Although it would be possible to just rip a SID emulator, I’m more into
extending the design ideas into a virtual “SuperSID”. It would play on any
wave output device - speaker and Covox DAC included, provided you’re on a
platform with a driver for that. (It would not, however, play on a real
SID, such as a HardSID card, as it would rely on more channels, more
waveforms, more powerful filters etc.)
In fact, something’s really missing in todays multimedia environments is a
really good music and sound system. Maybe I’ve missed some developments,
but I think there is a big lack.
I’m approaching this from two directions; serious “real” music, as well as
games music and sound effects, and yes, I think something has gone terribly
wrong with synthetic audio in general the last few years.
The only way to get real sound is to use professional hardware, or possibly
VST Instruments or other soft synths. There is Buzz and some other, less
succossful attempts at creating something less expensive than a Cubase VST +
VST Instruments and/or hardware setup, but that doesn’t in any way address
the problem that there is no file format standard that’s even remotely
usable. All you can do is renderd everything to mp3, or burn a CD-R.
The module formats seem to have stagnated on a level where they’re hardly
usable for serious music (games or other). They’re either way too limited, or
they rely on custom plugins rather than a portable instrument definition
standard.
The problem with most systems that use plugins in a useful way is that the
plugins either require way too much CPU power, or sound like crap - or
both! Another problem is that lots of them aren’t released under licenses
that make them usable in games and other multimedia. The licenses also
frequently prevent any form of porting or translation of the plugins to a
form where they would fit in a system like this “professional modular
tracker”.
Another problem is that it’s just incredibly hard to come up with a design
that’s powerful, flexible and comprehensible enough to get people to switch
from modules or mp3/CD-DA.
How about creating an easy to use sound
generation system with FM synthesis and sample effects? Something, like the
good old IMuse of Lucasarts, where the music was affected by gameevents or
like the MusyX system from Factor5 (Chris Huelsbecks company, they even
co-designed the game cube sound system). Shure, this is definitly nothing
for SDL, but it would be a nice 2nd level project I would like to
contribute to. Sadly, I don’t have enough music knowledge to define, what
would be needed, so if someone wants to join, we could do some brain
storming. I think of something like a parallel thread, which reads some
kind of instruction memory. That would make the system a virtual sound
processor, which can be programmed with all kinds of effects (Or so, just
bringing some thoughts up…), including some kind of mod successor, which
is implemented by default with some other library effects and may be some
kind of default sample set, like in general midi…
Well, in theory, this would be great, and indeed, something like that is part
of the motivation behind MAIA. However, many have tried, but all attempts
have more or less failed so far; most of them miserably… Maybe many of the
hackers set out to solve the wrong problems, or simply didn’t know what they
were doing, but I still find it hard to believe that it’s just a matter of
getting started with a design based on the right vision.
Anyway, FWIW, here are some observations I’ve made when trying to figure out
why all these projects have failed to become my favourite music solution:
* Everything must be Free/Open Source.
* There must be no dependancy on any specific editor or input method.
* Smooth integration with existing solutions must be possible.
* Scalability cannot be stressed enough; preferably, everything
should scale from microcontrollers to GHz class SMP systems.
* There is no such thing as a professional/non professional
distinction WRT demands on tools, performance and quality.
* Game music and other music have slightly different requirements,
but these will *not* be satisfied by separate, different tools.
* "MIDI [or similar] *or* Audio?" is the wrong question. Most
serious music work will require both at some point, if not all
the way from idea to CD master/game track.
* There is no superior way of writing/recording/editing music. Most
musicians prefer to use a combination of different methods, and
there are as many approaches as there are musicians.
//David
.- M A I A -------------------------------------------------.
| Multimedia Application Integration Architecture |
| A Free/Open Source Plugin API for Professional Multimedia |
----------------------> http://www.linuxaudiodev.com/maia -' .- David Olofson -------------------------------------------. | Audio Hacker - Open Source Advocate - Singer - Songwriter |--------------------------------------> david at linuxdj.com -'On Thursday 08 March 2001 19:48, Andreas Podgurski wrote: