I have no objection, as my impression is that nobody was actually using OSC in files that anybody would have shared or used for teaching.

Regards,
Mike

-----Original Message-----
From: jpff@cs.bath.ac.uk
Sent: May 19, 2005 8:22 AM
To: csound-devel@lists.sourceforge.net
Subject: [Cs-dev] OSC on CS5

The second model is now working.  I think it is much better and
cleaner, but it is not compatible with the previous version(s).
 the syntax is
    OSCsendl    kwhen, Shost, iPort, Sdest, Stype, xxx....

When kwhen changes in a kcycle an OSC message is send to the host
Shost (empty string means this machine), to port iPort with
destination address Sdest.  The type string is made from the
characters ifdsc (I may add others) which treat the corresponding
argument as an 32-bit integer, float, double, string or character.
Maximum of 25 arguments in a single call.

The incompatibility is that the type argument is a string when it was
an integer.

Are there any objections to renaming this newer version OSCsend and
deleting the earlier attempts?

==John ffitch


-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel





-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I second that.
Oeyvind


> From: Michael Gogins [gogins@pipeline.com]
> 
> I have no objection, as my impression is that nobody was actually using OSC in files that anybody would have shared or used for teaching.
> 
> Regards,
> Mike
> 
> -----Original Message-----
> From: jpff@cs.bath.ac.uk
> 
> Are there any objections to renaming this newer version OSCsend and
> deleting the earlier attempts?
> 
> ==John ffitch

I added the AltiVec flags to gcc as a Scons option: useAltivec=1 (defaults 
to 0).
Tried it, but I'm not sure it made much difference.

Any suggestions for more optimisation?

Victor


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



-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

On May 20, 2005, at 4:32 AM, Victor Lazzarini wrote:

> I added the AltiVec flags to gcc as a Scons option: useAltivec=1 
> (defaults to 0).

doesn't that just allow you to call altivec routines?

are your performance issues during file rendering or real-time?
and you are sure it is a general, "system-wide" problem?
[ what comes to mind to me is, when checking out poor performance in
  richard's initial pvs opcodes, i made a *huge*
speed improvement in the pvoc-ex file i/o adding a buffer because
because the unix file routines didn't buffer themselves on osx ==
i added it to csound4 but not to cs5 yet because it uses some global
variables and wanted to wait to do it right ]

you are using the api, right?  any difference between calling
performksmps() and performbuffer()?

-m

> Tried it, but I'm not sure it made much difference.
>
> Any suggestions for more optimisation?
>
> Victor
>
>
> Victor Lazzarini
> Music Technology Laboratory
> Music Department
> National University of Ireland, Maynooth
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Oh, no this is the csound command, which of course uses the API, not csoundapi~
that I am testing mainly.

Csoundapi~ is also slower than on Linux. But also made worse
because PD uses portaudio.

I'm mainly talking about real-time; I haven't tested much file rendering.

I added the altivec stuff because Anthony Kozar suggested trying it. But
I don't know much about it.

Victor
At 17:04 20/05/2005, you wrote:

>On May 20, 2005, at 4:32 AM, Victor Lazzarini wrote:
>
>>I added the AltiVec flags to gcc as a Scons option: useAltivec=1 
>>(defaults to 0).
>
>doesn't that just allow you to call altivec routines?
>
>are your performance issues during file rendering or real-time?
>and you are sure it is a general, "system-wide" problem?
>[ what comes to mind to me is, when checking out poor performance in
>  richard's initial pvs opcodes, i made a *huge*
>speed improvement in the pvoc-ex file i/o adding a buffer because
>because the unix file routines didn't buffer themselves on osx ==
>i added it to csound4 but not to cs5 yet because it uses some global
>variables and wanted to wait to do it right ]
>
>you are using the api, right?  any difference between calling
>performksmps() and performbuffer()?
>
>-m
>
>>Tried it, but I'm not sure it made much difference.
>>
>>Any suggestions for more optimisation?
>>
>>Victor
>>
>>
>>Victor Lazzarini
>>Music Technology Laboratory
>>Music Department
>>National University of Ireland, Maynooth
>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by Oracle Space Sweepstakes
>>Want to be the first software developer in space?
>>Enter now for the Oracle Space Sweepstakes!
>>http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
>>_______________________________________________
>>Csound-devel mailing list
>>Csound-devel@lists.sourceforge.net
>>https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by Oracle Space Sweepstakes
>Want to be the first software developer in space?
>Enter now for the Oracle Space Sweepstakes!
>http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
>_______________________________________________
>Csound-devel mailing list
>Csound-devel@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/csound-devel

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



-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

matt wrote:

> [ what comes to mind to me is, when checking out poor performance in
>  richard's initial pvs opcodes, i made a *huge*
> speed improvement in the pvoc-ex file i/o adding a buffer because
> because the unix file routines didn't buffer themselves on osx ==

Is there any particular reason for not using the (buffered) ANSI C
functions for file I/O instead ?


-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

On 5/20/05 12:17 PM, Victor Lazzarini  etched in
stone:

> I added the altivec stuff because Anthony Kozar suggested trying it. But
> I don't know much about it.

Neither do I !  I just know that it is supposed to be good for
single-precision floats.  I've also heard that it might require changing
code, but I thought there was a possibility that GCC might be able to
convert some loop operations into vector instructions.  I am really just
guessing here.

On 5/20/05 7:32 AM, Victor Lazzarini  etched in
stone:

> I added the AltiVec flags to gcc as a Scons option: useAltivec=1 (defaults
> to 0).  Tried it, but I'm not sure it made much difference.

Sorry, but I did not see this before sending my post to the other list about
not making Altivec the default.  This sounds good.  Thanks!

Anthony



-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net