Search options (Page 8 of 21)

Released the whole source code for PeskyTone, the music player with VIC-20, C64, and NES support. I also have a half-working Plus/4 support, but there is another mysterious issue that I couldn't figure out yet, so it is omitted for now.

While I was blindly experimenting with the extracted data trying to figure how to make it work, Ken Shirriff made a huge post full of precious information on the chip internals. In fact it has been explored back in 80s in an electronic magazine. So now it is totally possible to emulate the thing. I think the Arduino will be a good host - readily available, easily programmable. A more hardcore option would be an ATTiny implementation in assembly, it would function just like the original - one chip and no external components besides the battery and the speaker.

In order to make a software emulator with very authentic sound and limitations, we need to read out the contents of a chip, and make a parser that would yield the melody that is contained in the chip.

There is a bit older but high resolution photo of the UM66 die, so I used it to read out the mask ROM contents visually. This is the UM66T-19L variant that contains the For Alice (Für Elise) song. This video features a music card with this song, I guess that's is that encoded in the binary data below.

We know from the die photos that the LFSR is 7 bit, but the taps are unknown. There are different possibilities on that, and I can't figure out the actual one from the die photo (yet). The input frequency is presumably 32768 Hz, no intermediate dividers. There is a 9 bit divided (or that's a multiple of 9) for the clock frequency that somehow gets converted into the note durations (using another LUT), though.

One ROM on the die is 7x16 bits, that's presumably a pitch LUT for 16 pitches with 7 bits to initialize the LFSR. The weird part is that there is 3 of 16 entries are all 0. The addressing is horizontal (0..15):

0000110001000011
0010001011110000
0001100001010101
0011110000100010
0011001110110101
0010100011000011
0011110010100010

Another ROM, the largest on the die, is 16x24 bit. That's presumably the song data that is presented with 6 bits per note (4 bits for pitch, 2 for duration), 64 notes total. 16 is 1/4 of the sequence length, so the addressing is at partially horizontal (0..15 in the lowest 4 bits of the address), but the arrangement of the chunks is unknown:

1000001001111101
0000001001111101
0000010011111011
1100010001000001
1101010110100100
1101010110100110
1010101101001100
0011101110111110
1000000110000010
1000000110000010
0000001100000100
0011010011011110
0111110011010000
1111110011010010
1111100110100101
1010100101101010
1111111110111111
0111111110111111
1111111101111110
1111111111110110
1111111111110111
1111111111110111
1111111111101111
0111011101111111

Yet another ROM is 4x4 bits. That's presumably the note duration ROM, although presumably a song position is encoded as 4 bits pitch and 2 bits duration. Not sure why there is 16 bits then:

0110
1000
1010
1001

Side note, SjAsmPlus actually has support for 8080 with Z80 syntax, it basically limits valid supported opcodes to the 8080 subset. Activated with --i8080 flag. As there isn't many (if any) good modern 8080 cross assemblers, it comes very handy.

Wow, that's unexpected. Some of my engines guilty with the same thing, though. Squat, for one. Easy to fix, though, but it'll reduce the internal sample rate a bit.

A guy reported a very weird issue with this engine. It is completely messed up on the ZX Spectrum +3. Not too sure what causes it, but the result looks like it runs in the slow memory. Booting in 48K mode does not help, I suppose it should have a fast RAM page at 8000 in the default configuration anyways?

Thanks!

No, I barely managed to get this basic on/off stuff to work properly in time for the compo, so I didn't investigate it further than that. I guess there can be more, and in general Atari XL/XE platform is worthy enough to pay more attention in regards of the 1-bit stuff.

Pretty good one. It sounds kinda oldschool, not filled with fast movements or tiny details that is typical for modern compositions, so it sets up the proper classic mood, although with much improved melodic content and sound quality of (not so) modern engine(s).

Yeah, sounds much like it.

I did implement the viznut waveforms for the VIC-20. It was full of nasty surprises for sure, given the difficulty of making well timed 6502 code (added cycles here and there depending on code/vars locations), and really lacking debug tools. Even worse, VICE has some weird audio filtering going on, and MAME does not support expanded RAM for VIC (5K default model only), so I was not able to hear the sound properly, and my test code didn't fit to run in MAME.

It worked in the end, however, it wasn't much suitable for music. The thing is that these waveforms always contains 8 zeroes and 8 ones, just arranged in different ways. So it can't do the usual 50-25-12-7 duty cycle, or arbitrary bit patterns of PET, it just adds high pitched harmonics into square wave. So a special arrangement for music is a must, that would mostly rely of square wave and emphasize some parts using waveforms.

Here is the key part of the code (waveform_table there is to attempt to arrange waveforms in order from regular bold square to the thinest one):

set_pitch_and_wave:

    pha    
    tya
    bne :+
    
    sta <PV_WAVE_PREV,x
    lda #$7e
    sta VIC20_LOW,x
    pla
    rts
:
    
    asl a
    tay
        
    lda waveform_table,y
    iny
    
    cmp PV_WAVE_PREV,x
    beq @skip_wave
    
    sta PV_WAVE_PREV,x
    sta PV_WAVE_PTN
    
    lda #$7e
    sta VIC20_LOW,x
    
    lda waveform_table,y
    
    ldy #100/5                ;delay t-states/5
:
    dey
    bne :-

    tay

@loop:

    lda <PV_WAVE_PTN        ;3
    and #$80                ;2
    ora wavecnt_table,x        ;4+
    sta VIC20_LOW,x            ;4+
    asl <PV_WAVE_PTN        ;5
    lda <PV_WAVE_PTN        ;3
    nop                        ;2
    nop                        ;2
    nop                        ;2
    dey                        ;2
    bne @loop                ;2/3=32t

@skip_wave:

    pla
    tay
    lda (PV_NOTE_TABLE),y
    sta VIC20_LOW,x
    
    rts

wavecnt_table:

    .byte $7d,$7b,$79
    
waveform_table:

    .byte %11000000,3    ;00
    .byte %10010000,6    ;14
    .byte %10100000,5    ;09
    .byte %10100000,4    ;12
    .byte %10110000,6    ;11
    .byte %10110000,5    ;07
    .byte %10000000,2    ;01
    .byte %10000000,3    ;04
    .byte %11010000,5    ;08
    .byte %10101000,6    ;13
    .byte %10010000,5    ;06
    .byte %11100000,4    ;03
    .byte %10110000,4    ;02
    .byte %11000000,5    ;10
    .byte %10000000,4    ;05
    .byte %00000000,0    ;15

In fact, I think I'll release the whole sound code sometime later. Released album code contains most of it anyways, it only missing VIC-20, C64 and NES dependent parts.