[Kst] extragear/graphics/kst/src/libkstmath

[barth.netterfield at utoronto.ca]

Makes sense - my "math" was wrong (I used 8.5+11 = 20 at 100 pt/inch = 2000).

I see that the default kprinter margins seem to be 0.5 top and bottom, and 
0.25 left and right.  

So, following your suggestion, we should use ((11-2*0.5) + (8.5 - 2*0.25))*72 
= 1296 or something like that.   So I would say that if 1296 looks OK, then 
change to that, and document it.

thanks for clarifying this!

cbn

Quoting Andrew Walker <arwalker at sumusltd.com>:

> With about a dozen plots in front of me it is pretty clear that using 2000
> gives lines that are too narrow. A width of 0 or 1 is too narrow, and is
> hard to distinguish from a width of 2.
> 
> The value of 2000 doesn't seem correct from a theoretical viewpoint either.
> The value should be approximately (8+11)*72 (i.e. 1400). Assuming a
> reasonable margin this will drop to 1200. This value gives reasonable
> results as well.
> 
> Obviously this becomes a matter of opinion at some point, but I think its
> clear that 2000 is too high.
> 
> Andrew
> 
> -----Original Message-----
> From: barth.netterfield at utoronto.ca
> [mailto:barth.netterfield at utoronto.ca]
> Sent: Friday, June 02, 2006 8:50 AM
> To: kst at kde.org
> Subject: Re: [Kst] extragear/graphics/kst/src/libkstmath
> 
> 
> So am I right in concluding that you have just increased line widths by
> 2000/800 in all cases?
> 
> If so, I am curious to know the motivation for this.  Was line with = 2 not
> a
> ~2pt line when printed?  Or was it desired to not be able to make a 2pt
> wide
> line?
> 
> Quoting Andrew Walker <arwalker at sumusltd.com>:
> 
> > From trial and error to give the best output when printing.
> >
> > -----Original Message-----
> > From: George Staikos [mailto:staikos at kde.org]
> > Sent: Friday, June 02, 2006 5:43 AM
> > To: kst at kde.org
> > Subject: Re: [Kst] extragear/graphics/kst/src/libkstmath
> >
> >
> > On Thursday 01 June 2006 12:00, Andrew Walker wrote:
> >
> > >  int KstPainter::lineWidthAdjustmentFactor() const {
> > > -  const QRect& w(window());
> > > -  /* For square windows, the line width is in units of 0.1% of window
> > size
> > > */ -  /* For printing to Letter or A10, line width is ~in points */
> > > -  const int factor = (w.width() + w.height()) / 2000;
> > > +  int factor = 1;
> > > +
> > > +  if (type() == P_PRINT || type() == P_EXPORT) {
> > > +    const QRect& w(window());
> > > +
> > > +    factor = (w.width() + w.height()) / 800;
> >
> >    Where does this number come from?  The previous algorithm was
> > documented,
> > and was even explicitly documented:
> >
> > ------------------------------------------------------------------------
> > r520834 | staikos | 2006-03-20 17:58:11 -0500 (Mon, 20 Mar 2006) | 2
> lines
> >
> > document the magic number and increment it by one
> >
> >
> >    Please document the new approach.  Thanks.
> >
> > --
> > George Staikos
> > KDE Developer				http://www.kde.org/
> > Staikos Computing Services Inc.		http://www.staikos.net/
> > _______________________________________________
> > Kst mailing list
> > Kst at kde.org
> > https://mail.kde.org/mailman/listinfo/kst
> >
> > _______________________________________________
> > Kst mailing list
> > Kst at kde.org
> > https://mail.kde.org/mailman/listinfo/kst
> >
> 
> 
> _______________________________________________
> Kst mailing list
> Kst at kde.org
> https://mail.kde.org/mailman/listinfo/kst
> 
> _______________________________________________
> Kst mailing list
> Kst at kde.org
> https://mail.kde.org/mailman/listinfo/kst
>