////////////////////////////////////////////////////////
//// MCK/MML BEGINNERS GUIDE ///////////////////////////
//// by nullsleep ////////// product of 8bitpeoples ////
//// version 1.0 /////////// Research & Development ////
////////////////////////////////////////////////////////
Download MCK/MML Beginners Guide v1.0 (.TXT)
Workspace Setup - Generating the NSF - Song Setup - Pulse - Triangle - Noise - DPCM - Macros - Troubleshooting
*******************************************************************************
** OBJECTIVE ******************************************************************
*******************************************************************************
This document is aimed at providing the reader with everything that is needed
in order to begin programming NES tunes with MCK/MML. It outlines all the steps
required in the process, the initial organization of the MCK/MML workspace, the
conventions in programming each of the sound channels of the NES including
some of the most common and useful MML commands and effects, and finally
generating an NSF file and troubleshooting any problems that might come up
during this process.
Thanks: Izumi, Manbow-J, Norix, Virt, Memblers, and everyone at MCK 2ch.
*******************************************************************************
** MCK/MML WORKSPACE SETUP ****************************************************
*******************************************************************************
Download Necessary Files
-------------------------------------------------------------------------------
mck_0228.zip
mckc025.zip
dmcconv005.zip
mckc-e.txt
- currently available from:
http://www.geocities.co.jp/Playtown-Denei/9628/
mkit251_dos.zip
- currently available from:
http://www.magicengine.com/mkit/download.html
Organize MCK/MML Workspace
-------------------------------------------------------------------------------
Create a folder called 'workspace' and unzip all the files contained in
mck_0228.zip into this folder. Next, extract all files from mckc025.zip into
this same folder. Now extract ONLY nesasm.exe from the mkit251_dos.zip into
the 'workspace' folder. None of the other magickit files will be needed. The
'source' folder that was extracted from the mckc zip into the 'workspace' may
also be deleted. Finally, create a new folder inside 'workspace' called
'DMCconv' and extract the files from dmcconv005.zip into this folder.
Create and Modify Files
-------------------------------------------------------------------------------
Create a new text file, call it songdata.mml and place it in your 'workspace'
folder. This is the text file where you will use MML (music macro language) to
program your tune. But first, there are some other things that must be setup.
Open make_nsf.txt and scroll to the end of the file where there will be a
number of .include statements. After the last one, add this line to the file:
.include "songdata.h"
also look for these lines in make_nsf.txt:
.org $800E
db "Song Name"
db $00
.org $802E
db "Artist"
db $00
.org $804E
db "Maker"
db $00
This is the part of the NSF header that identifies the tune. It should be
obvious from the hex values of the org statements that there is only a limited
amount of space for each of the fields, the max length being 31 characters. You
should modify these header fields appropriately as seen below, keeping in mind
the 31 character limit:
.org $800E
db "My First NES Chip"
db $00
.org $802E
db "Nullsleep"
db $00
.org $804E
db "2003 Jeremiah Johnson"
db $00
*******************************************************************************
** GENERATING THE NSF FILE ****************************************************
*******************************************************************************
Create another new text file, open it up and type the following:
mckc_e songdata.mml
del nesmusic.nsf
nesasm -raw make_nsf.txt
ren make_nsf.nes nesmusic.nsf
Save the file, close it, and rename it build.bat, this is a simple batch file
that will run all of the commands to generate an NSF file from the MML data.
First it uses mckc to convert songdata.mml into a chunk of data in songdata.h
that nesasm will then compile along with the MCK sound driver code into an NSF
file. By this point everything should be setup properly, so now you can get
started on programming MML.
*******************************************************************************
** BASIC SONG SETUP ***********************************************************
*******************************************************************************
Header Credits
-------------------------------------------------------------------------------
After opening up songdata.mml in your preferred text editor, the first thing
you should do is add the header lines to the top of the file to identify
yourself as the composer and note the title of the song. For example:
#TITLE My First NES Chip
#COMPOSER Nullsleep
#PROGRAMER 2003 Jeremiah Johnson
This is an optional step, however it is highly recommended, especially if the
MML files will be released publicly. On a side note, the incorrect spelling of
#PROGRAMER is a typo within MCK itself, and therefore must be used.
Channel Layout
-------------------------------------------------------------------------------
The NES has 5 channels to work with, and they are defined in MML as follows:
A this is the first pulse channel
B this is the second pulse channel
C this is the triangle channel
D this is the noise channel
E this is the dpcm channel
This guide will cover the programming conventions for each of these channels,
with A+B being covered together since their operation is identical.
Tempo Settings
-------------------------------------------------------------------------------
Tempo can be set individually for each channel, however its likely that you
will usually want all channels to be playing at the same speed to keep
everything in sync. The tempo for all channels can be set simultaneously like
this:
ABCDE t150
In MML notation this is saying, for channels A, B, C, D, and E set the tempo
to 150 beats per minute. The valid range of values for tempo is 1 to 255.
Volume Settings
-------------------------------------------------------------------------------
The pulse wave channels (A+B) and the noise channel (D) of the NES have volume
control, while the triangle wave channel (C) and the DPCM channel (E) can only
be turned ON or OFF. For the pulse and noise channels, there are two options
for setting volume. The first is to just set a constant volume level, by doing
something like this:
A v15
Which would set the volume level for channel A to 15, which is the highest
volume level available. However, in most cases using a volume envelope is
probably a much better choice than setting a constant volume.
Setting up a basic volume envelopes is simple. If neither a constant volume nor
a volume envelope is defined for the pulse channels (A+B) or the noise channel
(D), you will not hear any sound output on these channels. Here is a simple
volume envelope example:
@v0 = { 10 9 8 7 6 5 4 3 2 }
The volume envelope takes values between 0 and 15. The highest volume being 15,
and 0 being silence. This volume envelope starts off at a high volume and
quickly decays down to a low volume, this last value will be held until another
note is played. You can easily modify the volume envelope or setup more,
further examples will be given, including setting loop points within the
envelope.
*******************************************************************************
** PULSE WAVE CHANNELS (A+B) **************************************************
*******************************************************************************
Initialization
-------------------------------------------------------------------------------
The next step is setting up each channel individually with the properties it
requires, such as note length, octave, duty cycle (for the pulse channels), and
volume envelope. Here is a possible setup for one of the pulse wave channels:
A l8 o4 @01 @v0
Which translates to, for channel A set the default note length to eighth notes,
set the octave to the 4th octave, set the duty cycle setting to 01 (25% duty
cycle), and use volume envelope 0 (which was defined above). A quick
explanation of the duty cycle setting follows.
Duty Cycle Explanation
-------------------------------------------------------------------------------
You can think of a pulse wave as a square wave with a variable width. In a
square wave the width is fixed at 50% (half up and half down), but pulse waves
have more flexibility. This flexibility is referred to as the duty cycle (or
sometimes the timbre) of the pulse wave. Below are the 4 possible duty cycle
settings for the pulse wave channels on the NES.
_
00 | | | 12.5% thin raspy sound
| |_____________|
___
01 | | | 25% thick fat sound
| |___________|
_______
02 | | | 50% smooth clear sound
| |_______|
___________
03 | | | 75% same as 25% but phase-inverted
| |___|
Programming the Pulse Channel
-------------------------------------------------------------------------------
Now that the pulse channel (A) should be completely setup, here is a little
note sequence that can be programmed on it.
A c d e f g4 a16 b16 >c c d e f g4 a16 b16 >c<<
If you know standard music notation most of what you see should look atleast
somewhat familiar. Additionally, the use of sharps and flats is accomplished by
adding either a + or - (respectively) after the note value. Making the notes in
an octave:
c+ d+ f+ g+ a+
| # # | # # # | additionally:
| # # | # # # | r = rest
| # # | # # # | w = wait (rest without silencing the previous note)
|__|__|__|__|__|__|__|
c d e f g a b
Since the default note length for channel A was set to eighth notes, the above
melody is playing c d e f notes for an eighth length each, then the g4 which
plays g for a quarter length, followed by a16 b16 which play an a note and b
note for a sixteenth length each. Next is the > symbol, this switches an octave
UP (so now we are in the 5th octave) and then a 5th octave c eighth note is
played. Now the scale is repeated again, before finally switching back down 2
octaves (to the original 4th octave we set for the channel). Additionally, in
reference to note duration it is possible to use dotted notes, which should
again be familiar to those with knowledge of standard music notation. Dotting
a note increases the duration of that note by half of its value. These examples
should help illustrate this:
c8. = c note played for an eighth plus a sixteenth
d4. = d note played for a quarter plus an eighth
e4.. = e note played for a quarter plus an eighth plus a sixteenth
f2.. = f note played for a half plus a quarter plus an eighth
Now, getting back to the above example programmed on the first pulse wave
channel (A), notice that this sequence will only play once. The full sequence
or small portions of it can be looped using brackets followed by a loop count
as shown below:
A [c d e f g4 a16 b16 >c c d e f g4 a16 b16 >c<<]2
This will loop the entire sequence twice. This can be handy for keeping your
MML code clean and saving you some unnecessary typing. To give this sequence a
little more of a dynamic feel another volume envelope can be setup and the two
can be switched back and forth. Ending up with something like this:
#TITLE My First NES Chip
#COMPOSER Nullsleep
#PROGRAMER 2003 Jeremiah Johnson
@v0 = { 10 9 8 7 6 5 4 3 2 }
@v1 = { 15 15 14 14 13 13 12 12 11 11 10 10 9 9 8 8 7 7 6 6 }
ABCDE t150
A l8 o4 @01 @v0
A [c d e f @v1 g4 @v0 a16 b16 >c c d e f @v1 g4 @v0 a16 b16 >c<<]2
This volume envelope switching will put a slight emphasis on the quarter notes
because of the higher initial volume setting of the new volume envelope and its
slower decay rate. All of this can be applied identically to the second pulse
channel (B) as well.
*******************************************************************************
** TRIANGLE WAVE CHANNEL (C) **************************************************
*******************************************************************************
Initialization
-------------------------------------------------------------------------------
The operation of the triangle wave channel (C) is similar to the pulse wave
channels with the exception of volume envelope and duty cycle parameters. The
triangle channel has no volume control, it is either ON or OFF and therefore
volume envelopes can not be used. Likewise, duty cycle settings only apply to
the pulse channels and can be disregarded when using the triangle channel.
Taking these things into consideration, the initial setup of the triangle
channel is fairly straightforward:
C l4 o3 q6
Which translates to, for channel C set the default note length to quarter
notes, set the octave to the 3rd octave, and finally set q6. This setting might
be confusing, but what it does is cuts the notes played back on this channel
slightly. The q stands for quantize and it can take values from 1 to 8. Notes
are divided into 8 equal parts and what this value determines is how many
eighths of the note to play before cutting it. For example, the q6 setting we
used will cut the note after 6/8ths of it has played. This will help give
basslines a bit of a snappier rhythm than if the notes were just allowed to
sound continuously.
Programming the Triangle Channel
-------------------------------------------------------------------------------
Here is a little example of a bassline sequence on the triangle channel:
C c e g8 g8 a16 b16 >c8 c e g8 g8 a16 b16 >c8<<
Add this into the MML covered so far, and loop it 4 times. This way its first
heard along with whats playing on the pulse wave channel (A) and then it can be
heard alone.
#TITLE My First NES Chip
#COMPOSER Nullsleep
#PROGRAMER 2003 Jeremiah Johnson
@v0 = { 10 9 8 7 6 5 4 3 2 }
@v1 = { 15 15 14 14 13 13 12 12 11 11 10 10 9 9 8 8 7 7 6 6 }
ABCDE t150
A l8 o4 @01 @v0
A [c d e f @v1 g4 @v0 a16 b16 >c c d e f @v1 g4 @v0 a16 b16 >c<<]2
C l4 o3 q6
C [c e g8 g8 a16 b16 >c8 c e g8 g8 a16 b16 >c8<<]4
*******************************************************************************
** NOISE CHANNEL (D) **********************************************************
*******************************************************************************
Noise Channel Explanation
-------------------------------------------------------------------------------
The noise channel (D) can be a fairly versatile instrument with a bit of work.
Waves crashing on a beach, rocket engines roaring, dark fiery dungeon sounds,
and supplemental percussion to enhance your drum samples are a few possible
applications. Like the pulse wave channels, volume envelopes are used by the
noise channel and are an important part of getting good sounds out of it.
Additionally it has 2 modes of operation: normal and looped noise. The looped
noise setting can be used to generate interesting, somewhat metallic sounds.
The pitch range of the noise channel is very limited and loops every octave,
making octave changing unnecessary. The c note seems to be the high pitch, with
the pitch moving slightly downwards over e, f, g, a, and finally to the b note
which seems to be the lowest pitched.
Initialization
-------------------------------------------------------------------------------
Here are a couple of simple volume envelopes that can be used for some basic
percussion on the noise channel:
@v2 = { 15 12 10 8 6 3 2 1 0 }
@v3 = { 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 }
After setting up the volume envelopes the channel should be initialized:
D l4 o1 @0 @v2
Which translates to, for channel D set the default note length to quarter
notes, set the octave to the 1st octave, set the noise mode to normal (@1 would
be used to turn on looped noise), and use volume envelope 2.
Programming the Noise Channel
-------------------------------------------------------------------------------
Here is a little sequence of simple noise drums:
D @v2 b @v3 e @v2 b @v3 e @v2 b @v3 e @v2 b @v3 e8 @v2 b8
Add this into the MML covered so far, and loop it 4 times.
#TITLE My First NES Chip
#COMPOSER Nullsleep
#PROGRAMER 2003 Jeremiah Johnson
@v0 = { 10 9 8 7 6 5 4 3 2 }
@v1 = { 15 15 14 14 13 13 12 12 11 11 10 10 9 9 8 8 7 7 6 6 }
@v2 = { 15 12 10 8 6 3 2 1 0 }
@v3 = { 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 }
ABCDE t150
A l8 o4 @01 @v0
A [c d e f @v1 g4 @v0 a16 b16 >c c d e f @v1 g4 @v0 a16 b16 >c<<]2
C l4 o3 q6
C [c e g8 g8 a16 b16 >c8 c e g8 g8 a16 b16 >c8<<]4
D l4 o1 @0 @v2
D [@v2 b @v3 e @v2 b @v3 e @v2 b @v3 e @v2 b @v3 e8 @v2 b8]4
*******************************************************************************
** DPCM CHANNEL (E) ***********************************************************
*******************************************************************************
DPCM Channel Explanation
-------------------------------------------------------------------------------
The DPCM channel, or delta modulation channel (DMC) as it is also sometimes
referred to, is used for sample playback on the NES. This can be useful for
programming drums, sampled basslines, or even vocal samples. Its operation is
simple and fairly straightforward, with very few parameters that must be set.
Like the triangle wave channel there is no volume control, the DPCM channel is
either ON or OFF. Additionally, the NES uses its own 1-bit sample format, which
you will have to convert your samples to before doing anything else. This
process is described in the next section.
Creating DPCM Samples
-------------------------------------------------------------------------------
DMCconv is the program that will be used to convert your samples from .wav to
.bin for use with MCK. The documentation for DMCconv is not in english,
however its operation is simple enough so that this should not be a problem.
Below is the description of its usage:
Usage: DMCconv wavefile outfile
Options
-r? DMC Sampling rate(0-F) (Default:F 33.14KHz)
0: 4.18KHz 1: 4.71KHz 2: 5.26KHz 3: 5.59KHz
4: 6.26KHz 5: 7.05KHz 6: 7.92KHz 7: 8.36KHz
8: 9.42KHz 9:11.18KHz A:12.60KHz B:13.98KHz
C:16.88KHz D:21.30KHz E:24.86KHz F:33.14KHz
-v? Volume(Default:100)
-n Volume not adjust(Default:Adjust)
-b Bank size padding(Default:No padding)
For example: DMCconv kick.wav kick.dmc
would convert a kick.wav file into a kickdrum sample useable by the NES with
all of the default settings.
Initialization and Programming the DPCM Channel
-------------------------------------------------------------------------------
Once all desired samples are converted, create a directory called 'samples'
within the 'workspace' folder and include the samples and initialize the
channel in the mml as follows:
@DPCM0 = { "samples\kick.dmc",15 }
@DPCM2 = { "samples\snare.dmc",15 }
E o0 l4
The first sample, kick.dmc, will be mapped to @DPCM0 which corresponds to the
c note on octave 0. You will then notice that the second sample, snare.dmc, is
mapped @DPCM2 which corresponds to the d note ... @DPCM1 is skipped over to
avoid mapping samples to sharps/flats in order to keep the MML more readable.
For example, which of the following is easier to recognize as alternating kick
and snare drumbeats?
E c d c d c d c d8 c8
or
E c c+ c c+ c c+ c c+8 c8
It should be apparent that the first one is more recognizeable, and this
increased readability will be further appreciated when programming more complex
or lengthy drum sequences. Looping this and adding it to the MML programmed so
far gives us something like:
#TITLE My First NES Chip
#COMPOSER Nullsleep
#PROGRAMER 2003 Jeremiah Johnson
@v0 = { 10 9 8 7 6 5 4 3 2 }
@v1 = { 15 15 14 14 13 13 12 12 11 11 10 10 9 9 8 8 7 7 6 6 }
@v2 = { 15 12 10 8 6 3 2 1 0 }
@v3 = { 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 }
@DPCM0 = { "samples\kick.dmc",15 }
@DPCM2 = { "samples\snare.dmc",15 }
ABCDE t150
A l8 o4 @01 @v0
A [c d e f @v1 g4 @v0 a16 b16 >c c d e f @v1 g4 @v0 a16 b16 >c<<]2
C l4 o3 q6
C [c e g8 g8 a16 b16 >c8 c e g8 g8 a16 b16 >c8<<]4
D l4 o1 @0 @v2
D [@v2 b @v3 e @v2 b @v3 e @v2 b @v3 e @v2 b @v3 e8 @v2 b8]4
E o0 l4
E [c d c d c d c d8 c8]4
*******************************************************************************
** ADDITIONAL MACROS **********************************************************
*******************************************************************************
This section will cover a few useful macros that can be used to add more depth
and complexity to your tunes. The syntax for the macros will first be given,
followed by the valid range of values that can be used, concluding with
examples and explanations. For a comprehensive listing of MML commands, please
see the mckc-e.txt file by Virt that is mentioned at the beginning of this
document.
Volume Macro @v[num] = { }
-------------------------------------------------------------------------------
Valid range of values for num: 0 to 127
Valid range of values for parameters: 0 to 15
@v0 = { 10 8 6 5 4 3 2 | 3 4 5 6 5 4 }
A @v0 c d e f g a b >c<
B @v0 c d e f g a b >c<
D @v0 c d e f g a b >c<
The volume envelope macro has been covered extensively already, useable by both
the pulse wave channels (A+B) and the noise channel (D). However this example
illustrates a loop point within the macro. Notice the | following the 2, this
is the loop point. All values after this point will loop indefinitely. So for
this example, the channel begins playing at volume 10 down to 8, 6, 5, 4, 3,
and 2 and then increases again to 3, 4, 5, peaking at 6, then back down to
5, 4, and looping back to 3 stepping through the values again. Loop points in
volume macros can be very useful for things such as simulating delay/echo
effects.
Timbre Macro @[num] = { }
-------------------------------------------------------------------------------
Valid range of values for num: 0 to 127
Valid range of values for parameters: 0 to 3
@0 = { | 2 2 2 2 1 1 1 1 0 0 0 0 1 1 1 1 }
A @@0 c d e f g a b >c<
B @@0 c d e f g a b >c<
The timbre macro is useable only on the pulse wave channels (A+B) in order to
rapidly switch the duty cycles of the pulse waves. The above example sets a
loop point immediately following the opening bracket, meaning all values will
be looped. A 50% duty cycle is played for 4 frames, followed by a 25% duty
cycle for 4 frames, followed by a 12.5% duty cycle for 4 frames, followed
again by a 25% duty cycle for another 4 frames, and finally looping back to
the beginning with the 50% duty cycle. Timbre macros can be used to make the
pulse wave channels sound much more interesting. Use a looping timbre macro
like the one above to thicken your tune up a bit. Or use a non-looping macro to
simply give instruments a sense of being plucked.
Arpeggio/Note Macro @EN[num] = { }
-------------------------------------------------------------------------------
Valid range of values for num: 0 to 127
Valid range of values for parameters: -127 to 126
@EN0 = { 0 0 | 4 0 3 0 -7 0 }
A EN0 c f g >c< ENOF
B EN0 c f g >c< ENOF
D EN0 c f g >c< ENOF
The arpeggio (or note) macro is used to step through a series of notes very
rapidly, useable on the pulse wave channels (A+B) and the noise channel (D).
The most common use of this technique is probably for simulating chords on a
single channel. The parameters for the macro are not absolute, but instead
relative to the current note. So, going through the above example, first the
base note is played for 2 frames indicated by the two 0s, next the loop point
is set, following the loop point the note is increased by 4 semitones (so if
the base note is c this will now play e), then followed by a 0 to hold this
new note for a second frame, then this new note is increased by 3 semitones
(bringing it from e to g if we assume c as the base note), followed by a 0 to
hold this new note for a second frame, then finally decreased by 7 semitones
back to the base note and held for a second frame by the following 0 before
jumping back to the loop point. This example is simulating a major chord. The
macro is turned on by issuing the EN[num] command on the channel, and turned
off using ENOF.
Pitch Macro @EP[num] = { }
-------------------------------------------------------------------------------
Valid range of values for num: 0 to 127
Valid range of values for parameters: -127 to 126
@EP0 = { 8 }
@EP1 = { -64 }
A EP0 c EPOF r EP1 c EPOF
B EP0 c EPOF r EP1 c EPOF
C EP0 c EPOF r EP1 c EPOF
D EP0 c EPOF r EP1 c EPOF
The pitch macro, useable on all channels except for the DPCM channel (E),
functions by sliding the frequency of a note with a speed and direction based
upon the given parameters. When the number(s) within the brackets is positive,
the frequency will slide up in pitch. Likewise, when the number(s) within the
brackets is negative, the frequency of the note will slide downwards in pitch.
The farther the value is from 0 (in either direction) the faster the frequency
will sweep. In the above example, two pitch macros are setup, EP0 to give a
slow upward pitch bend and EP1 to give a very extreme pitch bend down. In this
second instance, the pitch is swept downward so quickly and so far that it
wraps around, generating an interesting effect. Both examples use a single
value within the brackets, but it is possible to use multiple values and set a
loop point as in the above arpeggio macro example. Like the arpeggio/note
macro, the pitch macro will remain ON when in use on a channel until it is
issued the EPOF command for turning it off.
Vibrato Macro @MP[num] = { [delay] [speed] [depth] }
-------------------------------------------------------------------------------
Valid range of values for num: 0 to 63
Valid range of values for delay parameter: 0 to 255
Valid range of values for speed parameter: 1 to 255
Valid range of values for depth parameter: 0 to 255
@MP0 = { 8 2 6 }
A MP0 c1 d1 e1 f1 g1 a1 b1 >c1< MPOF
B MP0 c1 d1 e1 f1 g1 a1 b1 >c1< MPOF
C MP0 c1 d1 e1 f1 g1 a1 b1 >c1< MPOF
D MP0 c1 d1 e1 f1 g1 a1 b1 >c1< MPOF
The vibrato (or pitch modulation) macro is useable on all channels except for
the DPCM channel (E). There are 3 parameters that must be set within the
brackets: delay, speed, and depth. Pitch modulation works by sliding the
frequency of a note in a sine wave pattern. The delay parameter determines how
long to wait before activating the vibrato, a value of 0 activating it
immediately. The speed parameter determines the period of the sine wave, the
larger the value the slower the pitch will change. The final parameter is
depth, which determines the amplitude of the sine wave, the larger the value
the greater the pitch range. Like the arpeggio/note macro and the pitch macro,
the vibrato will remain ON when in use on a channel until it is issued the
MPOF command for turning it off.
*******************************************************************************
** TROUBLESHOOTING ************************************************************
*******************************************************************************
DPCM Sample Problems
-------------------------------------------------------------------------------
The DPCM samples can sometimes cause pops in your tunes, little clicks that are
audible before and/or after the samples play. This cannot always be completely
eliminated, but there are some things that you can do to minimize the problem.
First off, 16bit 44100KHz Mono .wav files are recommended for use with DMCconv,
and make sure to check in your sound editor that the samples have a starting
and ending level of 0. Once this is verified, convert the samples with DMCconv
and open the converted samples in a hex editor. If the size of the file is not
a multiple of 16, it should be padded with $AA or $55 until it is.
Bank Overflows
-------------------------------------------------------------------------------
When working on long tunes or using particularly complex MML, you may encounter
the following error output from nesasm:
Bank overflow, offset > $1FFF!
Nesasm code must be organized in a way such that it uses 8K banks. If your
songdata takes up too much space, you will encounter this bank overflow problem
at compilation time. Fortunately, there is an MML command for bankswitching
that will effectively give you more space for your songdata. Its usage is as
follows:
#BANK-CHANGE [num]
Where the valid range of values for num is 1 to 14, each corresponding to a
track in your MML code (1=A, 2=B, 3=C, ... ). If the command is not successful
in eliminating the bank overflow the first time you use it, try bankswitching
different tracks. There is also an alternative usage of this command below:
#BANK-CHANGE [num1],[num2]
Where the valid values for num1 are 0 to 2 which denotes the bank number, and
for num2 they are again 1 to 14 corresponding to a track number. If you are
using the DPCM channel, num1 should be set to 0 to avoid problems since DPCM
samples are stored in banks 1 and 2.
- EOF -------------------------------------------------------------------------
../love_always/nullsleep
../www.8bitpeoples.com
../www.nullsleep.com
If you find any errors in this document or have any suggestions or requests for
additions, please contact me: nullsleep@8bitpeoples.com
Related Links:
http://www.geocities.co.jp/Playtown-Denei/9628/ ; the home of mck
http://www.2A03.org ; NES scene music archive
http://www.vorc.org ; relevant vgm & chiptune news
http://nesdev.parodius.com ; a wealth of NES docs
http://research.8bitpeoples.com ; NES and other 8bit love
|
