[Changeset] Re: "steps plot style" seems not to work

Sergei Steshenko sergstesh at yahoo.com
Thu May 1 14:38:22 CDT 2008


--- David Bateman <adb014 at gmail.com> wrote:

> Sergei Steshenko wrote:
> > I started using 'stairs' even before writing the original Email; so far it works for me.
> > 
> > ...
> > 
> > Which part of the graphing process ('octave' or 'gnuplot') is responsible for setting tics
> > and their numerical values on 'y' axis when 'stairs' is called ?
> > 
> 
> 
> The tics are a property of the axis. So something like
> 
> bar ([4, 12 13 11 12 9, 6])
> set(gca(), 'xticklabel', {'Sun','Mon','Tue','Wed','Thu','Fri','Sat'},
> 'xtickmode','manual')
> 
> might be what you are looking for.
> 
> D.
> 

Sorry, I poorly formulated the question.

When I use 'stairs', initially tics look well.

When afterwards I zoom around, say, y=1, for a time being the ticks look OK, e.g.
0.99, 1, 1.01; 0.9998, 1, 1.0002; 0.999995, 1, 1.0001 - this one, by the way, shows some
kind of rounding asymmetry; then just 1, 1, 1, i.e. no fractional part.

Probably 'gnuplot' recalculates tics values when zooming is performed, and AFAIK 'gnuplot' uses 
single rather than double precision math.

But I wanted to make sure this phenomenon/problem does not come from 'octave'.

Thanks,
  Sergei.


Applications From Scratch: http://appsfromscratch.berlios.de/


      ____________________________________________________________________________________
Be a better friend, newshound, and 
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ


More information about the Bug-octave mailing list