Why does jt10_acc.v discard FM data for channels 0 and 4?

[kunichiko]

I used this core in 6-channel mode (by setting .num_ch(6)), but I noticed that the sound of some channels was slightly different from the real chip.

Upon further investigation, I came across the following comment:

jt12/hdl/jt10_acc.v

Line 78 in 42b1918

// uses channels 0 and 4 for ADPCM data, throwing away FM data for those channels

At this line, there is a comment stating uses channels 0 and 4 for ADPCM data, throwing away FM data for those channels.

However, I wanted to utilize all 6 FM channels, so I commented out lines 82 to 103, and it seems to be working fine now.

Could you please explain why jt10_acc.v discards some FM data?"

[jotego]

As the comment says, the original Yamaha chip did that. I am not sure about YM2608 you're looking at. Maybe that one used all six channels. Review carefully Yamaha's documentation before assuming the behavior was different.

[kunichiko]

@jotego
Thank you for your response.

I've reviewed the comment written in the source code, but I couldn't find any reference to that behavior in the YM2610 application manual. Could you kindly provide the source of your information?

Additionally, I believe your intention is that the jt10 acts as a YM2610, not a YM2610B. It's worth noting that the YM2610 has four FM channels, while the YM2610B has six.

Based on my understanding, I would like to propose the following idea:

When the parameter .num_ch(4) is passed, it should function as a YM2610.
When the parameter .num_ch(6) is passed, it should function as a YM2610B.

Please let me know your thoughts on this suggestion.

[kunichiko]

To organize my thoughts, I have created the following table for the DT/MULTI register:

Port0 (A1=0)
        Jt12      YM2610    YM2610B(YM2608)
REG$30  op0:ch0   N/A       OP1:CH1
REG$31  op0:ch1   OP1:CH1   OP1:CH2
REG$32  op0:ch2   OP1:CH2   OP1:CH3
REG$33  N/A       N/A       N/A
REG$34  op2:ch0   N/A       OP3:CH1
REG$35  op2:ch1   OP3:CH1   OP3:CH2
REG$36  op2:ch2   OP3:CH2   OP3:CH3
REG$37  N/A       N/A       N/A
REG$38  op1:ch0   N/A       OP2:CH1
REG$39  op1:ch1   OP2:CH1   OP2:CH2
REG$3A  op1:ch2   OP2:CH2   OP2:CH3
REG$3B  N/A       N/A       N/A
REG$3C  op3:ch0   N/A       OP2:CH1
REG$3D  op3:ch1   OP4:CH1   OP4:CH2
REG$3E  op3:ch2   OP4:CH2   OP4:CH3
REG$3F  N/A       N/A       N/A

Port1 (A1=1)
        Jt12      YM2610    YM2610B(YM2608)
REG$30  op0:ch4   N/A       OP1:CH4
REG$31  op0:ch5   OP1:CH3   OP1:CH5
REG$32  op0:ch6   OP1:CH4   OP1:CH6
REG$33  N/A       N/A       N/A
REG$34  op2:ch4   N/A       OP3:CH4
REG$35  op2:ch5   OP3:CH3   OP3:CH5
REG$36  op2:ch6   OP3:CH4   OP3:CH6
REG$37  N/A       N/A       N/A
REG$38  op1:ch4   N/A       OP2:CH4
REG$39  op1:ch5   OP2:CH3   OP2:CH5
REG$3A  op1:ch6   OP2:CH4   OP2:CH6
REG$3B  N/A       N/A       N/A
REG$3C  op3:ch4   N/A       OP2:CH4
REG$3D  op3:ch5   OP4:CH3   OP4:CH5
REG$3E  op3:ch6   OP4:CH4   OP4:CH6
REG$3F  N/A       N/A       N/A

This is an example of the table for the DT/MULTI register, and there are other similar registers like that.

Regarding the current implementation, the case statement you wrote discards FM data only at op0:ch0 and op0:ch4. Therefore, I believe the other operators of ch0 and ch4 are still valid (e.g., op1:ch0, op2:ch4, etc.).

Therefore when I use all of the six FM channels, YM2608(YM2610B)'s CH1 and CH4 produces odd sound.