Actually you can use NSF in your NES program, the problem is that you can't choose where the code will be located, and you'll not be able to play sound effects. It is acceptable for a simple intro or a music disc. I guess there could be problems with expansion chips, because banking could be used, for 'linear' memory model it is very simple:

  lda #0                ;init music, call it once, A is song number
  jsr $ac00
...
  jsr $ac03            ;play music, call it every frame
...
  .bank 1
  .org $a2e2-$80
  .incbin "music.nsf"

i know you can export .bin files from famitracker, but how do you use that in your assembly code? is there a web page somewhere with a tutorial or code examples?

"DMC off! DMC on! DMC off! DMC on!"

Neil and Nick sitting in a tree...

...N - E - S - and - D - M -G!

yes, that makes perfect sense. i'll try it out and post my results.

and seriously, thank you so much for helping me out.

No worries Nick

$4015 is the voice enable register and uses bits 0 to 4 to enable Pulse 1, Pulse 2, Triangle, Noise and DCM respectively.

Binary    Hex Voice
%00000001 $01 Pulse 1
%00000010 $02 Pulse 2
%00000100 $04 Triangle
%00001000 $08 Noise
%00010000 $10 DCM

To figure out the value to write to $4015 just add up the values. For example, to enable the two Pulse voices, you write $03 to $4015 ($01 + $02). The APU is generally initialised with a value of $0F which enables the first four voices ($01 + $02 +$04 +$08 = $0F).

DCM differs slightly in that it requires you to turn the voice off and back on again to actually play the sample. In Don's code, he's not using any of the other voices so he doesn't need to turn them on, hence the $00 write (and probably $10 somewhere further down the code to start the sample playing).

So for you, instead of writing #$00 to $4015, write #$0F. This leaves the first four voices on but clears the enable bit for DCM, as is required. Then to start the sample write #$1F to $4015.

Make sense?

yes, that makes perfect sense. i'll try it out and post my results.

and seriously, thank you so much for helping me out.

Binary    Hex Voice
%00000001 $01 Pulse 1
%00000010 $02 Pulse 2
%00000100 $04 Triangle
%00001000 $08 Noise
%00010000 $10 DCM

To figure out the value to write to $4015 just add up the values. For example, to enable the two Pulse voices, you write $03 to $4015 ($01 + $02). The APU is generally initialised with a value of $0F which enables the first four voices ($01 + $02 +$04 +$08 = $0F).

DCM differs slightly in that it requires you to turn the voice off and back on again to actually play the sample. In Don's code, he's not using any of the other voices so he doesn't need to turn them on, hence the $00 write (and probably $10 somewhere further down the code to start the sample playing).

So for you, instead of writing #$00 to $4015, write #$0F. This leaves the first four voices on but clears the enable bit for DCM, as is required. Then to start the sample write #$1F to $4015.

Make sense?

i want to include some dpcm samples in one of my programs. so, i checked out the source code of no carrier's program "lickshot" - http://www.no-carrier.com/index.php?/lickshot/

it seems pretty easy to understand and i was able to get it working in my program just fine. but here's my problem. no carrier's code starts off by writing #$00 to $4015. this ends up muting whatever other audio i have going on in my program at that point.

i was working on a simple program where the buttons of the first player controller would play chords and single notes on the pulse and triangle channels, while the buttons of the second player controller would play drum sounds on the noise and dpcm channels. so, if the first player wants to hold a note out for an extended period of time, it gets silenced whenever the second player triggers one of the samples.

i was wondering what you'd suggest in this instance. i'm probably missing something extremely obvious. thanks for the help in advance!

Looks like you've got a grasp of it

Reading and writing to the .sav file is pretty straightforward. One of the big advantages of using them is that it massively expands the available RAM. You can treat it like all the other RAM in the NES - just remember that any variables you put in there will be restored to the last saved value, unlike 'normal' RAM which is volatile.

woo! next step - adding this feature to my music programs.

Reading and writing to the .sav file is pretty straightforward. One of the big advantages of using them is that it massively expands the available RAM. You can treat it like all the other RAM in the NES - just remember that any variables you put in there will be restored to the last saved value, unlike 'normal' RAM which is volatile.