ntg-context - mailing list for ConTeXt users
 help / color / mirror / Atom feed
* Arithmetic Overflow error in MetaFun
@ 2020-05-16 14:36 Keith McKay
  2020-05-16 14:48 ` Keith McKay
  2020-05-16 14:50 ` Hans Hagen
  0 siblings, 2 replies; 5+ messages in thread
From: Keith McKay @ 2020-05-16 14:36 UTC (permalink / raw)
  To: mailing list for ConTeXt users


[-- Attachment #1.1: Type: text/plain, Size: 2154 bytes --]

Colleagues

I'm having problems with an Arithmetic Overflow error when calculating 
the x and y values for functions of the form:

y = ax^4 + bx^3 + cx^2 + dx + e

to which I then add a bit of randomness and other embellishments. I have 
had a look in the Metafun and Metafont manuals but nothing jumps out at 
me. In the minimum working example below, this Arithmetic Overflow 
occurs when xord =13.5 but I can get the calculation to work if I 
separate out the equation and then add together, since it seems to be 
failing on the pow(xord,4) instruction. Is there a switch in MetaFont 
that should be set or a better way of getting round this Arithmetic 
Overflow error?

Thanks

Best Wishes

Keith McKay

%%%%%%%%%%%%%MWE%%%%%%%%%%%

\setuppapersize [A4,landscape]

\starttext

\startMPpage

StartPage;

width := PaperWidth ; height := PaperHeight ; unit := cm ;

path p;

p := unitsquare scaled .1cm ;

path pat;

pat := (0.0*cm,19.935*cm);

for xord = 0 step 0.1 until 25:

show xord;

%%%%%%%%fails at xord = 13.5%%%%%%%%%

yord := ((-0.000268117) * pow(xord,4)) +(0.0136949 * pow(xord, 3)) + 
((-0.16608) * sqr(xord)) + ((-0.771743) * xord) + 19.935;

%%%%%comment out above and uncomment below and it works%%%%

% yord1 := (-0.000268117) * sqr(xord);

% yord1a := yord1 * sqr(xord);

% yord2 := (0.0136949 * pow(xord, 3));

% yord3 := ((-0.16608) * sqr(xord));

% yord4 := ((-0.771743) * xord);

% yord := yord1 + yord1a + yord2 + yord3 + yord4 + 19.935;

xrand := (uniformdeviate(1) - 0.5)*2;

yrand := uniformdeviate(1) - 0.5;

xcoord := xord + xrand;

ycoord := yord + yrand;

if odd xord:

pat := pat ... (xcoord*cm, ycoord*cm);

else:

pat := pat --- (xcoord*cm, ycoord*cm);

fill p shifted (xcoord*cm, ycoord*cm) withcolor 
(uniformdeviate(1),uniformdeviate(1),uniformdeviate(1)) ;

fi;

draw pat withpen pencircle scaled (uniformdeviate(0.75)*mm) withcolor 
(uniformdeviate(1),uniformdeviate(1),uniformdeviate(1)); 
%withtransparency (1, .5);

drawdot (xcoord*cm, ycoord*cm) withpen pencircle scaled 
(uniformdeviate(2.5)*mm) withcolor 
(uniformdeviate(1),uniformdeviate(1),uniformdeviate(1));

endfor;

StopPage;

\stopMPpage

\stoptext


[-- Attachment #1.2: Type: text/html, Size: 6978 bytes --]

[-- Attachment #2: Type: text/plain, Size: 493 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Arithmetic Overflow error in MetaFun
  2020-05-16 14:36 Arithmetic Overflow error in MetaFun Keith McKay
@ 2020-05-16 14:48 ` Keith McKay
  2020-05-16 19:46   ` Hans Hagen
  2020-05-16 14:50 ` Hans Hagen
  1 sibling, 1 reply; 5+ messages in thread
From: Keith McKay @ 2020-05-16 14:48 UTC (permalink / raw)
  To: mailing list for ConTeXt users


[-- Attachment #1.1: Type: text/plain, Size: 2575 bytes --]

Answering my own problem here.

I've just searched through the mailing list archives and it seems that 
if I add [instance=doublefun] I do not get the Arithmetic Overflow 
error. Joy!

Thanks

Best Wishes

Keith McKay

On 16/05/2020 15:36, Keith McKay wrote:
>
> Colleagues
>
> I'm having problems with an Arithmetic Overflow error when calculating 
> the x and y values for functions of the form:
>
> y = ax^4 + bx^3 + cx^2 + dx + e
>
> to which I then add a bit of randomness and other embellishments. I 
> have had a look in the Metafun and Metafont manuals but nothing jumps 
> out at me. In the minimum working example below, this Arithmetic 
> Overflow occurs when xord =13.5 but I can get the calculation to work 
> if I separate out the equation and then add together, since it seems 
> to be failing on the pow(xord,4) instruction. Is there a switch in 
> MetaFont that should be set or a better way of getting round this 
> Arithmetic Overflow error?
>
> Thanks
>
> Best Wishes
>
> Keith McKay
>
> %%%%%%%%%%%%%MWE%%%%%%%%%%%
>
> \setuppapersize [A4,landscape]
>
> \starttext
>
> \startMPpage
>
> StartPage;
>
> width := PaperWidth ; height := PaperHeight ; unit := cm ;
>
> path p;
>
> p := unitsquare scaled .1cm ;
>
> path pat;
>
> pat := (0.0*cm,19.935*cm);
>
> for xord = 0 step 0.1 until 25:
>
> show xord;
>
> %%%%%%%%fails at xord = 13.5%%%%%%%%%
>
> yord := ((-0.000268117) * pow(xord,4)) +(0.0136949 * pow(xord, 3)) + 
> ((-0.16608) * sqr(xord)) + ((-0.771743) * xord) + 19.935;
>
> %%%%%comment out above and uncomment below and it works%%%%
>
> % yord1 := (-0.000268117) * sqr(xord);
>
> % yord1a := yord1 * sqr(xord);
>
> % yord2 := (0.0136949 * pow(xord, 3));
>
> % yord3 := ((-0.16608) * sqr(xord));
>
> % yord4 := ((-0.771743) * xord);
>
> % yord := yord1 + yord1a + yord2 + yord3 + yord4 + 19.935;
>
> xrand := (uniformdeviate(1) - 0.5)*2;
>
> yrand := uniformdeviate(1) - 0.5;
>
> xcoord := xord + xrand;
>
> ycoord := yord + yrand;
>
> if odd xord:
>
> pat := pat ... (xcoord*cm, ycoord*cm);
>
> else:
>
> pat := pat --- (xcoord*cm, ycoord*cm);
>
> fill p shifted (xcoord*cm, ycoord*cm) withcolor 
> (uniformdeviate(1),uniformdeviate(1),uniformdeviate(1)) ;
>
> fi;
>
> draw pat withpen pencircle scaled (uniformdeviate(0.75)*mm) withcolor 
> (uniformdeviate(1),uniformdeviate(1),uniformdeviate(1)); 
> %withtransparency (1, .5);
>
> drawdot (xcoord*cm, ycoord*cm) withpen pencircle scaled 
> (uniformdeviate(2.5)*mm) withcolor 
> (uniformdeviate(1),uniformdeviate(1),uniformdeviate(1));
>
> endfor;
>
> StopPage;
>
> \stopMPpage
>
> \stoptext
>

[-- Attachment #1.2: Type: text/html, Size: 7609 bytes --]

[-- Attachment #2: Type: text/plain, Size: 493 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Arithmetic Overflow error in MetaFun
  2020-05-16 14:36 Arithmetic Overflow error in MetaFun Keith McKay
  2020-05-16 14:48 ` Keith McKay
@ 2020-05-16 14:50 ` Hans Hagen
  1 sibling, 0 replies; 5+ messages in thread
From: Hans Hagen @ 2020-05-16 14:50 UTC (permalink / raw)
  To: mailing list for ConTeXt users, Keith McKay

On 5/16/2020 4:36 PM, Keith McKay wrote:
> Colleagues
> 
> I'm having problems with an Arithmetic Overflow error when calculating 
> the x and y values for functions of the form:
> 
> y = ax^4 + bx^3 + cx^2 + dx + e
> 
> to which I then add a bit of randomness and other embellishments. I have 
> had a look in the Metafun and Metafont manuals but nothing jumps out at 
> me. In the minimum working example below, this Arithmetic Overflow 
> occurs when xord =13.5 but I can get the calculation to work if I 
> separate out the equation and then add together, since it seems to be 
> failing on the pow(xord,4) instruction. Is there a switch in MetaFont 
> that should be set or a better way of getting round this Arithmetic 
> Overflow error?

\startMPpage[instance=doublefun]

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Arithmetic Overflow error in MetaFun
  2020-05-16 14:48 ` Keith McKay
@ 2020-05-16 19:46   ` Hans Hagen
  2020-05-17  8:19     ` Keith McKay
  0 siblings, 1 reply; 5+ messages in thread
From: Hans Hagen @ 2020-05-16 19:46 UTC (permalink / raw)
  To: Keith McKay, mailing list for ConTeXt users

On 5/16/2020 4:48 PM, Keith McKay wrote:
> Answering my own problem here.
> 
> I've just searched through the mailing list archives and it seems that 
> if I add [instance=doublefun] I do not get the Arithmetic Overflow 
> error. Joy!
fwiw, we have also instance=decimalfun

concerning arithmic:

in metapost we have

scaled  : upto 4K bp
double  : 64 bit floats
decimal : configurable accuracy
complex : dual doubles (only in luametafun)

in lmtx lua we have:

integer  : lua tries to be clever but mixing leads to double
float    : lua tries to be clever
decimal  : userdata arbitrary precision (only in lmtx)
complex  : userdata dual doubles (only in lmtx)

(in lmtx we have an extended math library and these function are also 
available in luametafun)

Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Arithmetic Overflow error in MetaFun
  2020-05-16 19:46   ` Hans Hagen
@ 2020-05-17  8:19     ` Keith McKay
  0 siblings, 0 replies; 5+ messages in thread
From: Keith McKay @ 2020-05-17  8:19 UTC (permalink / raw)
  To: Hans Hagen; +Cc: mailing list for ConTeXt users


[-- Attachment #1.1: Type: text/plain, Size: 1339 bytes --]

Thanks for your help Hans and the pointer to the extended math library.
Lots more to explore now.
Best Wishes
Keith

On Sat, 16 May 2020, 20:46 Hans Hagen, <j.hagen@xs4all.nl> wrote:

> On 5/16/2020 4:48 PM, Keith McKay wrote:
> > Answering my own problem here.
> >
> > I've just searched through the mailing list archives and it seems that
> > if I add [instance=doublefun] I do not get the Arithmetic Overflow
> > error. Joy!
> fwiw, we have also instance=decimalfun
>
> concerning arithmic:
>
> in metapost we have
>
> scaled  : upto 4K bp
> double  : 64 bit floats
> decimal : configurable accuracy
> complex : dual doubles (only in luametafun)
>
> in lmtx lua we have:
>
> integer  : lua tries to be clever but mixing leads to double
> float    : lua tries to be clever
> decimal  : userdata arbitrary precision (only in lmtx)
> complex  : userdata dual doubles (only in lmtx)
>
> (in lmtx we have an extended math library and these function are also
> available in luametafun)
>
> Hans
>
> -----------------------------------------------------------------
>                                            Hans Hagen | PRAGMA ADE
>                Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
>         tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
> -----------------------------------------------------------------
>

[-- Attachment #1.2: Type: text/html, Size: 2004 bytes --]

[-- Attachment #2: Type: text/plain, Size: 493 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2020-05-17  8:19 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-16 14:36 Arithmetic Overflow error in MetaFun Keith McKay
2020-05-16 14:48 ` Keith McKay
2020-05-16 19:46   ` Hans Hagen
2020-05-17  8:19     ` Keith McKay
2020-05-16 14:50 ` Hans Hagen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).