Prev: Magnitude FFT block Simulink
Next: xcorr2 vs. normxcorr2
From: matt dash on 20 Mar 2010 16:43
I just solved a (minor) problem that's been (minorly) bugging me for years. I make a plot matrix (not using the plotmatrix command, but same idea) that's supposed to have exactly 1 pixel between each axes. However the resulting plot ends up having 0 pixels between some axes and 2 pixels between others. (I know this sounds ridiculous but the visual effect is actually really annoying) Sometimes the problem doesnt occur unless the axes are larger than a certain size. The positions returned with the get command are always the "correct" values, not showing the +/-1pixel error. So they axes think they're in the right place, they just appear wrong on the screen.

I just realized the problem is somehow related to the xlim/ylim values of the axes (as well as somehow being linked to the axes dimensions), and that if I set the xlim/ylim to *single precision* vectors the error never occurs. I still don't know why it happens in the first place, but on the off chance that anyone else has ever had this problem, try setting your axes limits to single precision and it should fix it. Also if anyone can explain why this happens in the first place, I'd like to hear it.
 | 
Pages: 1
Prev: Magnitude FFT block Simulink
Next: xcorr2 vs. normxcorr2