Wow, I had missed the introduction of these opcodes.
Very useful, Thanks Steven.
Oeyvind

2008/1/30, Steven Yi <stevenyi@gmail.com>:
> Hi All,
>
> Not sure if this is exactly related, but if you're working on a sample
> by sample basis, you may want to consider either working ksmps=1 or
> using the vaset and vaget opcodes, which will allow reading and
> writing on a per-sample basis of an a-rate variable.  The va- opcodes
> essentially allow you to write csound code in the same exact way you
> would if you were writing it in C, where you would be working with a
> vector of audio data, and were originally written to help port C audio
> code to pure csound code.
>
> Anyways, just wanted to mention those.
>
> Thanks!
> steven
>
>
>
> On Jan 29, 2008 8:25 PM, Brian Wong <brian.wong1@hotmail.com> wrote:
> >
> > Actually my ftsave question is about trying to do a very similar process,
> > except in my case I am trying to create wave files with a specific (small)
> > number of samples in the single-cycle waveform, to load into a synthesizer.
> > Any advice would be appreciated!
> >
> >
> >
> >  ________________________________
> >  > Date: Tue, 29 Jan 2008 17:13:57 -0500
> > > From: mailing-lists-1001@anthonykozar.net
> >
> > > To: csound@lists.bath.ac.uk
> > > Subject: [Csnd] Re: Re: generating single cycle waveforms to an audio file
> >
> >
> > >
> > > Csound note events always last a multiple of ksmps samples though. So, you
> > > have to be careful no matter which opcode you use to generate the sound.
> > > Settings ksmps = 1 is probably necessary to get the best results.
> > >
> > > This question is very similar to the ftsave/WAV question from the other
> > day.
> > > The other issue of course is that Csound's idea of a single-cycle waveform
> > > in an ftable is independent of pitch. The precision attainable when
> > reading
> > > such a waveform in a sampler will depend on what frequency you output it
> > at.
> > > I suggest that you may want to start with a Csound table size of 65536 and
> > a
> > > 1 Hz output frequency using an interpolating opcode (oscil3 or oscil1i
> > could
> > > work since kr = sr). That would give you 44100 samples in the WAV file
> > > (assuming 44.1 kHz sampling rate) for one cycle. You may even want more
> > > than that.
> > >
> > > Anthony Kozar
> > > mailing-lists-1001 AT anthonykozar DOT net
> > > http://anthonykozar.net/
> > >
> > >
> > > Rory Walsh wrote on 1/29/08 4:30 PM:
> > >
> > > > Perhaps I misunderstood but if you know how long one cycle is then just
> > > > run the instrument for one cycle. You could also catch a cycle during
> > > > performance using a conditional test with fout.
> > >
> > > > P R Figueiredo wrote:
> > > >> Hello,
> > > >>
> > > >> Is there any simple way of creating a single cycle waveform (from any
> > > >> table I could use with an oscillator) and save it to a .wav file to use
> > > >> in a sampler, for instance? Obviously I can use an oscillator to play
> > > >> the waveform, render to a file, and then edit to a single cycle. But I
> > > >> wonder if it's possible to generate a single cycle, and save me all the
> > > >> editing.
> > >
> > >
> > >
> > > Send bugs reports to this list.
> > > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> > csound"
> >
> > ________________________________
> >
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>