Re: [Gnuplot-bugs] SAMPLES, values at ticmarks

[Erik L. <eri...@gm...>]

Thank you. About the xtics: Mea culpa for not knowing that the default
format is "% h".

About SAMPLES, I would argue that any plotting program will have to
discreetly sample a function in order to plot it (i.e., I do not view it as
an important point to convey to new users), but I accept your perspective.
That being said, would I cause problems/incompatibilities if I set SAMPLES
to a different value in the precompiled binaries that I distribute for
macOS?

Kind regards,

Erik

On Fri, Jul 11, 2025 at 11:56 AM Hans-Bernhard Bröker <HBB...@t-...>
wrote:

> Am 11.07.2025 um 16:58 schrieb Erik Luijten:
> > Hi,
> >
> > I hope this is not sacrilege, but would it be possible to consider
> changing
> > the default value of SAMPLES in gadgets.h?  I understand that this was
> set
> > to this value in a time when the calculation was costly and screens had
> > much lower resolution, but nowadays there doesn't seem to be a good
> reason
> > to not set it to, say, 500.
>
> Computational cost is not the only reason.  It helps people understand
> the fact that the sampling is, in fact, discrete if the first few curves
> they see, as absolute newbies, show some visible sampling right off the
> bat.  It keeps them from forming an incorrect mental image of what this
> program will do.
>
> > Related to that particular demo: I noticed that the values at tics marks
> > are not located exactly where one would expect them.
>
> That depends, as always, on what one expects.  I think you'll understand
> better if you replace the default tic format with "%+h" and re-do your
> exercise.
>
>
>
> _______________________________________________
> gnuplot-bugs mailing list
> gnu...@li...
> Membership management via:
> https://lists.sourceforge.net/lists/listinfo/gnuplot-bugs
>