Hi, I am working on the Tcl examples and found possible causes for the rounding errors vis-a-vis the C examples, namely not using braces around all expressions. (The lack of braces means that the numbers are filled as strings before the expression is evaluated, whereas with braces the [expr] command can use the numerical value directly.)
Example 13 contained some deviations from the C version, but I noticed that the C version (and the Tcl version too) uses a loop with a floating-point variable. This in general is asking for rounding trouble. It did not happen in the C case, because effectively the numbers used are integer (they did happen in the Tcl case, as there a fraction 2pi/500 was used). I wonder though whether we should use this kind of loops at all. Changing it means that we need to propagate it to all languages, but leaving it might suggest that we accept this - IMHO rather undesirable - practice. What are your opinions? Regards, Arjen ------------------------------------------------------------------------------ _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
