	long PEGRateTable[128];

	/*Create PEG rate table (100 times the actual value)*/

	for(i=0L;i< 128L;++i)
		PEGRateTable[i]=(long)(((float)i*(float)i*(float)100)	
				/(float)16384*(float)100); /* Find the slope */

Delay table value	1 byte
OffsetLevel table value	1 byte
AttackRate table value	1 byte
AttackLevel table value	1 byte
DecayRate table value	1 byte
DecayLevel table value	1 byte
SustainRate table value	1 byte
SustainLevel table value	1 byte
ReleaseRate table value	1 byte
ReleaseLevel table value	1 byte

◆PLFO part format

When actually editing this parameter in the tone editor, do not use the actual value, but use the table value. The edited value will be written as is.

Delay table value	1 byte
PLFO frequency table value	1 byte
AMP table value	1 byte
Fade timetable value	1 byte

◆Voice part format

7	6	5	4	3	2	1	0
FM		Bend range width (0 -13)						1 byte
×		Portamento time (0 -127)						1 byte
×		Number of layers						1 byte
Volume Bias(-127 - 127)								1 byte
Layer data 0								32 bytes
:								:
Layer Data n								32 bytes

FM: For checking whether this voice is in FM format

◆Layer part format

Layer parts are included in Voice format.
Some parts of the format are written directly to SCSP registers.
This part is a little complicated, so I'll explain it later.
Data size is 32 bytes.

7	6	5	4	3	2	1	0
×	Start of MIDI note (0 -127)							1 byte
×	End of MIDI note (0 -127)							1 byte
SCSP register part 23 bytes								23 bytes
Base note (0 -127)								1 byte
Fine tune (-63 - 63)								1 byte
For FM connection 1 (0 -127)								1 byte
FM connection 2 (0 -127)								1 byte
Velocity number (0 -255)								1 byte
PEG number (0 -255)								1 byte
PLFO number (0 -255)								1 byte

●About SCSP register part

Since this register is written directly to SCSP, one byte is not reserved for each parameter.

reference

"HARDWARE MANUAL"/SCSP User's Manual/ ■Slot-specific control registers

15	14	13	12	11	10	9	8	7	6	5	4	3	2	1	0
PEG	PLFO	-	(1)	(2)	(3)		(4)		loop control		8Bit	StaetAddress(19:16)
Waveform Start Address(15-0)
Loop Start Address(15-0)
Loop End Address(15-0)
DecayRate2(4-0)					DecayRate1(4-0)					EGHold	AttackRate(4-0)
-	(Five)	KeyRateScaling(3-0)				DecayLevel(4-0)					ReleaseRate(4-0)
LFO M	PEG M	PLFO M	-	-	-	(6)	(7)	TotalLevel(7-0)
FM Modulation(3-0)				1(5-0) for FM connection						2(5-0) for FM connection
-	(8)				TMod	(9)
LFOR	LFO frequency (4-0)					PLFO waveform		PLFO depth (2-0)			ALFO waveform		ALFO depth (2-0)
-	-	-	-	-	-	-	-	-	EffectSelect(3-0)				EffectSend(2-0)
DirectSend(2-0)		DirectPAN(4-0)

Parameters written in abbreviations due to space limitations

Abbreviation	Contents
8Bit	Waveform sample size (sets when 16Bit)
PEG	PEG switch
PLFO	PLFO switch
LFO M	LFO modulation
PEGM	PEG modulation
PLFOM	PLFOM modulation
TMod	Total level modulation
LFOR	LFO reset

This table only shows the parameters used in SEGA SatToneEd Ver1.00. The unused parameters are (1) to (9) in the table, and their contents are as follows.

number	Name	Contents
(1)	KEYONEX	Key ON execution (written by driver)
(2)	KEYONB	Key ON, Key OFF registration (written by driver)
(3)	SBCTL	Source bit control (written by driver)
(4)	SSCTL	Control of sound source (written by driver)
(5)	LSLNK	Loopstart link (written by driver)
(6)	STWINH	Stack write protection (written by driver)
(7)	SDIR	Sound Direct (written by driver)
(8)	OCT	Octave (written by driver)
(9)	FNS	Frequency number switch (written by driver)

◆Waveform part format

The waveform format stores waveforms arranged in a predetermined waveform order.
There is no header part, it is pure audio data.
All playback information, waveform addresses, etc. are stored in layers.

┌─────────────────┐
│ Waveform data 1 │
├─────────────────┤
│ Waveform data 2 │
├─────────────────┤
│ Waveform data 3 │
├─────────────────┤
│        :        │
├─────────────────┤
│        :        │
├─────────────────┤
│ Waveform data X │
└─────────────────┘