Re: [Csnd-dev] [EXTERNAL] [Csnd-dev] pfields > 2000

Eduardo Moguillansky <[email protected]>
Newsgroups gmane.comp.audio.csound.devel
Message-ID <CAHFFssKmNB54JZYcQ4aj_Caz6qis7sNcraT2uu1F=8tqv0b1Wg@mail.gmail.com>
The fact that this went so long unnoticed is an indication that long events
are probably not the norm. In that case, it might be practical to just keep
the current design. Probably also the array version of the "ftgen" opcode
needs to be looked after, since it also uses the EVTBLK struct.



On Thu, May 8, 2025 at 10:17 PM Victor Lazzarini <
[email protected]> wrote:

> yes, I realised the schedule opcode needs to be able to access the extra
> field.
>
> I am considering now whether we may just go for a fully dynamic memory
> arrangement and dispense with the extra field. It may
> be the best solution now, but it’s a lot more involved.
> ========================
> Prof. Victor Lazzarini
> Maynooth University
> Ireland
>
> > On 8 May 2025, at 21:11, Eduardo Moguillansky <
> [email protected]> wrote:
> >
> > The schedule opcode itself also needs changes, that was not part of the
> original PR: There is also some wasteful copying of memory, where an EVTBLK
> is created and filled with an .extra chunk only to dispose the memory right
> after that.
> >
> > On Thu, May 8, 2025 at 8:37 PM Victor Lazzarini <
> [email protected]> wrote:
> > No need, I have all I require there. Leave it with me, there's some
> internal changes that I may need to look after (6->7)
> >
> > Thanks
> > Prof. Victor Lazzarini
> > Maynooth University
> > Ireland
> >
> >> On 8 May 2025, at 17:53, Eduardo Moguillansky <
> [email protected]> wrote:
> >>
> >>  I can post a PR later today
> >>
> >> On Thu, May 8, 2025 at 1:22 PM Eduardo Moguillansky <
> [email protected]> wrote:
> >> I think this might be it:
> >>
> >> https://github.com/csound/csound/pull/1764
> >>
> >> On Thu, May 8, 2025 at 12:59 PM Victor Lazzarini <
> [email protected]> wrote:
> >> Do you have the PR number?
> >> Prof. Victor Lazzarini
> >> Maynooth University
> >> Ireland
> >>
> >>> On 8 May 2025, at 11:33, Eduardo Moguillansky <
> [email protected]> wrote:
> >>>
> >>>  *Warning*
> >>> This email originated from outside of Maynooth University's Mail
> System. Do not reply, click links or open attachments unless you recognise
> the sender and know the content is safe.
> >>> I had submitted a PR long ago which fixed this problem, but it seems
> to not have been applied to csound 7. Whenever the number of pfields
> exceeds 2000 csound crashes.
> >>>
> >>> I don't have the code anymore
> >>>
> >>> Test code:
> >>>
> >>> <CsoundSynthesizer>
> >>> <CsOptions>
> >>> -odac
> >>> </CsOptions>
> >>> <CsInstruments>
> >>>
> >>> sr = 44100
> >>> ksmps = 64
> >>> nchnls = 2
> >>> 0dbfs = 1
> >>>
> >>> instr 10
> >>> ipfields[] passign 4
> >>> printarray ipfields
> >>> turnoff
> >>> endin
> >>>
> >>> instr 20
> >>> ipfields[] genarray 0, 2000
> >>> ipfields[0] = 10
> >>> ipfields[1] = 0
> >>> ipfields[2] = 1
> >>> schedule ipfields
> >>> turnoff
> >>> endin
> >>>
> >>> </CsInstruments>
> >>> <CsScore>
> >>>
> >>> i 20 0 1
> >>>
> >>> </CsScore>
> >>> </CsoundSynthesizer>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
>
>
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.