15 Jan
2003
15 Jan
'03
7:59 a.m.
also supports
MIDI, which I believe is a problem for LADSPA (Taybin?).
Yes. XAP is supposed to address this (at an abstract level), I believe
that explicit MIDI support ala VSTi is a serious mistake. 
15 Jan
15 Jan
8:48 a.m.
New subject: [linux-audio-user] VST plugins and linux
On Wed, Jan 15, 2003 at 11:50:26 +0000, nick wrote:
Everyone seems to love knocking MIDI, but MIDI is a
*real* standard,
supported by countless manufacturers for two decades now. Any
improvements or a new standard are going to arise from the MIDI
consortium itself and the support of manufacturers.. If MIDI were too
primitive for the job, then we wouldn't have all these neat hardware
synths!! MIDI's simplicity is also one of its greatest strengths along
with its low overhead and easy processing.
MIDI isn't too primitive to be useful (obviously), but there are
interesting/useful things you can do if you have more expressive power.
XAP doesn't ignore MIDI, its totaly midi compatible (for input), but it
can do things that MIDI can't (eg. playing overlapping notes at the same
pitch).
Obviously if you want to output things that MIDI can't represent then
you're going to loose something. No supprise.
Yes MIDI is useful, no that doesn't mean you have to be bound my its
limitations in everything you do.
- Steve
3:56 p.m.
New subject: MIDI multiple same-pitch notes / Re: [linux-audio-user] VST plugins and linux
Steve Harris wrote:
XAP doesn't ignore MIDI, its totaly midi
compatible (for input), but it
can do things that MIDI can't (eg. playing overlapping notes at the same
pitch).
MIDI can't play overlapping notes at the same pitch? I was wondering
myself, what the defined behavior was if you have two sucessive note
on's on the same pitch (with two matching note-off's later, of course).
Is this really illegal in MIDI? And what about this common situation:
Sustain-on, note-on, note-off note-on (same pitch), note-off,
Sustain-off>?
I had always assumed that, since I have never read anything in the MIDI
spec concerning this, that it was up to the device how to interpret such
things. I would be mildly surprised however, if the common behavior was
to turn off the last note and then immediately start the new one (when
the second note-on is received). I've always assumed that every box I
have would start a new note at the same pitch (but then I'm not sure if
the note-off's are handled last-note-first or last-note-last).
Does anybody know what the most common behavior is, or if indeed this is
addressed in the MIDI spec?
Larry Troxler
5:57 p.m.
New subject: MIDI multiple same-pitch notes / Re: [linux-audio-user] VST plugins and linux
Steve Harris wrote:
;-) OK, I'll chime in. None of this is directed at you! Truly.
MIDI doesn't play notes. MIDI synths play notes. All MIDI does is send
instructions to a device. The device then implements what it wants as a
response.
Some MIDI devices can play multiple notes at the same pitch on the same MIDI
channel. Manny cannot. I use GigaStudio. It can play 160 copies of the same
sample at the same time if you want it to. Not very interesting, but it does
have it's uses.
XAP doesn't ignore MIDI, its totaly midi
compatible (for input), but it
can do things that MIDI can't (eg. playing overlapping notes at the same
pitch).
MIDI can't play overlapping notes at the same pitch? I was wondering
myself, what the defined behavior was if you have two sucessive note
on's on the same pitch (with two matching note-off's later, of course).
Again, the answer is somewhat different if these two notes are on the same
channel or on different channels.
Is this really illegal in MIDI?
Don't know. That's a spec question.
And what about this common situation:
Sustain-on, note-on, note-off note-on (same pitch), note-off,
Sustain-off>?
It depends on the capabilities of the synth. In many synths this is the way
they operate. In many samplers they can operate differently.
I had always assumed that, since I have never read anything in the MIDI
spec concerning this, that it was up to the device how to interpret such
things.
I believe this is true.
I would be mildly surprised however, if the common
behavior was
to turn off the last note and then immediately start the new one (when
the second note-on is received). I've always assumed that every box I
have would start a new note at the same pitch (but then I'm not sure if
the note-off's are handled last-note-first or last-note-last).
Not every box I have!!
Does anybody know what the most common behavior is, or if indeed this is
addressed in the MIDI spec?
I do not know the most common behavior.
6:07 p.m.
New subject: MIDI multiple same-pitch notes / Re: [linux-audio-user] VST plugins and linux
Some MIDI devices can play multiple notes at the same
pitch on the same
MIDI channel. Manny cannot. I use GigaStudio. It can play 160 copies of the
same sample at the same time if you want it to. Not very interesting, but
it does have it's uses.
Are you running GStudio under WINE in Linux,
or are you simply running it under Windows? Just curious.
Sorry fo the OT, but the mention of GigaStudio caught my eye, and afterall
this is a Linux-Audio-User forum.
--
Jim Hines
Redhat Linux v8.0
1.4ghz Athlon/256mb DDR
8:17 p.m.
New subject: MIDI multiple same-pitch notes / Re: [linux-audio-user] VST plugins and linux
-----Original Message-----
From: linux-audio-user-admin(a)music.columbia.edu
[mailto:linux-audio-user-admin@music.columbia.edu]On Behalf Of Jim Hines
Sent: Wednesday, January 15, 2003 2:01 PM
To: linux-audio-user(a)music.columbia.edu
Subject: Re: MIDI multiple same-pitch notes / Re: [linux-audio-user] VST
plugins and linux
Are you running GStudio under WINE in Linux,
or are you simply running it under Windows? Just curious.
Jim,
Hi. I don't consider it off topic at all. I run two PCs. The Linux box
runs Ardour and Rosegarden, both using Jack, along with a few Linux soft
synths that are jackified. (right now RTSynth, a little SSM, and hopefully
amSynth soon...) This PC is the master machine that makes things tick.
The second PC runs Win ME and GigaStudio. It's dedicated to that purpose
and really just operates as an external synth. (Linux Sampler anyone??????)
I have the latest CVS for Wine and have built it with Jack enabled. I do
see Windows apps making jack connections in QJC, but I have not been able to
get a single sound out of that yet.
Even if I did, and presuming that GSt would even load under Wine, I am
not confident that Wine is efficient enough to run GSt correctly. Optimum
GSt operation requires a sample drive for GSt to get samples without the
Windows equivalent of an xrun in jack. Ardour requires an audio drive for
moving recorded audio in and out of the machine successfully. I somehow
think it's not the best idea to try and do all of this in the same machine.
I can promise you today that it won't work in Windows. I've tried it!
Also, GSt uses a special set of drivers for kernel level control, and I
haven't a clue if that would be acceptable under Wine.
All in all, just too much work for me right now. Maybe one day though...
Cheers,
Mark
16 Jan
16 Jan
12:42 a.m.
New subject: MIDI multiple same-pitch notes / Re: [linux-audio-user] VST plugins and linux
Mark Knecht wrote:
<snip stuff supporting this view>
Yup. That's what I thought - that it's entirely up to the MIDI device to
decide what to do with overlapping notes on the same key number.
If this is indeed the case (and like you, I certainly don't remember
anything in the MIDI spec pertaining to this), then it would fit with
the general principle that apart from some general guidelines, the MIDI
spec generally leaves it up to the instrument to determine how a
specific message is to be interpreted.
Larry Troxler
Steve Harris wrote:
;-) OK, I'll chime in. None of this is directed at you! Truly.
MIDI doesn't play notes. MIDI synths play notes. All MIDI does is send
instructions to a device. The device then implements what it wants as a
response.
XAP doesn't ignore MIDI, its totaly midi
compatible (for input), but it
can do things that MIDI can't (eg. playing overlapping notes at the same
pitch).
MIDI can't play overlapping notes at the same pitch? 
15 Jan
15 Jan
11:08 a.m.
New subject: [linux-audio-user] VST plugins and linux
nick wrote:
I still have trouble understanding this view (except
for non-standard tunings..)
Consider the 0-127 steps of MIDI volume control (ctrl #7). Does that
really seem adequate for a continuous controller ? The granularity of
most MIDI controllers is a real problem for some kinds of composition.
Consider too that MIDI was designed as a keyboard-oriented protocol, so
it's entire underlying assumption is that you'll be making music with a
keyboard. MIDI is notoriously "difficult" for MIDI controllers other
than keyboards, drum machines, and sequencers.
Everyone seems to love knocking MIDI, but MIDI is a
*real* standard,
supported by countless manufacturers for two decades now. Any
improvements or a new standard are going to arise from the MIDI
consortium itself and the support of manufacturers.. If MIDI were too
primitive for the job, then we wouldn't have all these neat hardware
synths!! MIDI's simplicity is also one of its greatest strengths along
with its low overhead and easy processing.
All true. However, we had great synths before MIDI (take a look at what
a Matrix 10 still gets on the used market) and we've had some real
clunkers since MIDI (though generally I agree with you: MIDI did bring
about cheaper and better hardware).
MIDI does certain kinds of work beautifully well. It can be twisted into
doing some things it wasn't really designed for, and there are many
things it simply isn't good for at all. If I want to write a
humanly-playable (or maybe not) piece for piano then I'm likely to use a
MIDI sequencer. If I want to write a piece with a sort of timbral flow
(klangfarbenmelodie) as its compositional basis, with no discernible
pulse or beat rhythm, then I'm probably not going to use MIDI.
I hope that one day some commercial manufacturers will
port their apps
to Linux, but does anyone honestly expect their going to use XAP? No, as
soon as this happens XAP will be forgotten. And XAP is just going to
alienate potential audio developers from Linux - from where i stand it
just looks too complicated and over-engineered. I can't help but feel a
lot of energy is being spent here by well-meaning individuals which
could be better spent by settling on a simpler standard and getting on
with application support, the biggest gap for Linux.
Again, there is a gap with regard to *certain kinds* of music software,
namely the big packages from Steinberg and Cakewalk. But there are other
tools available for Linux musicians, tools that do not assume that the
user is coming from a MIDI-oriented aproach to music.
I believe you've oversimplified the argument that the big manufacturers
are just going to roll in and wipe out the work already done by the LAD
people (and others). I doubt that Steinberg et al. have a team boning up
on their Linux GUI toolkit skills, and they're definitely going to need
such a team. They'll have to determine how to handle the existence of
two primary graphics toolkits (GTK and Qt; I know there are others but
those two are most commonly encountered in Linux applications). Plus
they're going to have to learn a great deal about Linux itself and the
existing (and evolving) Linux audio infrastructure; I doubt they'll want
to replace it with their own, and if they do then that's another number
of years and quantity of manpower that they will not commit to Linux
ports unless they're sure to make money from them.
Well, they obviously don't see the money here now, so they're not coming
to the show. Meanwhile, Paul Davis, Kai Vehmanen, the ALSA team, Werner
Schweer, the Rosegarden guys, Andrew Morton, and so many others are
going straight ahead designing and implementing a professionally-capable
Linux audio system and applications for it. It's quite possible that by
the time Steinberg releases Cubase4Linux no-one will want it: Ardour,
XAP, ALSA, and JACK are likely to have evolved to the point where a
purely open-source alternative to the big packages will exist. Not much
of a contest then: pay $$$ for a closed-source single-vendor package, or
go with open-source GPL'd software in constant development by a talented
community of programmers dedicated to the improvement and enrichment of
the Linux audio world. This is a development model that even Microsoft
admits is too tough for them to meet head-on, so I don't think it'll be
any easier for Steinberg and Co.
Btw, I should emphasize too that I think Steinberg, Logic,
Cakewalk/Sonar, and other similar manufacturers produce damned fine
software. But so do we, and just like them we're getting better and
better at it. The future for Linux audio looks pretty bright from
here...
Best regards,
== Dave Phillips
The Book Of Linux Music & Sound at http://www.nostarch.com/lms.htm
The Linux Soundapps Site at http://linux-sound.org
7985
days inactive
7985
days old
linux-audio-user@lists.linuxaudio.org
7 comments
6 participants
participants (6)
-
Dave Phillips -
Jim Hines -
Larry Troxler -
Mark Knecht -
nick -
Steve Harris