No, in all the tests I have heard between Csound and PD, csound would be
miles more efficient, as far as DSP is concerned. I am not sure about SC,
because it does not run on all platforms that Csound supports (only half
runs on Windows). The FLTK widgets performance varies from system
to system, but on Windows is not more CPU intensive than PD's GUI.
 
Victor

----- Original Message -----
From: Peiman Khosravi <peimankhosravi@gmail.com>
Date: Thursday, December 6, 2007 12:22 am
Subject: [Csnd] Re: Re: Re: teaching csound: textbook, PD, etc
To: csound@lists.bath.ac.uk

>


> A problem I used to have with an older machine was that adding FLTK gui increased cpu and resulted in glitches. The same I would say applies to some FFT instruments I use (on my older ppc which was quite fast for it's time!). In my experience csound sucks more processing power than max or supercollider, particularly
> with an added GUI. Of course there is always the option of using
> external midi controllers without any GUI I suppose. 

Having said these I am dying to read an actual survey on this. Say using the same DSP processing types on different platforms, and with different languages (csound, sc3, maxmsp, pd, bidule) and comparing the cpu usage and realtime performance. Maybe someone with access to all these systems could do an article for the csound journal? or it could be a collective work.


Just some ideas :-)


Peiman
>

PS I have started using bidule recently as well as csound and I would imagine it's a perfect tool for teaching, specially the fft handling is phenomenally easy and accessible with surprisingly good sound quality. 

http://www.plogue.com/


The demo version is free and fully functional but expires after a few month (should last for a whole term!).
On that note bidule can also read and synthesis pvoc files made with csound! 


On 5 Dec 2007, at 23:56, Chuckk Hubbard wrote:

On Dec 6, 2007 12:59 AM, barry threw <bthrew@gmail.com> wrote:
i also see that he says:

>
"Csound is better adapted
than Pd for batch processing and it handles polyphony much better than
Pd does. On the other hand, Pd has a better developed real-time
control
structure than Csound."

>
is that last statement really true (anymore)?

>

>

>
If better means "easier to figure out how to do" then yes, it is
still true.

>
I would also say easier to implement.  I'm exploring other
possibilities than Pd now, but it is still far quicker to hit
Ctl-Shift-V and place a vertical slider than it is to type all the
necessary FLTK commands, or Tk.  Do you want to teach your students
how to write FLTK code to construct sliders, or how to manipulate
sound?
I have always liked Pd too for its data structures and arbitrary
drawing of elements, but honestly that's pretty hard to do for
anything practical.  Probably just as simple to use Tcl/Tk in that
case.

>
Pd is useful for showing how stuff works for 2 simple facts: you can
see the entire program flow visually, and anything useful you do has
to be assembled from the ground up.  I tend to prefer Csound even for
low-level stuff just because it's quicker to type than to mouse; but
as far as others understanding what you're doing, there's something to
be said for graphical connections.  The algorithms in Csound's
higher-level opcodes are completely hidden from those not in the know.

>
I'm not convinced Csound is the best educational tool for all
occasions, but for certain subjects and certain students, definitely.
I tend to think people who are turned off by sine wave equations will
be turned off by code too.  Non-csounders who look over my shoulder
while I'm using Csound usually have no idea or interest in what I'm
doing.

>
But if I were working with pretty motivated and intelligent people,
I'd push them towards Csound.
How important is real-time control to your class?

>
-Chuckk

>
-- 
http://www.badmuthahubbard.com

>

>
Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"


>