JPH0248699A - Envelope controller - Google Patents

Abstract

PURPOSE:To damp musical sounds without causing an unnatural noise by holding key-off operation information at the time of the occurrence of this information in a prescribed period and generating a natural damping waveform thereafter. CONSTITUTION:When one of plural depressed keys on a keyboard 2 is released in the period of a quick damping waveform part in the repeat playing mode, a CPU 5 performs the control so that a musical sound signal generating part (TG) 8 is not turned off at the key-off time, and this state is kept for a while. Thereafter, the CPU 5 performs the key-off control of the TG 8 with respect to all sound producing channels in a timer interrupt handling routine to naturally damp a repeat waveform along the natural damping waveform part. Thus, the CPU 5 naturally damps musical sounds without unnaturally changing them.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an envelope control device, and is suitable for application to electronic musical instruments that generate musical tones with effects such as repeat effects and twin mallet effects. .

[Summary of the invention]

The present invention is an electronic musical instrument designed to generate an envelope musical tone having a plurality of repeating waveforms, in which when key-off operation information occurs during a predetermined period, it is held and then a naturally decaying waveform is generated. As a result, the musical tones can be attenuated without causing unnatural noise.

[Conventional technology]

In conventional electronic musical instruments, as a method of imparting a repeat effect to musical tones, a repetitive waveform is generated by resetting the address of an envelope counter with repeat pulses generated at a predetermined period in a repeat pulse oscillator. has been proposed (
(Special Publication No. 57-29715).

[Problems to be Solved by the Invention] However, with this configuration, when a repeat pulse is generated from the repeat pulse oscillator, if the musical tones generated by the previous performance have not been completely attenuated and remain. Then, when a new repeated waveform is generated, there is a problem in that a click sound is generated because a discontinuous point is created in the envelope between the previously remaining musical tone and the new musical tone generated by the repeated performance.

As one way to solve this problem, for example,
Utilizing the configuration disclosed in Japanese Patent No. 9-136788, the generation of click sounds can be prevented by rapidly attenuating the musical tones that have been generated up to that point immediately before the repeated musical tones are generated. It is possible to configure

However, in this case, when the performer releases the key from the state Qi in which the key is pressed (this is called the key-on operation state) (this is called the key-off operation),
Since the musical tone that has been generated up to that point decays at a high speed along the rapid decay waveform and then suddenly stops, there is a risk that the musical tone may end up giving the impression that it cuts off unnaturally.

This invention was made in consideration of the above points, and it is possible that when a performer performs a key-off operation, an unnatural change occurs in the musical tone (such as a change that generates unnecessary noise or suddenly stops). This paper attempts to propose an envelope control device that prevents this from occurring.

[Means for solving problems]

In order to solve this problem, in the first invention, key-on operation information 5TBUF (i) 145m obtained by key-on operation and key-off operation and key-off operation information 5TBUF (i) obtained by key-off operation
) □, multiple repetitive waveforms WAVEI, WAV
In an electronic musical instrument designed to generate an envelope musical tone with E2, key-on operation information 5TBUF
(i) Musical key-on processing information KEYBU based on □.
F(i)0. musical tone key-on processing information generating means (5, RT3) that generates key-off operation information 5TBUF (i);
□, musical tone key-off processing information KEY B U
F (i) musical tone key-off information processing means (5, RT5) that generates xts, and musical tone key-on information KEYBUF
(i). 3 and musical key-off information KEYBLJF (
i). , given multiple repetitive waveforms WAVEI
SWAVE2 also includes musical tone signal generating means (5, 8) that generates a Zenherop musical tone signal, and a repetitive waveform WAVEI.
, when key-off operation information 5TBUF (i) □ occurs at timing TIM=0 to TIM=15 corresponding to a predetermined period in the generation period of WAVE2, the predetermined period TIM=0 to TIM=15 elapses. Musical tone key until
Off processing information KEYBUF (i) 145m is held and a musical tone signal S having a naturally decaying waveform W15 is generated from the musical tone signal generating means (5, 8) at the elapsed time TIM=15.
Musical tone key-off processing information retention means (5,
RT5.5P28).

Further, in the second invention, a plurality of repetitive waveforms WAVE
I, a rapid decay waveform forming means (5, 5P26) that forms a rapid decay waveform Wll in the decay waveform portion W14 of WAVE2.
), musical tone key-off processing information retention means (5, RT5
．． 5P28) holds the tone key-off processing information KEYBUF (i) ms++ during the period TIM=0 to TIM=2 during which the rapid decay waveform Wll is formed.

Further, in the third invention, the musical tone key-off processing information holding means (5, RT5.5P28) stores a plurality of repetitive waveforms W.
During the period in which the attack waveform portion W12 of AVEI and WAVE2 is being formed, the musical tone key-off processing information KEYBUF
(i) Put Mss on hold.

[Effect]

Among the repetitive waveforms WAVEI and WAVE2, the musical tone key-off processing information KEYBUF (1)xsm is transmitted to the musical tone signal generating means (5,
8), and then the musical tone is attenuated so as to exhibit a natural attenuation waveform, so that the musical tone can be naturally attenuated without causing unnatural changes in the musical tone.

〔Example〕

An embodiment of the present invention will be described in detail below with reference to the drawings.

[1] First embodiment (1) Overall configuration In FIG. 1, l indicates an electronic musical instrument as a whole, and key information KIN input from the keyboard section 2 and the panel operation section 3
While processing the operator information PIN input from the central processing unit (CPU) 5 which executes the main routine shown in FIG. 4 based on the program data stored in the program memory section 4 having a ROM configuration, The data is loaded via a bus 6 into a register section 7 (FIG. 2) consisting of a group of 8-bit registers having a RAM configuration.

In the case of this embodiment, the electronic musical instrument 1 can be used for a plurality of devices at the same time, for example, 16
When a plurality of keys are pressed simultaneously, the CPU 5 assigns the data of each key to one of the 16 sound generation channels to correspond to each key pressed. It is designed to be able to generate musical tones. The CPU 5 sends the musical tone control information MUS corresponding to each key operated by the key-on operation to the bus 6.
By supplying the signal to the musical tone signal generating section (TG) 8 via the musical tone signal generating section (TG) 8, the musical tone signal SND corresponding to each key is generated from the musical tone signal generating section (TG) 8, and this is converted into a musical tone in the sound system 9.

In the case of this embodiment, as shown in FIG. 3, the pitches CI, ..., cz % C3 corresponding to 61 keys are
”,...,B, ,C,respectively, key code 24,...,48.49・old...,8
3.84 has been assigned.

When the power is turned on, the CPU 5 executes the main routine RT shown in FIG.
0 and executes an initialization routine in step RTI, thereby initializing the electronic musical instrument 1 as a whole. At this time, each register of the register section 7 is reset.

After that, in step RT2, the CPU 5 determines whether or not there is a key-on operation for any key (this is called a key-on event) based on the key information KIN of the keyboard section 2, and obtains an affirmative result. When the key-on event processing routine RT3 is executed, the process moves to step RT4, and when a negative result is obtained, the process jumps to step RT4.

In step RT4, the CPU 5 determines whether or not there is a key that has been operated as a key-off event (this is called a key-off event) based on the key information @KIN of the keyboard section 2, and when a positive result is obtained, the CPU 5 turns off the key. After executing the event processing routine RT5, the process moves to step RT6, and when a negative result is obtained, the process jumps to step RT6.

In this step RT6, the CPU 5 determines whether or not the performer has performed an operation to change the tone designation (this is called a tone change operation) based on the operator information PIN of the panel operation section 3, and if the result is affirmative, When a negative result is obtained, the process executes the tone color change process routine RT7 and then moves to another process routine RT8, whereas when a negative result is obtained, the process jumps to another process routine RT8.

In other processing routine RT8, the CPU 5 executes processing other than key-on event, key-off event, and tone color change processing, such as volume processing, and then returns to step RT2 described above.

Then, the CPU 5 forms a cyclical processing loop LOOP1, and by repeating the processing of this cyclical processing loop LOOPI, it repeatedly generates musical tones corresponding to the performer's performance operations.

(2) Processing of key-on event When the performer performs a key-on operation while the loop LOOP 1 is being circulated, the CPU 5 executes the processing shown in FIG. 5 in the key-on event processing routine RT3.

First, the CPU 5 stores the key code of the key in which the key-on event occurred in the step SPI in the register section 7 (second
The event key code data NKYL! is stored in the event key code register REG3 in the figure). Import as m,,4m.

Event key code register REG3 stores event key code data N consisting of 7-bit key code data in the 0th to 6th bits of the 8-bit memory area.
KYL, ~1. remember.

Next, the CPU 5 moves to step SP2 and registers the register section 7.
Performance mode data MO of performance mode register REG5
DE is read and it is determined whether the content is O or not.

In the case of this embodiment, the CPU 5 allows the player to play the musical tone signal 1; This can be specified by selecting a tone, and when the performer specifies normal performance, repeat performance, and twin mallet performance, the CPU 5 outputs three-value data as performance mode data MODE, that is, MOD.
E= rOJ, rlJ, "2" is maintained.

Here, the timbre parameter data for the input specified timbre and the conversion data for converting the timbre into a performance mode are stored in a conversion table of ROM configuration and the timbre parameter memory section 10, and the CPU 5 stores the timbre parameter data for the input specified timbre and the conversion data for converting the timbre into a performance mode. At the same time, the corresponding tone parameter data is transferred to the musical tone signal generator (TG) 8, and the performance mode data MOD
E is held in the performance mode register REG5 of the register section 7.

Here, the CPU 5 is in normal performance mode (MODE=rO
J) is specified, Figures 6(A) and 7(A)
As shown in (B), after the key-on operation and the key-on signal KON (included in the key information KIN) rises to logic "1", until the key-off operation is performed,
A sound system 9 generates a musical tone having an envelope such that musical tones of all keys are simultaneously and continuously produced.

Moreover, the CPU 5 is in the repeat performance mode (MODE=rlJ
) is specified, Fig. 6(B) and Fig. 8(A),
As shown in (J), during the period when the key-on operation is performed and the key-on signal KON rises to logic rlJ,
An envelope (rapid decay waveform part W11,
A musical tone having a repetitive waveform WAVEI having an attack waveform part W12, a sustain waveform part W13, and a decay waveform part W14 is generated from the sound system 9.

Furthermore, CPU5 is set to twin mallet performance mode (MODE).
= r2J') is specified, the key-on operation causes the key-on signal KON to rise to logic "1", as shown in Fig. 6 (C) and Fig. 8 (A), (J), and (K). During this period, the first musical tone is divided into a first musical tone consisting of the lowest musical tone among the keys that are being sounded simultaneously, and a second musical tone consisting of other musical tones, and the first musical tone and the second musical tone are divided into a predetermined musical tone. (For the first musical tone, a repeating waveform WAVE1 is formed that has a rapid decay waveform part Wll, an attack waveform part W12, a sustain waveform part W13, and a decay waveform part W14.) Also, for the second musical tone, a repetitive waveform WAVE2 having a rapid decay waveform part W21, an attack waveform part W22, a sustain waveform part W23, and a decay waveform part W24 is formed)
A musical tone having the following values is generated from a sound system 9.

If a positive result is obtained in step SP2 of FIG. Among the sound generation channels of 0 to 15), the sound generation channel with the oldest key-off operation time is searched among the sound generation channels to which the currently key-on operation key is not assigned, and the channel number data representing the channel is stored in the register section 7. It is held in the allocated channel register REG9 as allocated channel data ASS.

Next, the CPU 5 moves to step SP4 and registers the register section 7.
Of the pronunciation channel key code register REGI,
Sound generation key code data KEYBUF (ASS
) In Lss-61, event key code register RE
In addition to writing the G3 event key code data NKYLsm~, the sound generation on/off data KEYBUF (
ASS) Write logic “1” data to 01m.

The sound generation channel key code register REGI holds data for controlling the sound generation of the musical tone signal generator (TG) 8, and stores sound generation key code data consisting of 7-bit key code data in the memory area of the 0th to 6th bits. KEY
BUF (i) Lsm~,.

is stored in the memory area of the 7th bit.
1” (represents the control state in which musical tones should be produced, that is, the sound generation key is on), or logic “0” (represents the control state in which musical tones are to be attenuated, that is, the sound generation key is off). The sound generation on/off data KEYBUF (+)□ is stored.

Thus, the 0th to 6th bits of the sound generation channel key code register REGI contain sound generation key code data KEYBUF (
Ass) Lsm ++4m is held, and at the same time, the sound generation on/off data KEYBUF ( A
SS) □ is held.

After completing the assignment of the key in which the key-on event has occurred to the sound generation channel in this way, the CPU 5 controls the musical tone signal generation section (TO) in the next step SP5.
By giving musical tone control information MUS to 8, the sound generation key code data KEYBUF (ASS)L is set for the sound generation channel of the assigned channel data ASS.
After performing key-on processing on the musical tone signal of the key code represented by SI-61 (ie, the key code of event key code data NKY where a key-on event has occurred), the process returns to the main routine via step SP6.

On the other hand, if a negative result is obtained in step SP2, this indicates that the currently designated performance mode is the repeat performance mode or the twin mallet performance mode, and at this time, the CPU 5 moves to step SP7. For all the sound generation channels (i=0 to 15), the sound generation channel operation status register R of the register section 7
EG2 key-on/key-off data 5TBUF (i)
□.

Select the assigned channel from among the sounding channels i for which is logical "0" and register the assigned channel register REG.
9 as allocated channel data ASS.

Here, the sound generation channel operation state register REC2 holds data representing the currently requested operation state for the key of the key code assigned to each sound generation channel (i=Q to 15), and the 7th bit In the memory area of the memory area, there is a logic "l" (representing the player's key-on operation, i.e., the key-on operation state) or a logic "O" (the performer's key-off operation, i.e., the key-off operation state). The key-on/key-off data consists of 1 bit data of 5TBUF (representing that the
i) Store xs*.

Also, the sound generation channel operation status register REG2 is the sixth
Logic “1” in the bit memory area (key-on/key-off data 5TBUF (i), 4ss is logic rOJ
Therefore, even though the performer has performed a key-off operation, the sound on/off data KEYBUF (i)
Plsm is in the logic "1" state, therefore the musical tone signal generator (
Key-off request data 5TBUF (i )6m is memorized.

Furthermore, the sound generation channel operation status register REG2 has a logic "l" in the memory area of the 0th bit (indicating that the tone signal generator (TG) 8 is in a key-off prohibited state in which the key-off operation is not performed, that is, in a disable state).
Alternatively, a key-off disable enable data 5TBUF(i) 131 consisting of 1-bit data of logic "O" (indicating that the key-off prohibited state is not in question) is stored.

In this way, the CPU 5 generates the key-on/key-off data 5TBUF (i) in step SP7. m”rO
Searching for the sound generation channel i under the condition J and selecting the assigned channel means that the key-on data is assigned to one of the sound generation channels to which no key-on data is assigned.

Next, the CPU 5 moves to step SP8 and transfers the event key code data NKYts*~1 of the event key code register REG3 to the sound generation channel key code register REG1.
s) Ll. ~l and write logic “1” as sound on/off data KEYBUF (ASS) MSm.
Write data.

Thus, the CPU is in a state where it can perform key-on processing on the musical tone signal generator (TG) 8 for the assigned sound generation channel.

Next, the CPU 5 moves to step SP9, reads the event key code data NKYtsm~, 1 from the event key code register REG3, and sets 'Lsm-' to the lowest note key code data LSTK' held in the lowest note key code register REG4. Determine whether it is smaller than all.

If a positive result is obtained in step SP9, this means that the lowest tone key code data LST currently held as the lowest tone in the lowest tone key code register REG4.
K YLIm~6. This means that the event key code data NKYL, . At this time, in step 5 PIO, the CPU 5 selects event key code data NKYL, . . . , 6 . Add the lowest note key code data L S to the lowest note key code register REG4.
After rewriting the information on the lowest note by writing as TK YLs*~, the process moves to step 5PII.

In this way, the CPU 5 is configured to always hold the key code data of the lowest note in the lowest note key code register REG4 among the plurality of keys that are currently being keyed on at the same time, so that when the twin mallet performance mode is selected, the lowest note key code data It is designed so that the pronunciation can be distinguished between the ``nobu'' and the ``nobu'' group.

On the other hand, if a negative result is obtained in step SP9, this means that the musical note of the key in which the event occurred is not the lowest note, and in this case, the CPU 5 jumps to step 5PII.

This step 5PII is a step for determining whether or not there is any key that has been pressed on other than the currently assigned key, and the CPU 5 registers the sound generation channel operation status register REG2 for all sound generation channels i except the allocated channel 1=Ass. The key-on/key-off data 5TBUF (i) □3 held in )
, the CPU 5 returns to the main routine from step 5P12 and again executes the process for the key that has been turned on.

On the other hand, if a positive result is obtained in step 5PII (this indicates that no other keys have been pressed), the CPU 5 moves to step 5P13 and sets the time information data TIM of the time information register REG6 to a value O. At the same time, in step 5P14, a reset signal R5 is sent to the interrupt timer 11 (FIG. 1) via the bus 6.

(3) Timer interrupt processing The interrupt timer 11 performs timer interrupt processing as shown in FIG. 10 by sending an interrupt signal INT to the CPU 5 at predetermined intervals ΔT as shown in FIG. 9(A). When the interrupt processing routine RTIO is executed and the reset signal R3 (FIG. 9(B)) is received from the CPU 5, a new timing operation for the interrupt signal INT is started from that point.

That is, each time the CPU 5 receives the interrupt signal INT, it adds "+1" to the time information data TIM of the time information register REG6 (Figure 2) (Figure 9 (C)), and thus the CPU 5 uses the external interrupt timer. 11 time series.

At the same time, the CPU 5 clears the time information register REG6 in step 5P13 (FIG. 5) so as to start counting from the initial value O, and then clears the time information register REG6 in step 5P13 (FIG. 5).
14, by resetting the interrupt timer 11 with the reset signals R and S, the time information data TIM
The phase of the change is matched with the interrupt signal JNT of the interrupt timer 11.

Subsequently, the CPU executes a timer interrupt processing routine in step 5P15, and then returns to the main routine via step 5P16.

Here, the CPU 5 normally executes the timer interrupt processing routine RTI when the interrupt signal INT is generated from the interrupt timer 11.
After synchronization is forcibly performed in steps 5PI3 and 5P14, the first rapid decay process (when TIM=O) is executed in step 5P15 using the timer interrupt processing routine RTIO.

That is, when the timer interrupt processing routine RTIO is entered, C
As shown in FIG. 10, PU5 first performs step 5P2.
At step 1, it is determined whether the performance mode data MODE in the performance mode register REG5 (FIG. 2) is a value O or not.

If a positive result is obtained here, this means that the normal performance mode (FIG. 6(A)) is designated by the performer, and at this time, the CPU 5 performs step 5P.
22, the program immediately returns to the main routine. As a result, the CPU
5 is the time information data T in the normal performance mode.
A musical tone having a normal waveform WAVEO (FIG. 7(B)) is generated without responding to changes in IM.

On the other hand, if a negative result is obtained in step 5P21, this means that the player can play the repeat playing mode or the twin mallet playing mode (FIG. 6(B) or (C).
)) is specified, and in this case, CP
U5 moves to step 5P23 and determines whether there is a sound generation channel that requires processing.

That is, the CPU 5 first reads the key-on/key-off data 5TBUF (i) Msm (
i=o~15), key-on/key-off data 5TBUF (i) xs in all sound generation channels i
It is determined whether there is a sounding channel with m being logic "1". If a positive result is obtained here, this means that the key assigned to the relevant sound channel is in a key-on state, and as a result, some processing is required. At this time, the CPU 5 5
Move on to P24.

If a negative result is obtained, it means that there is no key in the key-on operation state in the sound generation channel operation status register REG2, and at this time, the CPU
5 is the main routine RTO (fourth step) from step 5P25.
) or key-on event processing routine RT3 (5th
Return to figure).

Further, in step 5P23, the CPU 5 secondly determines whether the sound generation on/off data KEYBUF(i)xsm held in the sound generation channel key code register REGI is logical "1". If an affirmative result is obtained here, this means that the CPU 5 is in a state in which the musical tone signal generator (TG) 8 is operating the sound generation section (TG) 8 in the corresponding sound generation channel i (that is, the sound generation key is on). cpus moves to step 5P24.

On the other hand, if a negative result is obtained in step 5P23, this means that the sound generation channel key code register REGI holds only data for the sound generation key off state, and at this time, the CPU 5
From P25, the process returns to the main routine RTO (FIG. 4) or the key-off event processing routine RT3 (FIG. 5).

Step 5P24 is a step of determining the value of time information data TIM of time information register REG6 (FIG. 2).
The CPU 5 generates time information data TI based on the determination result.
M is TIM=O1TIM=2, TIM=15, respectively.
Step 5P26.5 followed respectively when TIM=20
At P27.5P2B and 5P29, processing for adding an envelope to the musical tone is executed.

In this embodiment, the CPU 5 generates a first repetitive waveform WA as shown in FIG. 8(J) in the repeat performance mode.
The time information data TIM (8th
When TIM=O (FIG. 8(B)), as shown in FIG. 8(C), by executing the rapid decay subroutine (FDE EXE) in step 5P26, the musical tone is transmitted through the rapid decay waveform part Decay rapidly.

Here, the attenuation speed of the rapid attenuation waveform portion Wll is the attenuation speed (
In other words, the value is selected to be significantly larger than the damping speed (attenuation rate during natural damping).

Further, when the time information data TIM is TIM=2 (FIG. 8(B)), the CPU 5 executes the repetitive waveform key-on subroutine (RPKONl) in step 5P27 (FIG. 8(C)).
By executing the above, an attack waveform portion W12 and a sustain waveform portion W13 (FIG. 8(J)) are formed.

Also, when the time information data TIM is TIM=20 (the 8th
(B)) The CPU 5 executes the repetitive waveform key-off subroutine (RPKOFFl) in step 5P29 (Fig. 8 (C).
)), the decay waveform part W14 (eighth
Figure (J)) is formed.

Further, when the CPU 5 executes the rapid decay subroutine 5P26 in step 5P26 (TIM=0), as shown in FIG. After that, when the time/information TIM is TIM=15, step 5P28 is executed.
Disable end subroutine DISOF
By executing FI (FIG. 8(C)), key-off disable data 5TBUF (i) vsIl
is lowered to a logic "0" level.

(Then, the CPU 5 determines the timing (TIM=O) when the repetitive waveform WAVEI begins to form the rapid decay waveform portion Wll.
, attack waveform part W12, sustain waveform part W13
Even if a key-off event occurs, 5 TBU of key-off disable data of logic "1" is generated until the timing (T I M = 15) when the key-off event is first formed.
F (i) Musical tone signal generator (TG) 8 by LSll
In this case, the key-off processing is prohibited from being commanded.

In this way, the CPU 5 executes TIM=O12, 15, 2
At the timing of 0, step SP26.27.28
．． After executing the process of 29 or at other timing T
I M = 1.2~14.16~19.21~25 (
= 0) is confirmed in step 5P24, the time information register R is set in step 5P30.
Remainder number when the number obtained by adding “+1” to the time information data TIM of EG6 is divided by 25 (=(TIM+1) ・MO
After rewriting D25) into the time information register REG6 as new time information data TIM, the process returns to the main routine RTO (FIG. 4) or the key-on event processing routine RT3 (FIG. 5) from step SP31.

As described above, in the repeat performance mode, the CPU 5 controls the rapid decay waveform portion W11 and the attack waveform portion W12 while the time information data TTM cycles through TIM=0 to 25 (=O).
, the decay waveform part W1 passes through the sustain waveform part W13.
4, and then further time information data TIM forms TI
Decay waveform part W during the cycle from M=O to 25 (=0)
14 is attenuated at the natural attenuation rate, and thus one repetitive waveform WAVEI is formed while the time information data TIM changes for two periods.

On the other hand, when the CPU 5 is in the twin mallet performance mode, the CPU 5 uses the first repetitive waveform WAVEI (FIG. 8 (J)) described above for the repeat performance mode as an envelope waveform forming the lowest musical tone, and at the same time As shown in (K), the first repetitive waveform WAVEI
A repetitive waveform WAVE2 shifted by 172 cycles from the second repetitive waveform WAVE2 is formed based on the time information data TMM, and this second repetitive waveform WAVE2 is used for envelope control for musical tone groups other than the lowest note.

In this embodiment, the CPU 5 generates the second repetitive waveform WAVE.
2, as shown in FIG. 8(D), one period of time information data TIM (=O
~25 (=O)), when TIM=O, step SP2 of the timer interrupt processing routine RT10 (FIG. 10)
1, the rapid decay subroutine (FDEXE2) is executed to form the rapid decay waveform portion W21, and
At IM=2, execute the repetitive waveform key-on subroutine (RPKON2) of step 5P27 to form an attack waveform portion W22 and sustain waveform portion W23, and at TIM=20, execute the repetitive waveform key-off subroutine (RPKOFF2) of step 5P29. A decay waveform portion W24 is formed by this.

At the same time, as shown in FIG. 8(G), the CPU 5 raises the key-off disable data STB UF(i)t, ss to the logic rlJ level in the rapid decay subroutine (FDEXE2), and then at the timing of TIM=15. By executing the disable end subroutine (DISOFF2) in step 5P28, key-off disable data 5TBUF (i) L
Processing such as lowering sm to the logical rOJ level is executed.

Thus, in the twin mallet performance mode, the first repetitive waveform WAVEI is used to repeatedly and intermittently form an envelope for the lowest tone group, and at the same time, the second repetitive waveform WAVE1, which is shifted by 172 cycles from the repetitive waveform WAVE1 of the tube 1, is generated. Envelopes of musical tone groups other than the lowest tone are repeatedly and intermittently formed using the repetitive waveform WAVE2.

(4) Rapid decay subroutine In the timer interrupt processing routine RTIO (Fig. 1O), when the rapid decay subroutine (FDEXEiFDEXE2) of step 5P26 is entered at the timing of TIM=0, the CPU 5 executes the step 5P3
In step 5, it is determined whether the performance mode data MODE is 1 or not.

If a positive result is obtained here, this means that the performer has specified the repeat performance mode, and in this case, the CPU 5 moves to step 5P36 and sets the key-on/key-off data of the sound generation channel operation status register REG2.
TBUF (i)□.

5TBUF out of all pronunciation channels i
(i) Search for pronunciation channel i of □m=rlJ,
Key-off disable data 5TBUF held in the relevant sound generation channel i (i) Logic “1” in Lsm
After writing the data, the musical tone control information MUS is transferred to the musical tone signal generating section (TG) 8 to perform a process of rapidly attenuating the corresponding sound generation channel, and then the timer interrupt processing routine RTIO is executed from step 5P37. (
Return to Figure 10).

(Then, when the performer specifies the repeat performance mode, the CPU 5 executes envelope control such that the musical tones of all sound generation channels are rapidly attenuated at the timing when the time information data TIM is TIM=O.

On the other hand, if a negative result is obtained in step 5P35, this means that the player has designated the twin mallet performance mode.

At this time, the CPU 5 moves to step SP3B and determines that the lowest tone key-on data L S T K YHs* held in the lowest tone key code register REG4 is logic "1".
It is determined whether or not the repeating waveform for the lowest note is currently being sounded (FIG. 8(E)), and when an affirmative result is obtained, the process moves to step SP39.

In this step 5P39, the CPU 5 first determines that the key-on/key-off data 5TBOF (i) □ held in the sound generation channel operation status register REG2 is logical rlJ, and that the sound generation channel code register R
Pronunciation key code data held in EGI K
EYBUF (i) L12-61 is the lowest note key code data L S of the lowest note key code register REG4
TK YL51-68 not equal to pronunciation channel i
Determine whether or not there is.

If a positive result is obtained here, this means that in the sound generation channel where the key is operated by the performer (i.e., in the channel of 5TBUF (i) sss = rl), the key code other than the lowest note key code is In other words, it indicates that both the lowest musical tone and the other musical tones are specified.

At this time, the CPU 5 moves to step 5P40, and sets the key-off disable data 5 of the sound generation channel held in the sound generation channel operation status register REG2.
TBUF (i) After writing logic "1" data to Ill to inhibit the sound key-off process (Fig. 8 (G)),
By transmitting the musical tone control information MLJS to the musical tone signal generator (TG) 8, the tone generation channels of the musical tone group other than the lowest note are rapidly attenuated, and then from step 5P41 to the timer interrupt processing routine RTIO (FIG. 10). return.

Thus, in the twin mallet performance mode, the CPU 5 generates the repetitive waveform WAV of the musical tone having the lowest key code.
When the EI is in the generating state, the envelopes of musical tones other than the lowest note are rapidly attenuated.

On the other hand, if a negative result is obtained in step 5P38, this means that the key code of the currently sounded musical note is not the lowest note, and at this time, the CP
U5 moves to step 5P42 and reads the key-on/key-off data 5TBUF (i) ssm = r IJ held in the sound generation channel operation status register REG2.
, and the pronunciation key code data held in the pronunciation channel key code register REGI KEYB
UF (i) ~6. is the lowest key code register R
Lowest key code data LST held in EG4
Search for channel i equal to KYtsi~, and set key-off disable data of sound channel operation status register REG2 to 5TBUF (i)L! After writing logic "1" data to the musical tone signal generator (TG)
At step 8, the envelope of the musical tone of the i-channel is subjected to rapid attenuation processing, and then the process returns to the timer interrupt processing routine RTIO (FIG. 10) from step 5P43.

In addition to this, when a negative result is obtained in step 5P39 described above (this indicates that there is no key code other than the lowest note key code among the keys that are being operated), the CPU 5 similarly performs Step 5 above
After executing the process of P42, the process returns to the timer interrupt process routine RTIO (FIG. 1θ) from step 5P43.

(5) In the repeated waveform key-on subroutine timer interrupt processing routine RTIO, the time information register REG6
When the time information data TIM becomes TIM=2 and enters the repetitive waveform key-on subroutine (RPKONI, RPKON2) in step 5P27 (Fig. 10), C
As shown in FIG. 12, PU5 first performs step 5P4.
In step 5, it is determined whether or not the performance mode data MODE is 1, and when a positive result is obtained (that is, in the repeat performance mode), the process moves to step 5P46, where the tone generation channel operation status register REG2 holds the data. Key-on/key-off data 5TBUF (i). m
=rlJ (that is, the key code of the key that is being operated as a key is assigned) for sound generation channel i, sound generation on/off data KEYBUF of sound generation channel key code register REGI (i) logic r in ssm
After writing the lJ data, the musical tone signal generator (TG) 8
key code KE in the corresponding pronunciation channel i.
After performing key-on processing on the musical tone having the key code YBUF (i) Lsm~, the program returns to the timer interrupt processing routine RTIO (FIG. 10) from step 5P47.

Thus, in the repeat performance mode, the CPU 5
Musical tones corresponding to all keys being turned on by a performer are generated at timing M=2.

On the other hand, if a negative result is obtained in the above-mentioned step 5P45 (that is, in the twin mallet performance mode), the CPU 5 moves to step 5P48 and inputs the lowest key-on data LSTKY, which is held in the lowest key code register REG4. , is the logic "1" or not.

If a positive result is obtained here (that is, if the repetitive waveform WAVEI is currently being sounded for the musical tone of the lowest note key code), the CPU 5 executes step 5P49.
Step 5P50 After switching to the state in which the lowest note is not generated by writing data of logic rOJ to the lowest note key-on data LSTKY□ of the lowest note key code data register REG4, step 5P50
Move to.

In this step 5P50, the CPU 5 reads the key-on/key-off data 5TBUF (i) held in the sound generation channel operation status register REG2. m=rlJ and the sounding key code data KEYBUF of the sounding channel key code register REGI (i) LSI-bl
It is determined whether there is a sound generation channel for which the lowest tone key code data L5TKY Ls, .about.ill of the lowest tone key code register REG4 does not match.

If a positive result is obtained here, the CPU 5 proceeds to step 5P.
After moving to step 51 and writing logic "1" data into the sound generation on/off data KEYBUF (i) and 4sm of the sound generation channel key code register REGI for the sound generation channel i, the musical tone signal generation section (TO) 8 writes the sound generation on/off data KEYBUF (i) of the sound generation channel key code register REGI to 4sm. Pronunciation key code data for i KE
YBUF(i) The musical tone of the LSI-41 is subjected to key-on processing, and then the process returns to the timer interrupt processing routine RT10 (FIG. 10) from step SP52.

Thus, the CPU 5 determines that the time information data TrM is TIM=
At the timing of 2, when the musical tone of the lowest tone is currently being generated (the state in which the decay waveform part W14 of FIG. 8 (J) is being generated), the musical tone of the lowest tone is repeated next. After controlling to a state where no generation processing is performed during the period (TIM=0 to 25 (=0)), by generating musical tones other than the lowest note, the repetitive waveform WAVE2 is generated.
The generation of the attack waveform portion W22 (FIG. 8(K)) is started.

On the other hand, when a negative result is obtained in the above-mentioned step 5P48 (that is, when the currently generated musical tone is not the lowest note), the CPU 5 moves to step 5P53 and inputs logic to the lowest note key-on data LST K Yws*. After controlling the conditions to generate the lowest tone by writing data "1", the process moves to step SP54.

In this step 5P54, cpus generates 5TB of key-on/key-off data for all sound generation channels.
UF Mouth) □6 is logic “1” and pronunciation key code data KEYBUF (i) L! After searching for a sounding channel i whose m~, matches the lowest note key code data L 5TKYLss~, and writing logic "1" data into the sounding on/off data KEYBUF (i) xs* of the sounding channel i, , musical tone signal generator (TG
) 8 is key-on-processed for the musical tones of the sound generation key code data KEYBLJF (i), .

In this manner, the CPU 5 executes processing to generate the lowest tone at the timing of TIM=2 in a state where the lowest tone is not generated.

In addition, when a negative result is obtained in step 5P50 described above, this indicates that the conditions for generating musical tones other than the lowest note are not set, and in this case, the CPU 5 returns to step 5P53 described above. After moving on and rewriting the logical rlJ data as the lowest note key-on data LSTKYMSI, in step 5P55, processing is performed to generate the musical tone of the lowest note.

As a result, if the performer has specified the twin mallet performance mode but has not pressed a key with a key code other than the lowest note, the next repetition period T
Even when IM=0 to 25 (=O), the lowest tone repetitive waveform WAVEI is generated.

(6) In the disenable end subroutine timer interrupt processing routine RTIO (Figure 10), TIM=1
Disable end subroutine (DISOFFI, DISOFF2) of step 5P28 at timing 5
As shown in FIG. 13, the CPU 5 first inputs the key-off disable data 5TBUF held in the sound generation channel operation status register REG2 for all sound generation channels in step 5P61.
(i) By rewriting the LSI to logic "0" data, the musical tone signal generating section (TG
) 8, control is made to enable key-off processing.

Next, the CPU 5 moves to step 5P62, where the key-off request data 5TBUF (i) 6□ held in the sound generation channel operation status register REG2 is set to logic rl.
It is determined whether or not it is the pronunciation channel of J. If a negative result is obtained here, this means that the sound generation channel for which key-off processing is requested is not pending, and in this case, the CPU 5 executes the timer interrupt processing routine RTIO (FIG. 10) from step 5P63. Return to

On the other hand, if a positive result is obtained in step SP62, this means that the channel for which key-off processing is requested in any of the sound generation channels i is on hold, and at this time, the CPU 5 proceeds to step 5P64. Move.

In this step 5P64, for the sound generation channel i that has the current key-off request, the key-off request is erased by writing logical "0" data to key-off request data 5TBUF (i) 6m, and key-on/key-off data 5TBUF (i) □ By writing logic "0" data to , the performer's key-off operation is captured as data, and by writing logic "0" data to sound generation on/off data KEYBUF N) Mss, the condition for keying off the corresponding sound channel is set. After the settings are made, key-off processing is executed for the relevant tone generation channel of the musical tone signal generator (TG) 8.

Here, key-off request data 5TBUF (j)b
m is key-off disable data 5TBUF (
i) When a key-off operation is performed while Lss is in the logic "1" state, it becomes logic rlJ (Figure 15)
, the pronunciation key-off processing is not performed immediately and is put on hold.

In this way, the CPU 5 executes a process of naturally attenuating the musical tone for the sound generation channel i for which the key-off request is made.

Incidentally, in the key-off process at step SP64, the musical tone signal generating section (TG) 8 generates the following signals as shown by the decay waveform sections W15 and W25 in FIGS. 8(J) and (K).
Rapid decay waveform portions Wll and W21. ! : Performs natural attenuation key-off processing that attenuates musical tones so as to draw an attenuation waveform with a smaller attenuation rate compared to It is possible to naturally attenuate musical tones without causing noise in them.

In step 5P65, the &CPU 5 generates pronunciation key code data KEYBUF (i) LS that is equal to the lowest note key code data L S T K Ytsm~ of the pronunciation channel i that has undergone the key-off processing.
It is determined whether there is a sound generation channel i of I-61.

If a negative result is obtained here, this means that the musical tone subjected to the key-off process is not the lowest note, and in this case, the CPU 5 returns from step 5P66 to the timer interrupt processing routine RTIO (FIG. 1O).

On the other hand, if an affirmative result is obtained in step SP65, this means that the musical tone subjected to key-off processing is the lowest note, and at this time, the CPU 5 moves to step 5P67 and performs all sound generation channels i. Key-on/key-off data 5TBUF (i). , is a logical rOJ.

If a positive result is obtained here, this means that all the sound generation channels are in the key-off operation state, and at this time, the CPU 5 moves to step 5P68 and inputs the logic into the lowest note key code data LSTKYt, ss~. "0
” data, the timer interrupt processing routine RTIO (10th
Return to figure).

On the other hand, if a negative result is obtained in step 5P67 described above, this means that there are remaining sound generation channels in the key-on operation state, and at this time, the CPU 5 moves to step 5P70 to check all sound generation channels. About i Key on/key off data 5TBUF
(i) Pronunciation channel i where , ss is logic “1”
At the same time, the sound generation key code data KEYBUF (i) LS1 to 4. of the sound generation channel i is searched. Search for the pronunciation channel i that has the minimum value.

In this way, the CPU 5 finds the sound generation channel i having the lowest note key code among the sound generation channels for which the key-on operation has been performed, and then moves to step 5P71 to obtain the sound generation key code data KE for the sound generation channel with the lowest note.
Y BUF (i) L□~6° as the lowest key code data L S T K Y Lsm~6. is stored in the lowest note key code register REG4, and then from step 5P72 the timer interrupt processing routine RTIO (first
Return to Figure 0).

Thus, when the musical tone subjected to key-off processing in step 5P64 is the lowest note, the CPU 5 designates the musical tone with the smallest key code from among the remaining musical tones as the lowest note, and stores this in the lowest note key code register REG4. .

(7) In the repetitive waveform key-off subroutine timer interrupt processing routine RTIO (Fig. 10), the CPU 5
When the repetitive waveform key-off subroutine (RPKOFF1, RPKOFF2) is entered at the timing of TIM=20,
As shown in FIG. 14, in step 5P75, the sound generation on/off data KEY of all sound generation channels is
BUF (i) For the tone generation channel i whose nss is logic rlJ, the musical tone signal generator (TG) 8 is subjected to key-off processing, and the tone generation on/off data KEYBUF(t), 4ss of the tone generation channel that has undergone the key-off processing is set to logic "0". '' After writing the data, the process returns to the timer interrupt processing routine RT10 (FIG. 10) from step 5P76.

Thus, at the timing of TIM=20, the CPU 5 converts the musical tone into the decay waveform section W14 for the tone generation channel of the musical tone generated by the musical tone signal generator (TG) 8.
, W24 (FIG. 8 (J), (K)).

(8) Key-off event processing The CPU 5 executes the processing shown in FIG. 15 in the key-off event processing routine RT5 of the main routine RTO (FIG. 4).

That is, in step SP81, the CPU stores the key code of the key where the key-off event occurred in the event key code register REG3 as event key code data N.
After writing as K YLsm +&l, step 5
At P82, it is determined whether the performance mode data MODE is 0 or not.

When a positive result is obtained here (in other words, in the normal performance mode), the CPU 5 moves to step 5P83 and sets the sound generation on/off data KEYBUF for all sound generation channels i (i) Nim is logic "1" and the sound generation key Code data KEYBUF(i)Lsm~6. is event key code data NKYLsm~6. It is determined whether there is a pronunciation channel i equal to .

If a negative result is obtained here, this means that among the musical tones currently being produced, there is no musical tone that has the key code for which a key-off event has occurred, and therefore, in effect, the CPU 5 Since there is no need to process the event, the CPU 5 returns to the main routine RTO from step 5P84.

On the other hand, if a positive result is obtained in step 5P83, this means that the CPU 5 needs to perform key-off processing for the key code of the key in which the key-off event has occurred, and in this case, the CPU 5 returns to step 5P85. Then, the channel number data of the sounding channel i where the key-off event occurred is held in the key-off data register REGIO (FIG. 2) of the register section 7 as key-off data OFF.

Next, the CPU 5 moves to step 5P86, and checks the sound generation on/off data KEYBUF (i) held in the sound generation channel key code register REGI. F) xs
After rewriting m to logical “O” data, step 5P8
7, the musical tone signal generating section (TG) 8 executes key-off processing for the sound generation channel in which the key-off event has occurred, and then returns to the main routine RTO from step 5P88.

Here, the musical tone signal generating section (TG) 8 performs the key-off processing in step 5P87 to reach the point t KOF in FIG.
When a key-off operation is performed at F, the musical tone is naturally attenuated by the decay waveform portion WO3 having a natural attenuation slope.

On the other hand, if a negative result is obtained in the above-mentioned step SP82, this means that the performer has specified the repeat performance mode or the twin mallet performance mode, and at this time, the CPU 5 in step 5P89 For i, step 5P above
When the same judgment as in step 83 is made and a negative result is obtained, there is no need to perform key-off processing for the relevant key-off event, so the main routine RT starts from step 5P90.
Return to O.

On the other hand, when an affirmative result is obtained in step 5P89, the CPU 5 moves to step 5P91, stores the channel number data of the sound generation channel i in which the key-off event has occurred in the key-off data register REGIO as key-off data OFF, and then proceeds to step 5P91. 5P92
Move to.

In this step 5P92, the CPU 5 selects key-off disable data 5TBUF (OFF)LS of the sound generation channel in which the key-off event has occurred, among the key-off disable data 5TBUF(i)tsa stored in the sound generation channel operation status register REG2.
It is determined whether I is logical "1" or not.

If a positive result is obtained here, this means that the key-off process is currently prohibited for the corresponding sound generation channel 1 = OFF (Fig. 8 (F) and (
G)) At this time, the CPU 5 moves to step 5P93 and stores the key-off request data 5 of the corresponding sound generation channel 1=OFF in the sound generation channel operation status register REG2.
After writing logic “1” data as TBUF (OFF>61), the main routine RTO starts from step 5P94.
Return to

Thus, the CPU 5 suspends the key-off processing of musical tones during the period in which the key-off processing of musical tones is prohibited when the sound generation channel 1=OFF in which the key-off event occurs.

This pending state is based on key-off request data of 5TBUF.
(OFF) The process continues until a negative result is obtained in step 5P92 by rewriting Lsg to logic "0" data.

On the other hand, if a negative result is obtained in step 5P92, this means that the period in which the prohibition of key-off processing of musical tones has been lifted for tone generation channel 1 = OFF where the key-off event has occurred, and at this time C.P.
U5 moves to step 5P95 to obtain key-on/key-off data 5TBUF (OFF) of sounding channel 1=OFF where a key-off event has occurred. By writing logical "0" data into the register REG2, information indicating that the performance operation state has become the key-off operation state is written in the sound generation channel operation state register REG2.

At this time, the CPU 5 moves to step 5P96 and sets the sound generation on/off data KEY regarding sound generation channel 1=OFF.
BUF (OFF) Determines whether Hsm is logic "1" or not. If a positive result is obtained here (meaning that musical tones have been generated up to now), the CPU
Step 5 moves to step 5P97, where the musical tone signal generator (TG)
8 to execute the key-off processing of the musical tone for the sound generation channel 1 = OFF, and then in step 5P98, turn on/off the sound generation for the sound generation channel 1 = OFF that has undergone the key-off processing.
Set off data KEYBUF (OFF) + sa to logic “0”
” data (meaning that the information obtained by keying off the musical tone is retained).

On the other hand, if a negative result is obtained in the above-mentioned step SP96, this means that the key that was turned off is not already in the state of producing a musical tone, and at this time, the CPU 5 performs step SP97 and 5P98
The process jumps to step 5P99.

Here, the CPU 5 judges whether the event key code data NKYLSI~6m of the key generated by the key-off event is equal to the lowest note key code data L S T K Ytss~68, and when a negative result is obtained (i.e. When a key with a key code other than the lowest tone is turned off), the CPU 5 returns to the main routine RTO from step 5P100.

On the other hand, when an affirmative result is obtained in step 5P99 (that is, when the lowest note key is operated as a key-off), the CPU 5 moves to step SP1.01 and generates key-on/key-off data 5TBUF ( i) 0. It is determined whether or not the logic is "1".

If a negative result is obtained here, this means that there is no longer a key in the key-on operation state in the sound generation channel operation state register REG2.
After clearing the lowest key code information by writing all O data in S T K YLss and bm, the process returns to the main routine RTO from step 5P103.

On the other hand, if a positive result is obtained in step 5PIOI, the CPU 5 moves to step 5PIO4, and calculates key-on/key-off data 5TBUF (i), 4sm for all sounding channels i among sounding channels i whose logic is "1". Pronunciation key code data KEYBU
F (i) Search for the pronunciation channel i with the minimum Lsg-ha to search for the pronunciation channel i having the lowest note key code, followed by step 5P105.
In, the pronunciation key code data KEYBUF for the pronunciation channel i of the lowest note (i) Ls proverb ~ 4. New lowest key code data L S T K Yll
l Lowest key code register REG4 as m&j+
After this, the process returns to the main routine RTO from step 5P106.

(9) Tone change processing When cpus enters the tone change processing routine RT7 of the main routine RTO, it generates a musical tone of the corresponding tone according to the processing procedure shown in FIG. 16, and performs music based on the tone of the generated musical tone. A process is executed to select one of the following modes: normal performance mode, repeat performance mode, and twin mallet performance mode.

That is, in step 5PIIO, the CPU 5 writes the timbre information data TONE representing the timbre newly specified by the performer to the timbre information register REG7 of the register section 7, and then in step 5Pill, the CPU 5 writes the timbre parameters of the timbre information data to the musical tone signal generation section. (TG) Send to 8.

In this way, the musical tone signal generating section (TG) 8 is controlled to generate a musical tone of a tone specified by the player.

Subsequently, the CPU 5 moves to step 5P112 and inputs the performance mode data M held in the performance mode register REC5.
After rewriting ODE to the old performance mode register REG8 as the old performance mode data 0M0DE, step 5P1
At step 13, the tone color information data TONE is converted into a performance mode conversion data device (TONE) and stored in the performance mode register REG5 as performance mode data MODE.

Here, the performance mode conversion data TBL (TONE) is the performance mode conversion table and the tone parameter memory section 10.
(FIG. 1) is stored in advance as a conversion table.

The CPU 5 continues (in step 5P114, the old performance mode data 0M0DE is the numerical value 0 (this means that the performance mode before the tone change is the normal performance mode), and the performance mode data MODE is not 0 ( This means that the currently designated performance mode is the repeat performance mode or the twin mallet performance mode.

If a positive result is obtained here, it means that the performance mode changed according to the tone specified by the performer was from normal performance mode to repeat performance mode or twin mallet performance mode, and at this time, the CPU 5
Moving on to step 5, the sound on/off data KEYBUF (i) □, held in the sound channel key code register REGI for all sound generation channels, is transferred to the sound channel operation status register REG2 as key on/key off data 5TBUF (i), 4ss. At the same time, the sound generation on/off data of the sound generation channel musical tone processing register REG1 is transferred.KEYBUF (i)9. Clear l to all 0.

In this way, the cpus obtains a state in which the state of the sound generation channel that was producing sound before the tone color conversion is transferred to the sound generation channel operation state register REG2 as operation information.

Subsequently, in step 5P116, the CPU 5 attenuates the musical tone by causing the musical tone signal generator (TG) 8 to perform key-off processing on the musical tone for all of its sound generation channels, and then in step 5P117, it outputs key-on/key-off data for all sound generation channels. 5TBUF
(i) Determine whether □1 is the sound generation channel of logic rlJ.

If a negative result is obtained here, this means that there is no sound generation channel to be processed, and therefore the lowest note cannot be specified, so step 5P11
Moving on to 8, the lowest note key code data L S T K Y
After writing all 0 data to Ls-s~, 1, the process returns to the main routine RTO from step 5P119.

On the other hand, if a positive result is obtained in step 5P117, this means that the sound generation channel 1 (=O~15)
This indicates that the key that was turned on is assigned to one of the keys, and at this time the CPU 5 executes step 5.
Moving on to P120, for all pronunciation channels i,
Among the sound generation channels i whose key-on/key-off data 5TBUF (i) □ is logic "1", the sound generation channel whose sound generation key code data KEYBUF (i) Lsm~ is the minimum key code, that is, the sound generation channel of the lowest note. Search for i.

Next, the CPU 5 moves to step 5P121, and generates the pronunciation key code data K for the pronunciation channel i of the lowest note.
EYBUF (i) After writing LSI~h to the lowest note key code register REG4 as the lowest note key code data L S T K Ytsm~, 1, step 5P.
At step 122, the process returns to the main routine RTO.

In this manner, the CPU 5 selects the key code data L S T K Y Ls*~1. is selected and held in the lowest key code register REG4,
This executes the process of entering repeat performance mode or twin mallet performance mode.

On the other hand, when a negative result is obtained in the above step 5P114, the CPU 5 moves to step 5P123 and determines whether the old performance mode data 0M0DE is not 0 and the performance mode data MODE is O. .

If a positive result is obtained here, this means that the performance mode before the tone change has been switched from repeat performance mode or twin mallet performance mode to normal performance mode, and at this time, the CPU 5 Moving on to 5P124, key-on/key-off data 5TBUF (i) of all sounding channels is sounding on/off data KEYBUF (i) for sounding channel i where ssm is logic rlJ. After writing logic "1" data to 3, the musical tone signal generator (TG) 8 generates the sound generation key code data KEYBUF in the corresponding sound generation channel i.
(i) After performing key-on processing on musical tones with key codes of Lsm~ and lI, step 5P126
Return to the main routine.

Thus, the CPU 5 registers the sound generation channel operation status register R.
Based on the key press operation information stored in EG2, musical tones are generated in normal performance mode using the sound generation key code data held in the sound generation channel key code register REG1.

On the other hand, if a negative result is obtained in the above-mentioned step 5P123, this means that the performance mode has not been switched, and in this case, the CPU 5 immediately returns to the main routine RTO from step 5P125.

(10) Operation The electronic musical instrument having the above configuration operates as follows.

(10a) Key-on event processing operation When the performer performs a tone specifying operation on the panel operation section 3, the CPU determines this in step RT6 of the main routine RTO (Figure 4) and performs a tone change processing routine RT7 (Figure 16). Enter step 5P113
Performance mode data MO·DE=0.1.2 representing the normal performance mode, repeat performance mode, and twin mallet performance mode is stored in the performance mode register REGS&, so as to correspond to the timbre specified in .

Here, when the normal performance mode is specified, the CPU5
is step 5P114-3'P123-3P124-
In the loop of 3P126, when a key on the keyboard section 2 is being played, a musical tone corresponding to the key is generated in the musical tone signal generating section (TG) 8, and in response to this, a key is not yet pressed on the keyboard section 2. If not, the process returns to the main routine from step 5P125 without generating musical tones.

On the other hand, when the repeat performance mode or the twin mallet performance mode is specified, the CPU 5 performs step 5PI.
L4-3P115-3P116-SP117-SP12
In the loop of 0-3P121-5PI22, after resetting the musical tone signal generator (TG) 8 as a whole to the key-off state, the musical tone of the minimum key code among the keys that are operated on is generated as the lowest note key code data LST
KYL, ~4. (or if no key has been pressed, the lowest key code data LSTK Y
Set L□ to 611 to all rOJ), repeat waveform WAVEI (in repeat performance mode), or WAV by timer interrupt processing routine RTIO (Fig. 10).
Prepare for the formation process of El and WAVE2 (during twin mallet performance mode).

In this state, when the performer specifies the normal performance mode and presses the key on the keyboard section 2 (see Fig. 17 (A)
)), the CPU 5 executes this step RT2 (Fig. 4)
The judgment is made in key-on event processing routine RT3.
(Figure 5), step 5PI-3P2-3P3
- After assigning the key information of the key pressed by the loop of SF3-3P5-3P6 to the sound generation channel i = ASS, a musical tone is generated in the musical tone signal generating section (TG) 8, and thus the musical tone signal generating section (TG) 8 is shown in Figure 17 (
As shown in C), an attack waveform portion WOI and a sustain waveform portion WO2 are sequentially formed at the key press operation time point tKON.

In this normal performance mode, when the interrupt signal INT is generated from the interrupt timer 10, the CPU 5 executes step 5P21 of the timer interrupt processing routine RTIO (FIG. 10).
-3P22 loop immediately returns to the main routine. Therefore, the musical tone signal generator (TC;) 8 performs envelope control to maintain the sustain waveform section WO2 for the musical tone being generated, so that the tone signal generator (TC;) continues to generate musical tones corresponding to.

In contrast, when the performer specifies the repeat performance mode or the twin mallet performance mode and presses the keys on the keyboard section 2 (Fig. 18 (A1) to (A4), or
Figure (A)'), the CPU 5 executes this in step RT2 (
4), enters the key-on event processing routine RT3 (FIG. 5), and enters step 5PI-3P2.
-3P? -3P8-3P9-3PIO loop generates key information of the pressed key Channel j=Ass
As well as assigning the lowest note key code data L S T
KYLIm～6. After selecting, step 5PII-3
The interrupt timer 11 and the time information data TIM are synchronized by a loop of P13-3P14-3P15-3P16.

Here, in step 5P15, the CPU 5 executes step 5 of the timer interrupt processing routine RTIO (FIG. 10).
P21-3P23-3P24-3P26-3P30-3
Through the loop of P31, the repetitive waveform WAVEI (Fig. 18 (C1) to (C4)), or WAVEI, WAV
Rapid decay waveform part W of E2 (Fig. 19 (E), (F))
11, or Wl 1, W21, attack waveform part W12,
or Wl2, W22, sustain waveform section W13, or W
l3, W23, decay waveform part W14, or Wl, l,
We will form W24.

Thereafter, the CPU 5 receives the interrupt signal INT from the interrupt timer 11.
By executing the timer interrupt processing routine RTIO (Fig. 10) each time the waveform WAV is obtained,
EI, WAVEI, and WAVE2 are formed, and the waveforms WAVEI, WAVEI, and WAV are repeated until the player performs a key-off operation.
Form E2 repeatedly.

When the player eventually performs a key-off operation, the CPU 5 executes a key-off process corresponding to the designated performance mode.

(10b) Key-off processing in normal performance mode When the performer releases a key at time tll0FF in FIG.
(Fig. 4), the key-off event processing routine RT5 (Fig. 15) is detected, and the key-off event processing routine RT5 (Fig. 15) is detected, and steps 5P81-3P82-SP83-3P8
5-3P86-3P87-SPB8 loop attenuates the musical tone for the key where the key-off event occurs in the musical tone signal generator (TG) 8 (Fig. 17(C))
).

Here, the attenuation waveform of the musical tone in normal performance mode is
As mentioned above, the decay waveform section WO3 in FIG. 7 is selected as a natural decay waveform with a small slope, so that the musical tone subjected to key-off processing decays naturally without causing any unnatural changes such as noise. .

(10c) Key-off processing in repeat performance mode First,
As shown in FIG. 18 (A1), the rapid decay waveform portion W11
(Figure 18 (CI)) period [. Time t between ~t8
Yo. When one of the plurality of keys being pressed in the FFI is released, the CPU 5 executes the key-off event processing routine R of the main routine RTO (FIG. 4).
Steps 5P81-3P82-3 of T5 (Figure 15)
Through the loop of P89-3P91-3P92-3P93-3P94, key-off request data 5TBUF(OFF)i is generated for the key where the key-off event has occurred.
A state is obtained in which n is held in the sound generation channel operation state register REG2, thus key-off time t,. In the FFI, the musical tone signal generator (TG) 8 is controlled so as not to perform a key-off operation, and this state is maintained until time tlS.

Incidentally, at the timing from time t0 to time t, key-off disable data 5TBUF (OFF)tsm
has risen to logic “1” (Figure 18 (B)
), the CPU 5 obtains an affirmative result in step 5P92 (FIG. 15) so that it does not immediately perform the sound generation key-off process, and also sets the key-off request data 5TBUF (OFF) ha in step SP93.
The logical "l" data is held as .

Then time t1. Then, the CPU 5 executes step 5P6 of the disable end subroutine 5P28 (FIG. 13) of the timer interrupt processing routine RTIO (FIG. 10).
1-3P62-3P64, the repetitive waveform WAVEI (FIG. 18 (
C1)) is naturally attenuated along the naturally attenuated waveform portion W15. (Thus, the CPU 5 can naturally attenuate the musical tone so as not to cause unnatural changes in the musical tone.

Incidentally, if the generation of musical tones is interrupted at the key-off time tKOFF+, noise will inevitably be generated in the musical tones because the musical tones will rapidly attenuate with the rapid decay waveform Wll; Following the rapid decay waveform Wll corresponding to the timing at which the event occurred, there is an attack waveform portion W12 and a sustain waveform portion W for one more wave.
13 and the natural attenuation waveform portion W15, the musical tone can be attenuated naturally.

Second, as shown in FIG. 18 (A2), when a key-off operation is performed at time t KQFF2 between time t0 and t+s when the attack waveform portion W12 and sustain waveform portion W13 are formed, the CPU 5 The key-off event processing routine RT5 is performed in the same way as in the case of FIG.
(Figure 15) Steps 5P81-3P82-3P89
- Key-off request data 5TBUF (OFF
) 611, the key-off request data STBUF (OFF) is set at a subsequent time tls by a loop of steps 5P61-3P62-3P64 of the disable end subroutine 5P2B (Fig. 13).
The sound generation channel in which am is held is subjected to sound key-off processing. As a result, the key-off time t war is reached as shown in FIG. 18 (C2).

After continuing to control the musical tone signal generating section (TG) 8 to generate sound, the musical tone can be naturally attenuated by the natural attenuation waveform section W15.

Thus, even in this case, the CPU 5 can attenuate the generated musical tone without causing any unnatural changes.

Incidentally, if the musical tone is immediately attenuated in the middle of the attack waveform portion W12 at the key-off time t KOFF2, an incomplete attack waveform may remain, which may cause a change in the musical tone that gives an unnatural impression that the sound is being pulled back. However, according to the embodiment described above, this can be prevented.

Thirdly, as shown in FIG. 18 (A3), the time t between the time t's and t2° corresponding to the sustain waveform portion W13.
When a key-off event occurs in KOFF+,
The CPU 5 stores 5 TBU of key-off disable data.
Since F (OFF) Lm has already fallen to the logic "0" level, steps SP81-3P82 of the key-off event processing routine RT5 (FIG. 15) are executed.
-3P89-3P91-3P92-3P95-3P96
-3P97-3P98 loop causes the key-off process to be immediately executed for the sound generation channel 1=OFF in which the key-off event has occurred among the sound generation channels i of the musical tone signal generator (TG) 8.

As a result, the CPU 5, as shown in FIG. 18 (C3),
At key-off time t. Naturally damped waveform part W immediately from FFI
16 performs envelope control for natural attenuation.

Fourth, as shown in FIG. 18 (A4), the time tza corresponding to the decay waveform portion W14 and the time t0 of the next cycle.
When a key-off event occurs at time t KOFF4, the CPU 5 has already executed the key-off event at time t2 by steps 5P75-3P76 of the repetitive waveform key-off subroutine 5P29 (Fig. 14) of the timer interrupt processing routine RT10 (Fig. 10). Musical tone signal generator (
TG) Performs key-off processing on 8 and generates sound on/off data KEYBUF (i). , is controlled to be cleared to the logical "0" level, so step 5P81- of the key-off event processing routine RT5 (Fig. 15)
The loop of 3P82-3P89-3P90 performs envelope control so that the tone signal generator (TG) 8 continues to naturally attenuate.

As a result, the repetitive waveform WAVEI is a decay waveform portion W14 generated from time t2°, as shown in FIG. 18 (C4).
As a result, the CPU 5 can attenuate the musical tone without causing an unnatural change in the musical tone.

(10d) Key-off processing in twin mallet performance mode In the twin mallet performance mode, the CPU 5 executes steps 5P45-3P48-3P49-3P50-3P5 of the repetitive waveform key-on subroutine 5P27 (Fig. 12).
1-3P52 loop and steps 5P45-3P48
- By switching between the loops of 3P53, 3P54, and 3P55, a group of musical tones opposite to the group of musical tones generated immediately before, that is, the lowest note or other musical tones, is alternately generated, whereas the lowest note or other musical tones are alternately generated. When a key-off event occurs in one of the other musical tone groups, as shown in FIG. 19 in the same manner as in the repeat performance mode described above with reference to FIG.
FFI % t KOFFI , t I[0FF3,
t Decay waveform W1 that naturally attenuates according to KOFF4
5. Attenuate the musical tone along W16 and W17.

At this time, the key-off prohibition period of the repetitive waveform WAVEI or WAVE2 is set using key-off disable data 5TB.
UF (OFF) ILSm or 5TBUF (OFF)
2Lsm (FIG. 19(C) or (D)).

This prevents unnatural changes in musical tones when key-off processing is performed.

Here, in the twin mallet performance mode, when a key-off event occurs on the lowest note key, the CPU 5 performs the steps 5P96-5P99-3P10.1-3P104-3 of the key-off event processing routine RT5 (FIG. 15).
Through the loop of P105-3P106, by reselecting the musical tone that is the lowest tone among the musical tones of the other musical tone groups, the lowest tone repeating waveform WAVEI (Fig. 19 (E)) is created for the reselected musical tone. Perform envelope processing.

On the other hand, if a key-off event occurs for one of the other musical tones, the rapid decay subroutine 5P26
By executing the rapid attenuation process described above with respect to FIG. 11, only the musical tone in which the relevant key-off event has occurred is attenuated.

According to the embodiment described above, even in the twin mallet performance mode, it is possible to naturally attenuate a musical tone in which a key-off event has occurred without causing any unnatural changes in the musical tone.

[2] Other embodiments (1) In the above embodiment, the repetitive waveform WAVEl
, WAVE2, key-off disable data 5TBUF (OFF) tsm is used from the start of the rapid decay waveform portions Wll and W21 to the sustain waveform portions W13 and W23 via the attack waveform portions W12 and W22.
(Figure 18), 5TBUF (OFF)ltsa
An example has been described in which key-off processing of musical tones is not performed during the period of logic "1" by setting S5TBUF (OFF) 2L-1 (FIG. 19) to logic rlJ. The key-off prohibition period is not limited to this, and the same effect as described above can be obtained even if it is limited to, for example, the period of the rapid decay waveform portions Wll and W21, the period of the attack waveform portions W12 and W22, etc.

(2) The method of allocating sound generation channels when a key-on event occurs is not limited to the method described above, and other methods may be applied.

(3) In the above embodiment, musical tones are divided into the lowest note and other musical tone groups in the twin mallet performance mode, but the tone groups are not limited to this, for example, the highest note and other musical tone groups. Various changes may be made, such as dividing into musical tone groups.

(4) In the above embodiment, the period in which the sound generation on/off data KEYBUF (i) and 4sa are suspended is from the rapid decay waveform portion Wit, W21 to the attack waveform portion W.
12, W22, and the period TIM=0 to TIM=15 spanning the sustain waveform portions W13, W23 has been described, but the present invention is not limited to this, and does not have the rapid decay waveform portions Wll, W21. Repeated waveform WAV like
In the case of EI and WAVE2, the same effect as in the above case can be obtained even if the period of the attack waveform portion W12 is selected.

〔Effect of the invention〕

As described above, according to the present invention, when envelope control is performed to generate a repetitive waveform, if an unnatural change occurs in the musical tone when the musical tone is immediately interrupted at the timing when a key-off event occurs, By forming a natural attenuation waveform, the musical tone can always be attenuated naturally no matter what timing of the repeated waveform a key-off event occurs.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of an electronic musical instrument according to the present invention, FIG. 2 is a route map showing the detailed configuration of the register section, FIG. 3 is a diagram showing the relationship between keys and key codes, and FIG.
The figure is a flowchart showing the main routine of the CPU in Fig. 1, Fig. 5 is a flowchart showing the key-on event processing routine in Fig. 4, Fig. 6 is a route map for explaining the performance mode, and Fig. 7 is the normal performance mode. FIG. 8 is a signal waveform diagram for explaining envelope control in repeat performance mode and twin mallet performance mode. FIG. 9 is a signal waveform diagram for explaining timer interrupt processing. FIG. 10 is a flowchart showing the timer interrupt processing routine, and FIGS. 11, 12, 13, and 14 are the rapid decay subroutine 5P26, repeated waveform key-on subroutine 5P27, and disable end subroutine 5P28 of FIG. 10, respectively.
, a flowchart showing the repeated waveform key-off subroutine 5P29, FIG. 15 is a flowchart showing the key-off event processing routine in FIG. 4, FIG. 16 is a flowchart showing the tone change processing routine in FIG. 4, and FIG.
18 and 19 are signal waveform diagrams for explaining the operations in the normal performance mode, repeat performance mode, and twin mallet performance mode, respectively. 1...Electronic musical instrument, 2...Keyboard section, 3.
...Panel operation section, 5...CPU, ?・
... Register section, 8 ... Musical tone signal generation section (TG), 10 ... Performance mode conversion table and tone parameter memory section, 11 ... For interrupt timer. 7 register section register lambda oQt
ε) (C) 楕6 Please refer to the diagram.

Claims (3)

[Claims] (1) In an electronic musical instrument that is configured to generate an envelope musical tone having a plurality of repetitive waveforms based on key-on operation information and key-off operation information obtained by key-on operation and key-off operation, based on the key-on operation information, musical key-on processing information generating means for generating musical key-on processing information; musical key-off processing information generating means for generating musical key-off processing information based on the key-off operation information; and the musical key-on information and the musical key-off information are provided. a musical tone signal generating means for generating an enveloped musical tone signal having the plurality of repetitive waveforms, and when the key-off operation information is generated at a timing corresponding to a predetermined period among the generation periods of the repetitive waveforms, the predetermined period and musical tone key-off processing information holding means for holding the musical tone key-off processing information until the period of time has elapsed, and generating the musical tone signal having a naturally decaying waveform from the musical tone signal generating means at the time when the musical tone key-off processing information has elapsed. envelope control device. (2) comprising rapid decay waveform forming means for forming a rapid decay waveform in the decay waveform portion of the plurality of repetitive waveforms; 2. The envelope control device according to claim 1, wherein musical tone key-off processing information is held. (3) The musical tone key-off processing information holding means holds the musical tone key-off processing information during a period in which attack waveform portions of the plurality of repetitive waveforms are formed. Envelope control device as described in.