Per Iain's suggestion, I'm attempting to run csound from the DOS Command
Prompt. I'm barely competent to do this, so I may be doing any number of
things wrong, but I seem to be able to run CsoundVST.csd and send MIDI input
to its real-time process. Among the results in the DOS box are the following
lines, which print out when I play the MIDI keyboard while the .csd is
running:

SECTION 1:
  rtevent:         T754.220 TT754.220 M:      0.0      0.0
new alloc for instr 5:
  midiKeyOct:      pfield:   4  value:   8
  midiVelocity:    pfield:   5  value:  79
  rtevent:         T787.736 TT787.736 M:  14164.4  14164.4
  rtevent:         T806.719 TT806.719 M:  10456.8  10456.8
  midiKeyOct:      pfield:   4  value:   7
  midiVelocity:    pfield:   5  value:  79
  rtevent:         T836.654 TT836.654 M:  13442.7  13442.7
  rtevent:         T920.663 TT920.663 M:  12408.8  12408.8
  midiKeyOct:      pfield:   4  value:   8
  midiVelocity:    pfield:   5  value:  98
  rtevent:         T949.759 TT949.759 M: 128521.8 128521.8
         number of samples out of range:   555140   555140
  rtevent:         T974.774 TT974.774 M:  91585.5  91585.5
         number of samples out of range:      698      698
  midiKeyOct:      pfield:   4  value:   7
  midiVelocity:    pfield:   5  value:  96
  rtevent:         T999.497 TT999.497 M:  97347.7  97347.7
         number of samples out of range:   357631   357631
  rtevent:         T1069.081 TT1069.081 M:  87394.1  87394.1
         number of samples out of range:      367      367
  midiKeyOct:      pfield:   4  value:   8
  midiVelocity:    pfield:   5  value: 112
  rtevent:         T1090.065 TT1090.065 M: 545175.1 545175.1
         number of samples out of range:   818058   818058
  rtevent:         T1164.072 TT1164.072 M: 528496.9 528496.9
         number of samples out of range:     2584     2584
  midiKeyOct:      pfield:   4  value:   7
  midiVelocity:    pfield:   5  value: 121
  rtevent:         T1186.838 TT1186.838 M:1611879.11611879.1
         number of samples out of range:   953327   953327
B  0.000 ..3600.000 T3600.000 TT3600.000 M:1137203.01137203.0
         number of samples out of range:     3886     3886
Score finished in csoundPerform().
inactive allocs returned to freespace
end of score.              overall amps:1611879.11611879.1

This is good, because it shows note-ons and note-offs are being received by
csound. However, I'm not hearing anything. (Probably just as well, given the
amount of clipping shown above.) Conversely, Csound is definitely able to
find the audio output bus when run from the GUI (although the output is
complete crap today, though it was fine yesterday when I was using a
different interface ... something to do with buffering, I'm sure -- God only
knows what).

In addition to not sending audio anywhere, CSoundVST.csd from the command
line quits after a few seconds, even though there's an f0 3600 line in the
score. (That's the entire score, in fact.) So I still have at least three
problems that I'm aware of:

1) No audio output when csound is run from the command line, even though
there's MIDI input.

2) A score that quits when it shouldn't.

3) An audio buffer size/rate problem in real-time rendering that didn't
exist yesterday, before I switched from a USB interface to a FireWire
interface.

Just out of pure academic curiosity, why is it that none of the options in
the drop-down under "Real time audio module" are devices that are in my
system? Where is the GUI getting its confused misinformation from? What are
pa_cb and pa_ci? What, for that matter, is PortAudio?

When I run a different .csd from the command line, the audio output is just
fine, not glitchy -- but this .csd doesn't seem to want to accept MIDI input
from this line in an instr:

kctrl2  ctrl7   1, 4, 0, 127

It compiles, but when I wiggle the slider that sends control change 4, I get
no audible change and no print statement. I am, needless to say, unable to
figure out why. But here's a clue. In order to try to get my uncooperative
test file to to respond to MIDI, I next copied the CSOptions from
CSoundVST.csd into it, replacing the options I had been using before. The
line looks like this:

csound -f -h -d -n -m7 --midi-key-oct=4 --midi-velocity=5

When I run my test file again, it quits after a few seconds rather than
remaining active as it did before. It didn't quit before; it opened a slider
panel and kept right on running (thanks to an f0 3600 score statement).
Frolicking through the page that lists the command-line flags, I can't see
that any of the above items would cause csound to ignore an f0 3600 and bail
out after a few seconds ... but doubtless I'm misunderstanding something
basic. As usual.

Suggestions most humbly appreciated.

--Jim Aikin

-- 
View this message in context: http://www.nabble.com/MIDI-input%2C-audio-output-tp24749297p24749297.html
Sent from the Csound - General mailing list archive at Nabble.com.

Just some ideas to the commandline:
csound -f -h -d -n -m7 --midi-key-oct=4 --midi-velocity=5
-n means no sound to disk. This may be the reason of not hearing  
anything. Have you tried -odac instead?
Try removing -h (makes no sense here I think).
And also remove -f (to see whether your audio device has a problem  
with float output).
Perhaps the amplitude output is correct if you use --midi-velocity- 
amp=5 instead of --midi-velocity=5.
Hope that helps -
	joachim


Am 31.07.2009 um 02:20 schrieb Jim Aikin:

>
> Per Iain's suggestion, I'm attempting to run csound from the DOS  
> Command
> Prompt. I'm barely competent to do this, so I may be doing any  
> number of
> things wrong, but I seem to be able to run CsoundVST.csd and send  
> MIDI input
> to its real-time process. Among the results in the DOS box are the  
> following
> lines, which print out when I play the MIDI keyboard while the .csd is
> running:
>
> SECTION 1:
>  rtevent:         T754.220 TT754.220 M:      0.0      0.0
> new alloc for instr 5:
>  midiKeyOct:      pfield:   4  value:   8
>  midiVelocity:    pfield:   5  value:  79
>  rtevent:         T787.736 TT787.736 M:  14164.4  14164.4
>  rtevent:         T806.719 TT806.719 M:  10456.8  10456.8
>  midiKeyOct:      pfield:   4  value:   7
>  midiVelocity:    pfield:   5  value:  79
>  rtevent:         T836.654 TT836.654 M:  13442.7  13442.7
>  rtevent:         T920.663 TT920.663 M:  12408.8  12408.8
>  midiKeyOct:      pfield:   4  value:   8
>  midiVelocity:    pfield:   5  value:  98
>  rtevent:         T949.759 TT949.759 M: 128521.8 128521.8
>         number of samples out of range:   555140   555140
>  rtevent:         T974.774 TT974.774 M:  91585.5  91585.5
>         number of samples out of range:      698      698
>  midiKeyOct:      pfield:   4  value:   7
>  midiVelocity:    pfield:   5  value:  96
>  rtevent:         T999.497 TT999.497 M:  97347.7  97347.7
>         number of samples out of range:   357631   357631
>  rtevent:         T1069.081 TT1069.081 M:  87394.1  87394.1
>         number of samples out of range:      367      367
>  midiKeyOct:      pfield:   4  value:   8
>  midiVelocity:    pfield:   5  value: 112
>  rtevent:         T1090.065 TT1090.065 M: 545175.1 545175.1
>         number of samples out of range:   818058   818058
>  rtevent:         T1164.072 TT1164.072 M: 528496.9 528496.9
>         number of samples out of range:     2584     2584
>  midiKeyOct:      pfield:   4  value:   7
>  midiVelocity:    pfield:   5  value: 121
>  rtevent:         T1186.838 TT1186.838 M:1611879.11611879.1
>         number of samples out of range:   953327   953327
> B  0.000 ..3600.000 T3600.000 TT3600.000 M:1137203.01137203.0
>         number of samples out of range:     3886     3886
> Score finished in csoundPerform().
> inactive allocs returned to freespace
> end of score.              overall amps:1611879.11611879.1
>
> This is good, because it shows note-ons and note-offs are being  
> received by
> csound. However, I'm not hearing anything. (Probably just as well,  
> given the
> amount of clipping shown above.) Conversely, Csound is definitely  
> able to
> find the audio output bus when run from the GUI (although the output  
> is
> complete crap today, though it was fine yesterday when I was using a
> different interface ... something to do with buffering, I'm sure --  
> God only
> knows what).
>
> In addition to not sending audio anywhere, CSoundVST.csd from the  
> command
> line quits after a few seconds, even though there's an f0 3600 line  
> in the
> score. (That's the entire score, in fact.) So I still have at least  
> three
> problems that I'm aware of:
>
> 1) No audio output when csound is run from the command line, even  
> though
> there's MIDI input.
>
> 2) A score that quits when it shouldn't.
>
> 3) An audio buffer size/rate problem in real-time rendering that  
> didn't
> exist yesterday, before I switched from a USB interface to a FireWire
> interface.
>
> Just out of pure academic curiosity, why is it that none of the  
> options in
> the drop-down under "Real time audio module" are devices that are in  
> my
> system? Where is the GUI getting its confused misinformation from?  
> What are
> pa_cb and pa_ci? What, for that matter, is PortAudio?
>
> When I run a different .csd from the command line, the audio output  
> is just
> fine, not glitchy -- but this .csd doesn't seem to want to accept  
> MIDI input
> from this line in an instr:
>
> kctrl2  ctrl7   1, 4, 0, 127
>
> It compiles, but when I wiggle the slider that sends control change  
> 4, I get
> no audible change and no print statement. I am, needless to say,  
> unable to
> figure out why. But here's a clue. In order to try to get my  
> uncooperative
> test file to to respond to MIDI, I next copied the CSOptions from
> CSoundVST.csd into it, replacing the options I had been using  
> before. The
> line looks like this:
>
> csound -f -h -d -n -m7 --midi-key-oct=4 --midi-velocity=5
>
> When I run my test file again, it quits after a few seconds rather  
> than
> remaining active as it did before. It didn't quit before; it opened  
> a slider
> panel and kept right on running (thanks to an f0 3600 score  
> statement).
> Frolicking through the page that lists the command-line flags, I  
> can't see
> that any of the above items would cause csound to ignore an f0 3600  
> and bail
> out after a few seconds ... but doubtless I'm misunderstanding  
> something
> basic. As usual.
>
> Suggestions most humbly appreciated.
>
> --Jim Aikin
>
> -- 
> View this message in context: http://www.nabble.com/MIDI-input%2C-audio-output-tp24749297p24749297.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>

CsoundVST.csd is designed to be used within CsoundVST running as a VST
plugin, hence the host receives the audio and there is no direct audio
output.

CsoundAC.csd should do what you want.

Hope this helps,
Mike

On 7/30/09, Jim Aikin  wrote:
>
> Per Iain's suggestion, I'm attempting to run csound from the DOS Command
> Prompt. I'm barely competent to do this, so I may be doing any number of
> things wrong, but I seem to be able to run CsoundVST.csd and send MIDI input
> to its real-time process. Among the results in the DOS box are the following
> lines, which print out when I play the MIDI keyboard while the .csd is
> running:
>
> SECTION 1:
>   rtevent:         T754.220 TT754.220 M:      0.0      0.0
> new alloc for instr 5:
>   midiKeyOct:      pfield:   4  value:   8
>   midiVelocity:    pfield:   5  value:  79
>   rtevent:         T787.736 TT787.736 M:  14164.4  14164.4
>   rtevent:         T806.719 TT806.719 M:  10456.8  10456.8
>   midiKeyOct:      pfield:   4  value:   7
>   midiVelocity:    pfield:   5  value:  79
>   rtevent:         T836.654 TT836.654 M:  13442.7  13442.7
>   rtevent:         T920.663 TT920.663 M:  12408.8  12408.8
>   midiKeyOct:      pfield:   4  value:   8
>   midiVelocity:    pfield:   5  value:  98
>   rtevent:         T949.759 TT949.759 M: 128521.8 128521.8
>          number of samples out of range:   555140   555140
>   rtevent:         T974.774 TT974.774 M:  91585.5  91585.5
>          number of samples out of range:      698      698
>   midiKeyOct:      pfield:   4  value:   7
>   midiVelocity:    pfield:   5  value:  96
>   rtevent:         T999.497 TT999.497 M:  97347.7  97347.7
>          number of samples out of range:   357631   357631
>   rtevent:         T1069.081 TT1069.081 M:  87394.1  87394.1
>          number of samples out of range:      367      367
>   midiKeyOct:      pfield:   4  value:   8
>   midiVelocity:    pfield:   5  value: 112
>   rtevent:         T1090.065 TT1090.065 M: 545175.1 545175.1
>          number of samples out of range:   818058   818058
>   rtevent:         T1164.072 TT1164.072 M: 528496.9 528496.9
>          number of samples out of range:     2584     2584
>   midiKeyOct:      pfield:   4  value:   7
>   midiVelocity:    pfield:   5  value: 121
>   rtevent:         T1186.838 TT1186.838 M:1611879.11611879.1
>          number of samples out of range:   953327   953327
> B  0.000 ..3600.000 T3600.000 TT3600.000 M:1137203.01137203.0
>          number of samples out of range:     3886     3886
> Score finished in csoundPerform().
> inactive allocs returned to freespace
> end of score.              overall amps:1611879.11611879.1
>
> This is good, because it shows note-ons and note-offs are being received by
> csound. However, I'm not hearing anything. (Probably just as well, given the
> amount of clipping shown above.) Conversely, Csound is definitely able to
> find the audio output bus when run from the GUI (although the output is
> complete crap today, though it was fine yesterday when I was using a
> different interface ... something to do with buffering, I'm sure -- God only
> knows what).
>
> In addition to not sending audio anywhere, CSoundVST.csd from the command
> line quits after a few seconds, even though there's an f0 3600 line in the
> score. (That's the entire score, in fact.) So I still have at least three
> problems that I'm aware of:
>
> 1) No audio output when csound is run from the command line, even though
> there's MIDI input.
>
> 2) A score that quits when it shouldn't.
>
> 3) An audio buffer size/rate problem in real-time rendering that didn't
> exist yesterday, before I switched from a USB interface to a FireWire
> interface.
>
> Just out of pure academic curiosity, why is it that none of the options in
> the drop-down under "Real time audio module" are devices that are in my
> system? Where is the GUI getting its confused misinformation from? What are
> pa_cb and pa_ci? What, for that matter, is PortAudio?
>
> When I run a different .csd from the command line, the audio output is just
> fine, not glitchy -- but this .csd doesn't seem to want to accept MIDI input
> from this line in an instr:
>
> kctrl2  ctrl7   1, 4, 0, 127
>
> It compiles, but when I wiggle the slider that sends control change 4, I get
> no audible change and no print statement. I am, needless to say, unable to
> figure out why. But here's a clue. In order to try to get my uncooperative
> test file to to respond to MIDI, I next copied the CSOptions from
> CSoundVST.csd into it, replacing the options I had been using before. The
> line looks like this:
>
> csound -f -h -d -n -m7 --midi-key-oct=4 --midi-velocity=5
>
> When I run my test file again, it quits after a few seconds rather than
> remaining active as it did before. It didn't quit before; it opened a slider
> panel and kept right on running (thanks to an f0 3600 score statement).
> Frolicking through the page that lists the command-line flags, I can't see
> that any of the above items would cause csound to ignore an f0 3600 and bail
> out after a few seconds ... but doubtless I'm misunderstanding something
> basic. As usual.
>
> Suggestions most humbly appreciated.
>
> --Jim Aikin
>
> --
> View this message in context:
> http://www.nabble.com/MIDI-input%2C-audio-output-tp24749297p24749297.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

Michael Gogins-2 wrote:
> 
> CsoundVST.csd is designed to be used within CsoundVST running as a VST
> plugin, hence the host receives the audio and there is no direct audio
> output.
> 
> CsoundAC.csd should do what you want.
> 

Possibly. Difficult to be certain. When I attempt to run it (from the DOS
prompt) I get no immediate error messages. The first note-on I send causes
about 80,000 samples out of range in each channel, the DOS window now says,
"begin closing open devices...", and the process is hung. If I get rid of -f
(which is probably what's causing the samples out of range), there is no
longer _any_ response to MIDI input. In either scenario, the DOS window has
hung. Closing the DOS window seems just a trifle inefficient as a way to
shut down a running Csound process, more especially if the process _were_
producing music.

I did edit the file a bit by commenting out three #define lines, which were
obviously inappropriate on my system, and by substituting -M1 and -odac1 in
CSoptions. Quite possibly that breaks something instead of fixing it, but
those do seem to be the correct numbers for my devices.

As an aside, I can't help feeling that it would be a courtesy to folks who
are just starting to learn Csound if someone could provide them, in the
examples folder, with a simple MIDI synth and complete instructions on how
to get it running. A file that includes PianoTeq gives one, if you'll
forgive my saying so, rather the impression that one is seeing shovelware.

I'm determined to get it running on my system, but I can readily envision
the possibility that a lot of potential Csound users would simply give up at
this point.

--JA
-- 
View this message in context: http://www.nabble.com/MIDI-input%2C-audio-output-tp24749297p24759373.html
Sent from the Csound - General mailing list archive at Nabble.com.

...and of course, neither CsoundAC.csd nor CsoundAC-MIDI.csd can be run from
the GUI in Windows. They both crash the GUI.

--JA

-- 
View this message in context: http://www.nabble.com/MIDI-input%2C-audio-output-tp24749297p24759528.html
Sent from the Csound - General mailing list archive at Nabble.com.

The files you mention run without a hitch from the command line on my
systems (windows and Linux), once I set the macros to include or
exclude the pianoteq, etc.

I do not have the resources to test the files on the variety of
systems that might be used by others. For one thing, I do not even
have a Mac, although that seems to be the most used platform by Csound
musicians.

I agree that these issues are annoying to all and more than annoying
to some beginners. Dealing with them properly would require
re-engineering the input/output driver situation in Csound. it was
designed to function on a very wide variety of computers, operating
systems, and audio and MIDI devices. It was not designed to be easy to
use. One thing that could be added, perhaps, would be a utility to
identify all input and output devices on a system and print out the
Csound options that would access them.

If you come up with an orchestra that you feel works well for you,
please send it to me. I'm quite willing to add it to the distribution,
or to modify the existing files to reflect your experience.

I don't, however, appreciate terms like "shovelware." You are tempting
me to use terms that refer to people who either do not read
documentation, or pretend not not to read documentation.

Regards,
Mike



On 7/31/09, Jim Aikin  wrote:
>
> ...and of course, neither CsoundAC.csd nor CsoundAC-MIDI.csd can be run from
> the GUI in Windows. They both crash the GUI.
>
> --JA
>
> --
> View this message in context:
> http://www.nabble.com/MIDI-input%2C-audio-output-tp24749297p24759528.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com