Version: 1.0 (using KDE 3.0.5) Compiler: gcc version 2.95.4 20011002 (Debian prerelease) OS: Linux (i686) release 2.4.18-bf2.4 When plotting a function like f(x)=(-x+1)^x kmplot seems to stop plotting once f(x)==nan. In this case the last point drawed to the coordinate system is (1|0) as f(1)=(-1+1)^1=0 is a rational value. f(1.1), f(1.2) and so on of course are not plotted because the root of a negative number is not defined. The problem is, that kmplot even does not draw all f(x) where x is integer and x > 1 like f(2)=(-2+1)^2=1 or f(3)=(-3+1)^3=-6. When following the function with the mouse in the statusbar it says "x= +2.00" and "y= +nan" but this seems to happen only when f(x) is not to see anymore in the coordinate system. In every other case the correct value is displayed in the statusbar. Regards, Benedikt Gollatz
Is this bug still there in a recent version of KDE, such as 3.5.8 or KDE4.0 RC2?
I can reproduce the bug in KDE 3.5.8 (kmplot 1.2.0). Since I am unable find binary packages of KDE 4 for my distribution, I cannot make any statement about the KDE 4 version of kmplot. By the way, I think my previous idea about how the coordinate system used affects the occurrence of the bug seems wrong to me now. It seems more likely that kmplot displays a rounded x value in the status bar but uses an exact x value to calculate f(x). (Haven't looked at the sources, though.)
*** Bug 156368 has been marked as a duplicate of this bug. ***
Created attachment 29764 [details] plotted function with maxima
Comment on attachment 29764 [details] plotted function with maxima i think that the problem is that negative parts of the functions are not plotted. for example the third root of a negative real number is a negative real number. the square root of a negative real number is not defined. the same problem appears in KCalc and in gcalctool as well. qtiplot also can not handle this functions correctly. only in maxima i got the correct result.
The problem is that the above mentioned function is defined for (-∞,1] and {2,3,4,...}. See Bug 156368 for an explanation. Since the function is only pointwise defined in (1,∞), there's no curve one can see. Notice however, that the calculator correctly calculates f(i) where i is integer.