Matt,

I had a look at your frontend. The idea is great; I liked
it.
However, on OSX 10.3 it does not run in realtime. It does
write to disk OK, no problems, but with realtime it just
hangs and does not perform at all. I only have builtin audio
at the moment, so it's not as if I am trying to use a
different
soundcard.  That's either with or without -odac in the
options.

I am wondering whether this is because you have built it on.
OSX 10.4. You don't use either of the two RT IO modules,
do you? (but it actually reports that you are using
portaudio,
which I didn't think you were since it does not have the
limitation
of only using builtin audio).  Perhaps the program, built on
10.4
will not run on 10.3. What do you reckon?

Victor

Could this also be true of the current GUI app in CVS; in that if it is 
built on 10.3.x it will not be deployed correctly on 10.4.x ?


-David

On Feb 11, 2006, at 5:01 AM, Victor Lazzarini wrote:

> Perhaps the program, built on
> 10.4
> will not run on 10.3. What do you reckon?
>
> Victor

i am seeing it too on 10.4

could have something changed from the pre-release i was using to 
develop this and the official csound5 release?
i automatically send -odac and -iadc when "realtime" is selected, and 
assign my own rtaudio callbacks -- portaudio should NOT be called!!
i do not remember seeing any portaudio stuff in the pre-release version of 
the framework..

thanks,
matt

On Sat, 11 Feb 2006, Victor Lazzarini wrote:

> Matt,
>
> I had a look at your frontend. The idea is great; I liked
> it.
> However, on OSX 10.3 it does not run in realtime. It does
> write to disk OK, no problems, but with realtime it just
> hangs and does not perform at all. I only have builtin audio
> at the moment, so it's not as if I am trying to use a
> different
> soundcard.  That's either with or without -odac in the
> options.
>
> I am wondering whether this is because you have built it on.
> OSX 10.4. You don't use either of the two RT IO modules,
> do you? (but it actually reports that you are using
> portaudio,
> which I didn't think you were since it does not have the
> limitation
> of only using builtin audio).  Perhaps the program, built on
> 10.4
> will not run on 10.3. What do you reckon?
>
> Victor
> --
> Send bugs reports to this list.
> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>

Hi Matt,

After poking around in the ../InOut, and ../Engine directories of the 
Csound5 source tree, it seems likely that simply sending "-odac -iadc" 
without the additional -+rtaudio= flag will default to PortAudio. 
Perhaps "-+rtaudio=CoreAudio" should be appended when realtime is 
selected ?


-David

On Feb 11, 2006, at 7:59 PM, Matt J. Ingalls wrote:

>
> i am seeing it too on 10.4
>
> could have something changed from the pre-release i was using to 
> develop this and the official csound5 release?
> i automatically send -odac and -iadc when "realtime" is selected, and 
> assign my own rtaudio callbacks -- portaudio should NOT be called!!
> i do not remember seeing any portaudio stuff in the pre-release 
> version of the framework..
>
> thanks,
> matt
>
> On Sat, 11 Feb 2006, Victor Lazzarini wrote:
>
>> Matt,
>>
>> I had a look at your frontend. The idea is great; I liked
>> it.
>> However, on OSX 10.3 it does not run in realtime. It does
>> write to disk OK, no problems, but with realtime it just
>> hangs and does not perform at all. I only have builtin audio
>> at the moment, so it's not as if I am trying to use a
>> different
>> soundcard.  That's either with or without -odac in the
>> options.
>>
>> I am wondering whether this is because you have built it on.
>> OSX 10.4. You don't use either of the two RT IO modules,
>> do you? (but it actually reports that you are using
>> portaudio,
>> which I didn't think you were since it does not have the
>> limitation
>> of only using builtin audio).  Perhaps the program, built on
>> 10.4
>> will not run on 10.3. What do you reckon?
>>
>> Victor

but i do not want to use any rtmodule -- i am using my own callbacks!??

On Sat, 11 Feb 2006, David Akbari wrote:

> Hi Matt,
>
> After poking around in the ../InOut, and ../Engine directories of the Csound5 
> source tree, it seems likely that simply sending "-odac -iadc" without the 
> additional -+rtaudio= flag will default to PortAudio. Perhaps 
> "-+rtaudio=CoreAudio" should be appended when realtime is selected ?
>
>
> -David
>
> On Feb 11, 2006, at 7:59 PM, Matt J. Ingalls wrote:
>
>> 
>> i am seeing it too on 10.4
>> 
>> could have something changed from the pre-release i was using to develop 
>> this and the official csound5 release?
>> i automatically send -odac and -iadc when "realtime" is selected, and 
>> assign my own rtaudio callbacks -- portaudio should NOT be called!!
>> i do not remember seeing any portaudio stuff in the pre-release version of 
>> the framework..
>> 
>> thanks,
>> matt
>> 
>> On Sat, 11 Feb 2006, Victor Lazzarini wrote:
>> 
>>> Matt,
>>> 
>>> I had a look at your frontend. The idea is great; I liked
>>> it.
>>> However, on OSX 10.3 it does not run in realtime. It does
>>> write to disk OK, no problems, but with realtime it just
>>> hangs and does not perform at all. I only have builtin audio
>>> at the moment, so it's not as if I am trying to use a
>>> different
>>> soundcard.  That's either with or without -odac in the
>>> options.
>>> 
>>> I am wondering whether this is because you have built it on.
>>> OSX 10.4. You don't use either of the two RT IO modules,
>>> do you? (but it actually reports that you are using
>>> portaudio,
>>> which I didn't think you were since it does not have the
>>> limitation
>>> of only using builtin audio).  Perhaps the program, built on
>>> 10.4
>>> will not run on 10.3. What do you reckon?
>>> 
>>> Victor
>
> -- 
> Send bugs reports to this list.
> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>

On Sun, 12 Feb 2006, Istvan Varga wrote:

> On Sunday 12 February 2006 04:01, Matt J. Ingalls wrote:
>
>> but i do not want to use any rtmodule -- i am using my own callbacks!??
>
> Then set the callbacks between csoundPreCompile() and csoundCompile(),
> and append -+rtaudio=null to the command line options passed to

ok thanks. i think requiring rtaudio=null is the change.  i guess now 
portaudio is set by default?  would be nice to automatically turn 
rtaudio=null when you register callbacks, but i can live with this :)

-m