I've been using the csoundapi~ object in Pure Data from 
http://www.nuim.ie/academic/music/musictec/csound/csoundapi.zip (the 
4.23 version)
...on three separate WinXP machines and they all stop generating sound 
after one hour. (The patch can still make sound but the csound part is 
just giving residual whine from an unempty buffer.) One of the machines 
(for unknown reasons, they are all set up identically as far as I can 
tell) is printing the score progress and always at T3600 the score is 
still being processed but the time stops advancing. Is this because I 
didn't pay my registration fee (just kidding)?
I've seen some mention of a dummy note that plays this long for realtime 
but I don't have it in my score and it isn't in the csoundapi~ source code.
Can it be changed? I like to make sound installations so obviously this 
is no good at all for me.
Restarting the score or even closing the pd patch and reopening it both 
crash pd.
Is this changed in csound5?

Thanks for any help...
Martin


-- 
Send bugs reports to this list.
To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Sorry for the delay in replying. I haven't tested that myself, but from your
description, it looks like a bug in the csound lib 4.23, as the PD code
itself seems to still be running (with the same looped buffer). I have to 
try it
in csound 5 to see if it's the same.

Victor

At 23:51 22/09/2005, you wrote:
>I've been using the csoundapi~ object in Pure Data from 
>http://www.nuim.ie/academic/music/musictec/csound/csoundapi.zip (the 4.23 
>version)
>...on three separate WinXP machines and they all stop generating sound 
>after one hour. (The patch can still make sound but the csound part is 
>just giving residual whine from an unempty buffer.) One of the machines 
>(for unknown reasons, they are all set up identically as far as I can 
>tell) is printing the score progress and always at T3600 the score is 
>still being processed but the time stops advancing. Is this because I 
>didn't pay my registration fee (just kidding)?
>I've seen some mention of a dummy note that plays this long for realtime 
>but I don't have it in my score and it isn't in the csoundapi~ source code.
>Can it be changed? I like to make sound installations so obviously this is 
>no good at all for me.
>Restarting the score or even closing the pd patch and reopening it both 
>crash pd.
>Is this changed in csound5?
>
>Thanks for any help...
>Martin
>
>
>--
>Send bugs reports to this list.
>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 

-- 
Send bugs reports to this list.
To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

I have been running csoundapi~ with libcsound 5 for the past three hours
and it's still working, no timeout detected.

So it's probably a bug with 4.23

Victor

At 09:28 27/09/2005, you wrote:
>Sorry for the delay in replying. I haven't tested that myself, but from your
>description, it looks like a bug in the csound lib 4.23, as the PD code
>itself seems to still be running (with the same looped buffer). I have to 
>try it
>in csound 5 to see if it's the same.
>
>Victor
>
>At 23:51 22/09/2005, you wrote:
>>I've been using the csoundapi~ object in Pure Data from 
>>http://www.nuim.ie/academic/music/musictec/csound/csoundapi.zip (the 4.23 
>>version)
>>...on three separate WinXP machines and they all stop generating sound 
>>after one hour. (The patch can still make sound but the csound part is 
>>just giving residual whine from an unempty buffer.) One of the machines 
>>(for unknown reasons, they are all set up identically as far as I can 
>>tell) is printing the score progress and always at T3600 the score is 
>>still being processed but the time stops advancing. Is this because I 
>>didn't pay my registration fee (just kidding)?
>>I've seen some mention of a dummy note that plays this long for realtime 
>>but I don't have it in my score and it isn't in the csoundapi~ source code.
>>Can it be changed? I like to make sound installations so obviously this 
>>is no good at all for me.
>>Restarting the score or even closing the pd patch and reopening it both 
>>crash pd.
>>Is this changed in csound5?
>>
>>Thanks for any help...
>>Martin
>>
>>
>>--
>>Send bugs reports to this list.
>>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>
>Victor Lazzarini
>Music Technology Laboratory
>Music Department
>National University of Ireland, Maynooth
>--
>Send bugs reports to this list.
>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 

-- 
Send bugs reports to this list.
To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

I guess I need to use the csound5 version. What components are needed to 
run csoundapi~ besides the dll?
I've tried a few times with 
http://www.nuim.ie/academic/music/musictec/csound/csoundapi~5.zip
First I got a message about missing libsndfile.dll, which I found a copy 
of in another csound5 distibution, but then pd just outright crashes 
when csoundapi~ is created.
So I'm confused about the setup. Do I need an actual copy of csound5 as 
well as the csoundapi~? All the windows-compiled versions of csound5 I 
have come across don't include libcsound, for instance.

Martin


Victor Lazzarini wrote:

> I have been running csoundapi~ with libcsound 5 for the past three hours
> and it's still working, no timeout detected.
>
> So it's probably a bug with 4.23
>
> Victor
>
> At 09:28 27/09/2005, you wrote:
>
>> Sorry for the delay in replying. I haven't tested that myself, but 
>> from your
>> description, it looks like a bug in the csound lib 4.23, as the PD code
>> itself seems to still be running (with the same looped buffer). I 
>> have to try it
>> in csound 5 to see if it's the same.
>>
>> Victor
>>
>> At 23:51 22/09/2005, you wrote:
>>
>>> I've been using the csoundapi~ object in Pure Data from 
>>> http://www.nuim.ie/academic/music/musictec/csound/csoundapi.zip (the 
>>> 4.23 version)
>>> ...on three separate WinXP machines and they all stop generating 
>>> sound after one hour. (The patch can still make sound but the csound 
>>> part is just giving residual whine from an unempty buffer.) One of 
>>> the machines (for unknown reasons, they are all set up identically 
>>> as far as I can tell) is printing the score progress and always at 
>>> T3600 the score is still being processed but the time stops 
>>> advancing. Is this because I didn't pay my registration fee (just 
>>> kidding)?
>>> I've seen some mention of a dummy note that plays this long for 
>>> realtime but I don't have it in my score and it isn't in the 
>>> csoundapi~ source code.
>>> Can it be changed? I like to make sound installations so obviously 
>>> this is no good at all for me.
>>> Restarting the score or even closing the pd patch and reopening it 
>>> both crash pd.
>>> Is this changed in csound5?
>>>
>>> Thanks for any help...
>>> Martin
>>>
>>>
>>> -- 
>>> Send bugs reports to this list.
>>> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>>
>>
>> Victor Lazzarini
>> Music Technology Laboratory
>> Music Department
>> National University of Ireland, Maynooth
>> -- 
>> Send bugs reports to this list.
>> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>
>
> Victor Lazzarini
> Music Technology Laboratory
> Music Department
> National University of Ireland, Maynooth



-- 
Send bugs reports to this list.
To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

At this stage, you might need to build your own from CVS.
The reason for the crash might be the version of PD you are
using (Michael Gogin has reported something of the kind), but
I think the latest sources have included a fix by Michael.

Istvan's binary distribution might not include csoundapi~.dll, but
if they do, that's your best bet for a ready-to-go system.

Victor

At 23:52 28/09/2005, you wrote:
>I guess I need to use the csound5 version. What components are needed to 
>run csoundapi~ besides the dll?
>I've tried a few times with 
>http://www.nuim.ie/academic/music/musictec/csound/csoundapi~5.zip
>First I got a message about missing libsndfile.dll, which I found a copy 
>of in another csound5 distibution, but then pd just outright crashes when 
>csoundapi~ is created.
>So I'm confused about the setup. Do I need an actual copy of csound5 as 
>well as the csoundapi~? All the windows-compiled versions of csound5 I 
>have come across don't include libcsound, for instance.
>
>Martin
>
>
>Victor Lazzarini wrote:
>
>>I have been running csoundapi~ with libcsound 5 for the past three hours
>>and it's still working, no timeout detected.
>>
>>So it's probably a bug with 4.23
>>
>>Victor
>>
>>At 09:28 27/09/2005, you wrote:
>>
>>>Sorry for the delay in replying. I haven't tested that myself, but from your
>>>description, it looks like a bug in the csound lib 4.23, as the PD code
>>>itself seems to still be running (with the same looped buffer). I have 
>>>to try it
>>>in csound 5 to see if it's the same.
>>>
>>>Victor
>>>
>>>At 23:51 22/09/2005, you wrote:
>>>
>>>>I've been using the csoundapi~ object in Pure Data from 
>>>>http://www.nuim.ie/academic/music/musictec/csound/csoundapi.zip (the 
>>>>4.23 version)
>>>>...on three separate WinXP machines and they all stop generating sound 
>>>>after one hour. (The patch can still make sound but the csound part is 
>>>>just giving residual whine from an unempty buffer.) One of the machines 
>>>>(for unknown reasons, they are all set up identically as far as I can 
>>>>tell) is printing the score progress and always at T3600 the score is 
>>>>still being processed but the time stops advancing. Is this because I 
>>>>didn't pay my registration fee (just kidding)?
>>>>I've seen some mention of a dummy note that plays this long for 
>>>>realtime but I don't have it in my score and it isn't in the csoundapi~ 
>>>>source code.
>>>>Can it be changed? I like to make sound installations so obviously this 
>>>>is no good at all for me.
>>>>Restarting the score or even closing the pd patch and reopening it both 
>>>>crash pd.
>>>>Is this changed in csound5?
>>>>
>>>>Thanks for any help...
>>>>Martin
>>>>
>>>>
>>>>--
>>>>Send bugs reports to this list.
>>>>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>>>
>>>
>>>Victor Lazzarini
>>>Music Technology Laboratory
>>>Music Department
>>>National University of Ireland, Maynooth
>>>--
>>>Send bugs reports to this list.
>>>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>>
>>
>>Victor Lazzarini
>>Music Technology Laboratory
>>Music Department
>>National University of Ireland, Maynooth
>
>
>
>--
>Send bugs reports to this list.
>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth