Spec of 2.1 sound system
32 slots (FM and PCM using combinedly)/16 tracks
. 32 sound or less pronunciation simultaneously in case of one tone one layer.

251 songs or less
253 songs for together for "BEEP" and for the PLAY instruction.

26000 notes or less
. When you do not specify special effects of the pitch pipe bend etc.

Tone data area: Standard 340 Kbyte: About maximum 462 Kbyte
. Capacity that can be treated at the same time without replacing it by on memory

It is possible to use it up to 128 tones or less (one tone 128 layers or less) simultaneously.
. In the general synthesizer term, it corresponds to patch/performance.

Two maximum pipe bend cooking stove +- octaves
. It is possible to change by the TPROP instruction.

DSP control for effect
. The reverberation, DeLay, and the distortion can control with MML.

200 kinds or more of PCM sound source and 512 kinds of frequency modulation oscillator standard(The effect sound contains it. )
. It is possible to rearrange it by arbitrarily selecting it from 16 kinds of tone data files.

Frequency modulation oscillator edit command equipment
. Another of 47 kinds of modulation wave forms for which 32 operators are freely prepared by can connect wires and the standard and the original PCM sampling data can be used as a modulation wave form.

The spec is detailed.
Voice synthesis method
. PCM voice synthesis and FM voice synthesis. The noise sound source (Coexistence can be used).

Sound source part
. Ri sampling frequency 44.1KHz (fixed reproduction rate) fixation (CD quality at sampling frequency 44.1KHz and 16 bits).

Waveform data format
. Straight line and 16 bits straight line. eight bits
. Sampling frequency 0(DC) ~ 44.1KHz.
. For sign bit reversing function of waveform data

Number 32 of voice processing slots
. One slot corresponds to one operator a sound of the PCM sound source of the frequency modulation oscillator.
. Maximum and simultaneous ..pronunciation number.. 32 (eight sounds when FM of all four operator algorithm is generated).
. The number of slots used for one FM sound sound can arbitrarily set even 1~32. (It is possible to exist together about FM sound and the PCM sound. )
. It manages it collectively by the layer data.

Wavy loop function
Each layer
. Normality
. Reverse
. Oltanatib loop
Correspondence both (PCM/FM of ..twining.. selection)

EG (Emberorb generator)
. Four segments Amplitude-EG (correspondence both PCM/FM).

LFO (Or Frikenshe oscillator)
. For amplitude/frequency modulation (tremolo/vibrato)
. Kind of LFO output wave form
Saw tooth wave
Rectangular wave
Triangular wave
White noise
. It is ALFO/PLFO possible independently I (correspondence both PCM/FM).

Tone data(tone file of standard equipment)
. The memory capacity is a trade-off of the song, the data area, and the DSP program area (Refer to "2.3 sound memory map").
. It is possible to combine freely from among the tone file (frequency modulation oscillator file of 340 Kbyte*12= about 4MB+ array of GM).
. The reproduction that changes registration and the pitch of the original PCM sampling data is possible (It is also possible to combine with the data of the standard tone file).
The file of "WAVE" form of "AIFF" form can be linked with the tone data (automatic identification).
. The wave data without header information can be registered by "DATA" sentence.
. ALFO/PLFO/4 segment EG can control ..the unit of the layer (each slot)...

DSP/mixer control(input mixer/output mixer)
"DSP" At the mode (single effector mode)
DSP effector program
STEREO REVERB 2 (reverberation)
Kind
STEREO DELAY (DeLay)
DISTORTION (distortion)
It drinks and one kind can be selected from Tuu in real time by "Ec: the effect change" command of MML.
"DSPEXT" At the mode (multi effector mode)
DSP effector program
STEREO REVERB (reverberation)
STEREO DELAY (DeLay)
DISTORTION (distortion)
It is allotted to another DSP input mixer of Bets however, and the effector of each layer can be used ..choices.. at the same time. The bread jar of each output channel can effect level/be set.

Shape of waves for modulation of frequency modulation oscillator
The width of the expressible tone extends because 47 kinds of in total besides a general shape of waves such as a signature wave, Nocogiri wave, rectangular waves, and triangular waves shapes of waves for the modulation of the frequency modulation oscillator are prepared. Please refer to the explanation part of TARG instruction "Instruction related to 2.5 sounds" (FM modulation wave form chart) for the shape of the modulation wave form.

Feature of MML
Past MML and high rank compatible almost
. As for the MML command to operate the register directly, because the meaning of the register is different, it is the interchangeability none.
. It is possible to specify it up to 16 tracks or less.

The sign can be specified.
. There is a concept of the bar and the temporary change sign (If "S" command is not specified, it is compatible with past MML).

Suboption for note on command (sound name specification)
. Specification compared with velocity specification/Gate/Delta.
. Pitch pipe bend master-apprentice (auto bread and auto volume/using combinedly).
. Code name specification.
. The suboption is compatible with past MML if it doesn't specify it.

Damper switch of each track
Harmony specification in one track(It is possible to specify it even because of a different tone. )
. Ren, the Thailand specification, and the pitch pipe bend operation of harmony are possible.

DSP effector/mixer control command
. When the input mixer setting to the DSP block is done in each layer, the level and the location of the effect element can be controlled with MML.
"DSP" The effector can be selected by the effect change command "Ec" at the mode ("DSPEXT" Fix "Ec0" at the mode).

Macro-function that can be redefined
. Velocity macro (p.mp.mf.f etc.).
. Free definition 128 character *52 macros (effective in percussion definition).

Key transformer pose function
. Loop structure of each semitone

Loop structure of easy ceremony specification
. Loop structure based on abbreviation method of not no simple loop, and subroutine but easy ceremony.

There is a burial control command to take the main system and synchronization.
The performance for 16 tracks can be in real time observed.
. The interruption can be observed for the track "0".

About the sound system for GAMEBASICforSEGASATURN
Because the program that does the parsing of BASIC and the program that does an actual performance are controlled with a separate system in GAMEBASICforSEGASATURN, MML and some specifications that have been used with a past personal computer etc. are different.

As for past MML, the specification of the command is different in the hoop and this MML where the register setting command to control PSG and the frequency modulation oscillator directly is prepared according to the difference of constructional of hardware (PSG doesn't exist in hardware). The memory capacity of Sega Saturn is a design that also greatly enhances the function also by the sound system including MML because it is vast, includes the function of a sequencer on the market and the synthesizer as much as possible, and values music or more compared with the personal computer when past MML is installed. If the option is not specified because this is a method to add and to specify the option for a basic command, the same operation as past MML is done.

In the tone data file set, because there is an internal organs file, too making the music scale because of the effect sound, and making the effect sound by using tones of musical instruments oppositely are possible the effect sound data that can be used as it is.

The sound source hardware of Sega Saturn is a design that can be reproduced by individually specifying the tone for 32 slots. It doesn't allocate in one track one slot, and 32 slots are used effectively while dynamically doing assign though this system is adjusting the number of tracks that can be described with MML to 16 or less (As well as MIDI).

The sound data described with MML converts into the song data beforehand by PLAY, EFFECT, and the SETSONG instruction before it performs, and it forwards it to the sound memory. Therefore, the time lag by the tune change and the effect sound request is stopped at least.

The tune is limping while parsing is being done, and it drags on and timing never goes mad because the music performance and the parsing of BASIC operate in parallel by a separate system. However, it is likely to drag on only at the time of parsing on the way because the following instruction is not executed only by the PLAY instruction until each MML character string described in one line is performed, and the performance ends. When the program that assumes the game and the music performance to be the main is united, the timing gap between tracks doesn't happen, and the form that calls arbitrary song data by EFFECT and the BGM instruction after the song data is prepared in the sound memory beforehand by the SETSONG instruction (251 or less) and performs is taken. It never mutually influences even if it is made to perform while moving the polygon character because it performs completely independently of main BASIC system when it can be EFFECT and BGM dispirited and performs.

When the DSP effector is used, "Input mixer" setting is needed on the tone data side (There is a tone set by default, too).
Attention
Please use it after often reading "Block composition of 2.4 sound system".

There is "Burial control command" to take the main system and synchronization. The command of another and the trigger form to be able to acquire "Note", "Volume", and "Location" performing now in real time can be sent from the sound system to the BAISIC inter printer. Only the control command described in the track "0" can observe interrupt from the inter printer side. It is also possible to move the polygon while synchronizing with the karaoke program and music by this function.