Message for the May 2020


This is just a regular monthly reminder for the Csound list.

The Csound list welcomes posts from people with ALL levels of skill,
from the newest newbie to the most serious hacker or established
composer.  The subject and tenor of the posts varies dramatically
depending on what the current concerns are.  Newbies are sometimes
afraid to post because they read discussions about the
incomprehensible deep inner workings, and all they want to know is how
to get a sound to come out of their computer, or advice on how to get
mobile sounds.  Rest assured that your questions will be answered
quickly, and (usually) in a helpful and courteous manner.  We have all
been there.  Please post.

If you find a bug or want to request for a new feature for the csound
library please report it here and/or open a ticket by going to

    https://github.com/csound/csound/issues

Please note that front-ends and other csound-based systems (e.g blue,
CsoundQt, Cabbage) have their own reporting schemes.

The developers particularly love bug reports that say which platform,
operating system and version (we are currently maintaining 6.14 while
developing csound7) were involved and a simple test program.

If you are thinking of submitting new code please read the Submission
advice (Submitting_opcodes.md) and/or consult a core developer.

==List Administrator

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Culd you provie an exampleinut that would fail?  I cannot see te prolem 
yet o please enlighten me.

==John ff

On Mon, 11 May 2020, David Bowen wrote:

> I was looking at the code for cps2pch as a model for a possible oct2pch opcode
> and noticed what I think is a bug in the handling of the frequency table
> option.The relevant code is:
> 
> int32_t cps2pch(CSOUND *csound, XENH *p)
> {
>     double  fract;
>     double  loct;
> 
>     fract = modf((double)*p->pc, &loct);        /* Get octave */
>     if (*p->et > 0) {
>       fract = pow(2.0, loct + (100.0*fract)/((double)*p->et));
>       *p->r = (MYFLT)(fract * 1.02197503906); /* Refer to base frequency */
>     }
>     else {
>       MYFLT t = - *p->et;
>       FUNC* ftp = csound->FTnp2Finde(csound, &t);
>       int32_t len;
>       if (UNLIKELY(ftp == NULL))
>         return csound->PerfError(csound, &(p->h),Str("No tuning table %d"),
>                                  -((int32_t)*p->et));
> int32_t cps2pch(CSOUND *csound, XENH *p)
> {
>     double  fract;
>     double  loct;
> 
>     fract = modf((double)*p->pc, &loct);        /* Get octave */
>     if (*p->et > 0) {
>       fract = pow(2.0, loct + (100.0*fract)/((double)*p->et));
>       *p->r = (MYFLT)(fract * 1.02197503906); /* Refer to base frequency */
>     }
>     else {
>       MYFLT t = - *p->et;
>       FUNC* ftp = csound->FTnp2Finde(csound, &t);
>       int32_t len;
>       if (UNLIKELY(ftp == NULL))
>         return csound->PerfError(csound, &(p->h),Str("No tuning table %d"),
>                                  -((int32_t)*p->et));
>       len = ftp->flen;
>       while (fract>len) {
>         fract -= len; loct++;
>       }
>       fract += 0.005;
>       *p->r = (MYFLT)(1.02197503906 * *(ftp->ftable +(int32_t)(100.0*fract)) *
>                       pow(2.0, loct));
>     }
> 
>     /*  double ref = 261.62561 / pow(2.0, 8.0); */
>     return OK;
> }    }
> 
>     /*  double ref = 261.62561 / pow(2.0, 8.0); */
>     return OK;
> }
> 
> I believe the intent of the while loop is to handle the case where the pitch
> class is larger than the number of elements in the frequency table by
> incrementing the octave number and reducing the pitch class until the class is
> in the table range. However, fract is still a number between 0 and 1, so the
> while will always short circuit and then csound will access outside the table
> range. I think the corrected code should look something like this:
> 
>       len = ftp->flen;
>       fract *= 100.0;
>       while (fract>len) {
>         fract -= len; loct++;
>       }
>       fract += 0.5;
>       *p->r = (MYFLT)(1.02197503906 * *(ftp->ftable +(int32_t)fract) *
>                       pow(2.0, loct));
>     
> David Bowen
> 
> Csound mailing list Csound@listserv.heanet.ie
> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to
> https://github.com/csound/csound/issues Discussions of bugs and features can
> be posted here
>

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

I would expect to see zero i te second print as the table i 16 long anf 
you look at the 12 value which is empty; i also note the same problem 
exists in cpsxpch

If I chage yor 16 fr 10 it works in my version.

Will commit soon.

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here