MML is abbreviation of Music Macro Language, and the productivity package only for the music to express the score by using the character string such as "C4D8E8". It is matched it to the MML specification for a personal computer as past as possible for a lot of basic commands. Abundant special options in addition to a past MML command are prepared along with the difference point and the function addition of an internal function. When a peculiar option to this MML is not specified, it is possible to perform by inputting an old MML program list as it is (Or, some parameters are corrected) because it is almost compatible. The note number has been matched to the MIDI standard. Next, the correspondence of the sound name and the note number on the correspondence table and the score with the tone is shown.

It is C0~B8 to be able to specify the sound name with MML. Please pronounce the note outside the range by "N" command.
The position of "# (sharp sign)" is a sample for the MML explanation.
Sound name on music score and figure for note number
"S" command of reference ��MML (mixture specification).

It is a memory only for the sound processing of 512 Kbyte that CPU(MC68EC00) only for the music of Sega Saturn manages. The default song data for the STANDARD tone data and the BEEP instruction is set at the same time as BASIC's being started, and the MML development area is initialized. The tone data is loaded directly from CD-ROM. The song data is transmitted to the sound memory by inputting in the form of the MML character string, and executing the PLAY/EFFECT/SETSONG instruction. Because the variable area of BASIC is not occupied like past MML, the free area check function "SFRE()" and "TFRE()" only for the sound memory are prepared.

It is a set of data to sample musical instruments and the effect sounds by the PCM method. "GM_FM1" "GM_FM4 -" is a tone file (General MIDI array) assembled by the frequency modulation oscillator method. The tone data file is loaded from CD-ROM by the PLAYINIT instruction and used. The tone data of tens of kinds of is stored in one file. An easy explanation of each tone data can be displayed by the TLIST instruction (The list of the tone file that can be specified is displayed by TLIST "MENU"). Can it rearrange between each tone files, and the original PCM sampling data be registered by the TARG instruction.
The file of the AIFF form and the WAVE form is taken recognizing it automatically, and the original PCM sampling data can be used as tone data. Moreover, it is also possible to give wavy data without the header by the DATA initialization statement. There are four 12 PCM sound source sets + frequency modulation oscillator set tone data file prepared by the standard. Please refer to "Details of 2.8 tone data" for details.

The tone data registered in the tone data file is sampling of the shape of waves of the sound by the PCM method. The wave data registered by the TARG instruction is eight bits/16 bits of "AIFF" form or "WAVE" form (monaural). A standard sampling frequency of the tone data is 44.1 It is possible to correspond also to the data of a sampling frequency different from the Sega Saturn standard such as 11KHz and 22Khz because the pitch can be able to be bought freely by the TPROP instruction though it is Khz. When the wave data is uttered, LFO and EG, etc. can process various tones.

When PLAY/EFFECT or the SETSONG instruction is executed, the batch conversion is done by "Song data" to be able to recognize CPU that manages the sound memory as for the MML character string. The MML character string for 16 tracks or less is progressed to one song data, and it is stored in the sound memory. The number of song data that can be had in the sound memory (number of songs) and each size is decided at a time by the PLAYINIT instruction. It is possible to perform by being able to register up to 251 (for + PLAY instruction for "BEEP +"), and calling it even times how many by the EFFECT/BGM instruction as long as the sound memory is not initialized by the PLAYINIT instruction.

It is a serial number of the song data in the state registered in the sound memory. Only the 0th is for the BEEP instruction, and it is also possible to register and to sound usual MML here. Because the song data for the BEEP instruction is called at Break when making an error, the sound of the error (tune) can be freely changed. When the song number is omitted, the first song data becomes the object of the operation because the default of each instruction of which the parameter is the song number is "1".

It is a management number needed when the song data is sounded at the same time. This sound systems can be performed up to eight at the same time by arbitrarily choosing from among 251 or less (The effect sound is included). When it is allocated in either of eight sequences when the performance is begun by the PLAY/EFFECT/BGM instruction, and all the sequence control instructions (PLAYFADE and PLAYSTOP, etc.) are executed when the sequence control instruction (PLAYFADE and PLAYSTOP, etc.) is thereafter executed, it is possible to execute it in real time by specifying all the sequence numbers. The sequence number becomes "0~7" because there are eight kinds. The BGM instruction specially uses the sequence number "0" by the exclusive use. As for the sequence number "1~7", the place that has become vacant by the PLAY/EFFECT instruction is dynamically allocated. "Numeric variable" to return which sequence number the specified song data was allocated by the system can be specified for the EFFECT instruction.
For instance, the performance of the following format begins first to put the fade processing after it begins to perform.
"MML character string" EFFECT(I)
Because the allocated sequence number is stored in a numeric variable "I", the parameter passed to the fade processing instruction "PLAYFADE" is done specifying "Numeric variable" as follows.

PLAYFADE I, 127, 50
Thus, there is no necessity that the user considers because the sequence number mediates a numeric variable and is passed at all.

It is a number done to do it is time when in which the track where one song data is composed is distinguished. In this MML, it is possible to describe it up to 16 tracks or less. The harmony specification in the same track is enabled in this MML though it was necessary to divide the track in past MML when harmony is pronounced.
When two or more tracks are used, the youngest track number is especially called a base track.

The layer in the sound system is a unit of the data of a more detailed sound to compose one tone of the tone data. The layer is internally divided in this case by each wave data though this slot has 32 pronunciation slots in Sega Saturn, and has something that sounds one layer data a simultaneous plural it and pronounces. Therefore, a simultaneous number of pronunciations decreases more than 32 pieces because two layer data exists, and two slots are used at the same time when pronouncing in case of the tone in which two wave data is sounded at the same time. The wave data of a different sampling area is allocated in one tone like the tone of "PIANO" set in this case, the slot for a few minutes of the layer is never occupied because it doesn't sound it at the same time also though the layer has increased like the drum set that much in case of the tone into which the kind changes according to the note number. However, the occupation of the slot happens because the thing as which "Start Note" that is an internal parameter of the layer data and "End Note" are the same is pronounced at the same time mutually. Because each one of career/Mojurata is allocated to one slot when the frequency modulation oscillator is used, the tone made by four operator algorithm will have four layer data. A simultaneous number of pronunciations decreases more than 32 sounds to use two or more slots for one sound according to the method of specifying the slot algorithm. For instance, 32 slots are meant and when wires is connected, it is meant to pronounce only by one sound at the same time the entire series. If an extreme example is enumerated, sounding FM of 32 piece connecting wires when sounding it with PCM by one sound becomes impossible for FM because it cancels the pronunciation when the number of slots done to do by the algorithm connected is insufficient even by one.
A detailed edit of each layer can be done by the TARG instruction and the TPROP instruction.

Even if DSP is not passed or is passed, the sound in which effect is not put is called a direct data (Or, dry data).

To the sound generated in the sound source part  When some effect is put through DSP, the sound obtained as a result is called an effect data (Or, wet data).

It is a function to provide Localization: ..location.. ..(.. sound is a direction in which ..listening.. and a position). There are two kinds of this sound systems (the direct bread (location of the direct data) and the effect bread (location of the effect data)).
It divides into three kinds, and the priority level has been decided though direct bread information is set by the one and the PLAYPAN instruction set by "P" command of the one and MML set to the tone data beforehand. Please refer to the following "Priority level of the direct bread" for details.

The sound is made PCM data, it saves in the memory, and the sound source of the method to read from the memory when performing and to output the sound is indicated. PCM is to divide the sound (shape of waves) according to the time axis, and to replace each wave high price with digital data. The data obtained because of this work is called PCM data.

The frequency modulation is shown. In the meaning of sound source, the frequency modulation oscillator is indicated.

The function generator that utters the shape of waves of a frequency that is lower than the range where I hear to man's ear is indicated. A wavy output of this LFO is used to express the tremolo and Bibrard for the modulations of the amplitude and the frequency, etc.

The operation (multiplication and addition) is chiefly done at high speed, and the circuit only for a high-speed operation. Internal organs DSP is used for SCSP(Saturn Custom Sound Processor) by the one that it specialized in the function for audio as the means to produce the effect that is called sound effect such as echoes and reverberations.

The number of data of waveform data is shown. It is one sample 16 bits in case of one sample eight bits and sampling 16 bits in case of eight bit sampling.

It is a unit where the size of the volume etc. was shown by using the logarithm for the sake of comparison. Concrete..standard..become..volume..shape of waves..amplitude..object of comparison..become..volume..volume..compared with..become.Moreover, one time :.   [0dB] and three times become 10[dB] etc.Oppositely, when the volume that becomes an object of comparison is smaller than the volume that becomes a standard, this value [dB] takes a negative value (value of the minus). For instance, it ..6[dB].. increases for the standard by a factor of 1/2-. In this sound system, "@V" command of MML and [ "" the command" ] The unit at the command, the modulation level, and the total level ..".. hits this.

When two or more slots are connected in the frequency modulation oscillator, the slot on the side where the modulation is put is indicated.

When two or more slots are connected in the frequency modulation oscillator, the slot on the side on which the modulation is put is indicated.

The time that the key has been actually turned on is shown.

It uses it as a word that shows time length of one note in this manual though the interval between the command and the command through all tracks is shown. The unit is a direct length of the MML command "L".

The ratio at the gate time and the delta time is shown. For instance, it is shown to perform by the staccato because only 1/2 of the lengths of the note (delta time) the key is turned on in case of 1/2.

It is a waveform data of Macintosh standard format. It queues up in order in eight high rank bits and eight subordinate position bits at the time of 16 bit sampling.

It is a waveform data of Windows standard format. It queues up in eight subordinate position bits and order of eight high rank bits at the time of 16 bit sampling.

It is a standard to exchange the signal of the synthesizer and electronic musical instruments and computers such as keyboards. This sound system has matched the note number and the pronunciation method to this standard.

It is a note. The sound of C of octave 4 is allocated in 60, and the height of the sound is expressed by the number of 0~127.