[ scummvm-Feature Requests-742249 ] MI1VGA Roland support
SourceForge.net
noreply at sourceforge.net
Sat Aug 9 02:14:37 CEST 2003
Feature Requests item #742249, was opened at 2003-05-23 20:31
Message generated for change (Comment added) made by khalek
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=418823&aid=742249&group_id=37116
Category: None
Group: None
Status: Open
Resolution: Invalid
Priority: 5
Submitted By: Marek Roth (logicdeluxe)
Assigned to: Jamieson Christian (jamieson630)
Summary: MI1VGA Roland support
Initial Comment:
Only Adlib emulations works in the current version.
E.g. '-e windows' has no effect on this game. The
MT32-MIDIs should be translatable to GM like ScummVM
already does it in later games.
----------------------------------------------------------------------
>Comment By: Jonathan Gray (khalek)
Date: 2003-08-09 10:14
Message:
Logged In: YES
user_id=59347
on a similiar note the afore mentioned roland patch for loom
(adds some LFLs),
it would be nice if that could be supported as well :)
http://support.lucasarts.com/patches/loom.htm
It is a self extracting zip file.
----------------------------------------------------------------------
Comment By: Jamieson Christian (jamieson630)
Date: 2003-07-17 15:56
Message:
Logged In: YES
user_id=596642
Well, I downloaded the disk09.lec and ran a hex dump on the
non-room "RO" resources from ScummRev. One "RO" looks like
a valid room resource (albeit without any real useful
information, but there ARE child elements in there), while the
rest look like they COULD possibly pass as MIDI data. If I
were to write a parser off the top of my head, here's how I
would interpret the file:
Music starts at address + 6 bytes (just past the two-
character "RO" header, though it might start a few bytes
later).
Events seem to follow SMF closely. Cx (program change, 1
byte data); Bx (control change, 2 bytes data); 9x (note on, 2
bytes data); and 8x (note off, 2 bytes data); all look to
match up fine. Ax appears numerous time, with no data, and
so far all I've found is 0xA0 (i.e. x == 0 every time). Could
probably be ignored in a first-pass parser.
Events stream one right after another, and are assumed to
occur at the same point in time unless F0 xx is encountered.
This is a delay marker where xx == ticks (presumably) to
delay. xx == 0 means end of track. Also, FF is sometimes
used instead of F0 00 at end of track. Whether this type of
EOT should be treated separately (e.g. does it mean loop to
beginning instead?) is unknown.
So... yeah, I could throw together a parser for this format.
What I'm not so sure about is (1) whether an attempt is made
to load these songs in the MonkeyVGA scripts, and (2) how
we distinguish the music "RO" resources from the room "RO"
resources.
If somebody else can look at the resource
loading/identification issue, I'll see what I can do for parser
code.
Fingolfin, I will leave this tracker item's resolution and status
for you decide upon.
----------------------------------------------------------------------
Comment By: Andrej Sinicyn (andrej4000)
Date: 2003-07-17 08:10
Message:
Logged In: YES
user_id=508463
I think, that logicdeluxe is right!
Go to:
http://support.lucasarts.com/patches/monkey-
1.htm#Roland%20Upgrade
There you can get the Roland-Upgrade for MonkeyEGA
(DISK09.LEC), if you compare the only room in this file with
the last room in DISK04.LEC of MonkeyEGA, you'll find, that
they are identical!
And in fact, tha Roland-Sounds-Ressources have thje
Header "RO".
----------------------------------------------------------------------
Comment By: Max Horn (fingolfin)
Date: 2003-07-17 06:53
Message:
Logged In: YES
user_id=12935
"Check this out with ScummRV" assumes that a) I own that game
and b) I can use ScummRev, neither is true.
Maybe Jamieson wants to say something on this matter <shrug>
----------------------------------------------------------------------
Comment By: Marek Roth (logicdeluxe)
Date: 2003-07-06 04:37
Message:
Logged In: YES
user_id=123766
You are wrong. In fact there ARE Roland MIDI's with the VGA
disk version, though they are a bit hidden. They are
included in disk04.lec in the very last room there is a
bunch of Roland-MIDI's also called RO hence they can be
easily mistaken for rooms. Check this out with ScummRV. Also
noticable when playing directly from disk (despite it is not
recommended in the readme) you have to insert disk 4 every
time a music is to be loaded.
For Monkey EGA and Loom EGA there are free upgrades at
LucasArts for download, which add MT-32 support for those
versions, too. There the MIDI's are stored in separate files
as well. This is disk09.lec for Monkey EGA and several
further lfl files for Loom EGA. (Yes, those old Interpreters
do have an undocumented '-r' switch! Try it without those
upgrates and you will see a file not found error.)
Since you want to know:
ScummVM 0.5.0pre-cvs Windows
Built on Jul 5 2003 21:59:31
But there is no different since the first version supported
monkeyvga with music anyway, since there is simply no MT-32
support implemented in ScummVM yet, hence I did not write
those information in the opening post earlier.
Btw. "-e windows" has no effect with monkeyvga.
----------------------------------------------------------------------
Comment By: Max Horn (fingolfin)
Date: 2003-05-23 20:54
Message:
Logged In: YES
user_id=12935
There is no MT32 midi in MonkeyVGA. And anyway, did you actually try
with latest CVS? For the problem with "-e windows" not working with
MonkeyVGA was only present for a few days, and was fixed some time
ago.
Feel free to reopen this if you really think there is some justification,
but this time please include at least your specs (what ScummVM
version etc.)
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=418823&aid=742249&group_id=37116
More information about the Scummvm-tracker
mailing list