The way fcurve editing tools work right now the curves are clamped to whole numbers, so in practice I don't think users entering fractional fcurves for int inputs is such a problem.
We can make it so Ctrl+Clicking and the panel clamp the points to int too (current ways I could find to set fractions). float precision error makes me think rounding at 0.5 this is worth keeping, with larger values and some modifiers applied I bet values over say - 10000 can become 9999.999 which then would get rounded down to 9999.0. Rounding at 0.5 means just means its less error prone even after driver, modifiers, unit-scaling. Really large float values too can't represent whole numbers and get rounded down wrongly, managed to redo this but was somewhere around 220000 though this isn't really such a normal use case and at these values drawng fcurves starts to go strange too. Even so, IMHO rounding this way is best kept and any chance of users to encounter rounding as an issue they have to deal with can be solved on the tool/ui level. On Fri, May 20, 2011 at 5:12 PM, Dan Eicher <[email protected]> wrote: > On Thu, May 19, 2011 at 5:34 PM, Matt Ebb <[email protected]> wrote: >> On Fri, May 20, 2011 at 10:17 AM, Campbell Barton >> <[email protected]>wrote: >> >>> >>> Aligorith and I discussed this before committing, and are aware of the >>> implications. >>> >>> To me if you are editing floats, but have them converted to ints later >>> it makes most sense (on a user level), to round them. If the GUI makes >>> it unintuitive then it should be made intuitive (grid lines when >>> zoomed in for example), if you want I can add this. >>> >> >> Rounding them to whole numbers, sure, I totally agree, and if floor() works >> better then I'm all for it. But when animating bool values it makes perfect >> sense that as soon as your curve crosses from 0 to 1 (not 0 to 0.5), the >> value is switched on. Same goes for integer values, animating my curve from >> 4.0 to 4.51 should not switch the value to 5. >> >> cheers >> >> Matt >> _______________________________________________ >> Bf-committers mailing list >> [email protected] >> http://lists.blender.org/mailman/listinfo/bf-committers >> > > I totally agree, < 1 == false and >= 1 == true is how I always thought > it worked. > > Dan > _______________________________________________ > Bf-committers mailing list > [email protected] > http://lists.blender.org/mailman/listinfo/bf-committers > -- - Campbell _______________________________________________ Bf-committers mailing list [email protected] http://lists.blender.org/mailman/listinfo/bf-committers
