Re: [PD] Status of PD double precision

[Jonathan Wilkes via Pd-list]

> Though Pd 0.49 compiled as above shows those small and large numbers,
it can't show all significant decimals in the GUI. This of course
limits the usefulness of double precision and also complicates the
interpretation of other tests.

Hi Katja,
Have you taken a look at Purr Data? AFAICT the number precision is correctly 
displayed in the GUI for the double precision builds.
If you want to play around with it you can compile using the toplevel makefile 
with `make light`. This will give you a deb or tarball with which you can 
install 
the basic Purr Data (core, pddp, and some others). It's about a 5 minute 
process 
on my Chromebook.

One outstanding question that came up was wrt table sizes for sqrt~ when 
compiling with PDFLOATSIZE=64:
http://disis.music.vt.edu/pipermail/l2ork-dev/2018-June/001985.html
Atm the same table size is used for both floatsizes.
Best,
Jonathan
> Katja  ___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[katja]

On 12/17/18, Thomas Grill  wrote:

> No time to check deeper now, but to my surprise
> CFLAGS "-D PD_FLOATSIZE=64" ./configure && make
> does indeed compile from the vanilla git repo (but doesn't link with
> portmidi and portaudio for other reasons).
> The definitions in m_pd.h suggest that t_float becomes double...


Successful compilation with 64 bit floats is a first step but doesn't
tell so much in itself. By the way it is better to override CPPFLAGS,
not CFLAGS (which holds important flags):

$ make CPPFLAGS="-D PD_FLOATSIZE=64"

This builds Pd with 64 bit floats indeed. The question is, to what
extent does it work? This page provides a .zip with some test patches
that illustrate hotspots of precision problems:

https://katjaas.nl/doubleprecision/doubleprecision.html

The patches were made in 2011 to test pd-double but they are still
useful. For example patch "03precision-tests/limits.pd" gives proof of
double precision when showing very small and large numbers that single
precision can't do.

Though Pd 0.49 compiled as above shows those small and large numbers,
it can't show all significant decimals in the GUI. This of course
limits the usefulness of double precision and also complicates the
interpretation of other tests.

Katja



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[IOhannes m zmölnig]

On 12/16/18 1:16 PM, Giulio Moro via Pd-list wrote:
> It's been done this past summer for Purr-data/Pd-l2ork. Here is the report: 
> http://disis.music.vt.edu/pipermail/l2ork-dev/2018-August/002019.html


i did a check of the changes that went into purr-data, and afaict some
of them are already in Pd-vanilla, and the rest is lingering on my
harddisk waiting to be cleaned up.
it seems that purr-data isn't really much ahead of Pd-vanilla in this
respect.

there's still some open questions which i'd like to resolve before doing
a PR (@katja: did you get my mail? :-))


anyhow, at least for *me*¹, double precision is on the roadmap for the
the next Pd-release(s).
of course, i have no idea what will eventually end up in the final releases.

gf,mads
IOhannes



¹ https://github.com/pure-data/pure-data/projects



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[Miller Puckette]

Katja Vetter has tried this and compiled a list of changes needed to make it
work.  It compiles fine, but apprently quite a few of the objects have trouble
functioning correctly.

cheers
Miller

On Mon, Dec 17, 2018 at 11:48:15PM +0100, Thomas Grill wrote:
> > Am 16.12.2018 um 18:41 schrieb Alexandre Torres Porres :
> > Am I crazy or did I hear some talk about 64bit for pd vanilla soon, maybe 
> > as early as 0.50-0?
> 
> No time to check deeper now, but to my surprise
> CFLAGS "-D PD_FLOATSIZE=64" ./configure && make
> does indeed compile from the vanilla git repo (but doesn't link with portmidi 
> and portaudio for other reasons).
> The definitions in m_pd.h suggest that t_float becomes double...
> 
> best, Thomas
> 
> --
> Thomas Grill
> http://g.org
> 
> 



> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list




___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[Thomas Grill]

> Am 16.12.2018 um 18:41 schrieb Alexandre Torres Porres :
> Am I crazy or did I hear some talk about 64bit for pd vanilla soon, maybe as 
> early as 0.50-0?

No time to check deeper now, but to my surprise
CFLAGS "-D PD_FLOATSIZE=64" ./configure && make
does indeed compile from the vanilla git repo (but doesn't link with portmidi 
and portaudio for other reasons).
The definitions in m_pd.h suggest that t_float becomes double...

best, Thomas

--
Thomas Grill
http://g.org




signature.asc
Description: Message signed with OpenPGP using GPGMail
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[Dario Sanfilippo]

I've just tested this release:
https://github.com/jonwwilkes/purr-data/releases/download/2.6.0/purr-double-trouble-test1-osx_10.11-x86_64-dmg.zip
.

I've added 1 to 1e10 and then subtracted 1e10 expecting to get 1, but I get
0, so it doesn't seem double. Do you get the same behaviour?

D




On Sun, 16 Dec 2018 at 13:16, Giulio Moro  wrote:

> It's been done this past summer for Purr-data/Pd-l2ork. Here is the
> report:
> http://disis.music.vt.edu/pipermail/l2ork-dev/2018-August/002019.html
>
> Best,
> Giulio
>
> On Sunday, 16 December 2018, 11:29:25 GMT, Dario Sanfilippo <
> sanfilippo.da...@gmail.com> wrote:
>
> Hello, list.
>
> I'm just curious to know if anybody is actively working on a double
> precision version of PD. Please let me know if I can help in any way.
>
> Best,
> Dario
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management ->
> https://lists.puredata.info/listinfo/pd-list
>
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[Alexandre Torres Porres]

Am I crazy or did I hear some talk about 64bit for pd vanilla soon, maybe
as early as 0.50-0?

cheers

Em dom, 16 de dez de 2018 às 11:37, hans w. koch 
escreveu:

> and then there is also spaghettis:
> https://github.com/Spaghettis/Spaghettis
> which is double prec. but not as encompassing as purr data.
>
> best hans
>
> > Am 16.12.2018 um 13:16 schrieb Giulio Moro via Pd-list <
> pd-list@lists.iem.at>:
> >
> > It's been done this past summer for Purr-data/Pd-l2ork. Here is the
> report:
> http://disis.music.vt.edu/pipermail/l2ork-dev/2018-August/002019.html
> >
> > Best,
> > Giulio
> >
> > On Sunday, 16 December 2018, 11:29:25 GMT, Dario Sanfilippo <
> sanfilippo.da...@gmail.com> wrote:
> >
> > Hello, list.
> >
> > I'm just curious to know if anybody is actively working on a double
> precision version of PD. Please let me know if I can help in any way.
> >
> > Best,
> > Dario
> > ___
> > Pd-list@lists.iem.at mailing list
> > UNSUBSCRIBE and account-management ->
> https://lists.puredata.info/listinfo/pd-list
> >
> >
> >
> > ___
> > Pd-list@lists.iem.at mailing list
> > UNSUBSCRIBE and account-management ->
> https://lists.puredata.info/listinfo/pd-list
>
>
>
>
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management ->
> https://lists.puredata.info/listinfo/pd-list
>
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[hans w. koch]

and then there is also spaghettis: https://github.com/Spaghettis/Spaghettis
which is double prec. but not as encompassing as purr data.

best hans

> Am 16.12.2018 um 13:16 schrieb Giulio Moro via Pd-list :
> 
> It's been done this past summer for Purr-data/Pd-l2ork. Here is the report: 
> http://disis.music.vt.edu/pipermail/l2ork-dev/2018-August/002019.html
> 
> Best,
> Giulio
> 
> On Sunday, 16 December 2018, 11:29:25 GMT, Dario Sanfilippo 
>  wrote: 
> 
> Hello, list.
> 
> I'm just curious to know if anybody is actively working on a double precision 
> version of PD. Please let me know if I can help in any way.
> 
> Best,
> Dario
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list
> 
> 
> 
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list




___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Status of PD double precision

[Giulio Moro via Pd-list]

It's been done this past summer for Purr-data/Pd-l2ork. Here is the report: 
http://disis.music.vt.edu/pipermail/l2ork-dev/2018-August/002019.html

Best,
Giulio

On Sunday, 16 December 2018, 11:29:25 GMT, Dario Sanfilippo 
 wrote: 

Hello, list.

I'm just curious to know if anybody is actively working on a double precision 
version of PD. Please let me know if I can help in any way.

Best,
Dario
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list