So, Manny, based on the fact that you had to re-submit, am I to understand that my submission of
SI_DEFECT-0190 fell on deaf ears then?! Is Autodesk going to do anything at all more for Soft to fix what is broken? >.< On Mon, May 26, 2014 at 11:14 AM, Manny Papamanos < manny.papama...@autodesk.com> wrote: > Hello. > I do confirm that this particular functionality is broken since SI:2014 > and that it worked in SI:2013. > The issue seems specific with closed curves and happens with path and > param types of curve constraints. > > I have logged the issue as a regression so it can get immediate attention > here: > SPR-14215 Tangency on Path constraint (path and param) broken for closed > curves > > It’s possible this issue may have developed when resolving this fixed one: > SOFT-8622 Evaluation of fcurve tangents yield incorrect results in long > curve cases. > Thanks. > > > > Manny Papamanos > Product Support Specialist > Americas Frontline Technical Support > > > From: softimage-boun...@listproc.autodesk.com [mailto: > softimage-boun...@listproc.autodesk.com] On Behalf Of Thomas Volkmann > Sent: Monday, May 26, 2014 11:34 AM > To: softimage@listproc.autodesk.com > Subject: Re: BUG!?: Path-constrain tangency on closed curves > > Let's hope so! This feature is basic enough to be high on the priority > list.... > > > Eric Turman <i.anima...@gmail.com<mailto:i.anima...@gmail.com>> hat am > 26. Mai 2014 um 17:18 geschrieben: > Correct. I reported the bug at the end of the beta for 2015, it works fine > in 2013 but breaks after 2014. > > I hope that this regression bug is a low hanging fruit that they can get > to soon. > > On Mon, May 26, 2014 at 10:10 AM, Thomas Volkmann < > li...@thomasvolkmann.com<mailto:li...@thomasvolkmann.com>> wrote: > Hi, > > we just opened up an old scene and found that some objects were moving > strange on their curves. It seems some bug was introduced to the > path-constrain operator when using closed curves and tangency. Tangency is > doing something wrong. > > A little python-script to reproduce: > > > Application.NewScene("", "") > Application.SICreateCurve("crvlist", 3, 0) > Application.SIAddPointOnCurveAtEnd("crvlist", -4.12493302610738, 0, > -1.5985306969839, False, 0, "") > Application.SIAddPointOnCurveAtEnd("crvlist", -1.94612224821476, 0, > -2.72067807367415, False, 0, "") > Application.SIAddPointOnCurveAtEnd("crvlist", 2.89802986962416, 0, > -3.2076476899737, False, 0, "") > Application.SIAddPointOnCurveAtEnd("crvlist", 6.282590301302, 0, > 1.95846476120408, False, 0, "") > Application.SIAddPointOnCurveAtEnd("crvlist", 0.973061124107382, 0, > 4.54152098679297, False, 0, "") > Application.SIAddPointOnCurveAtEnd("crvlist", -2.87687636692617, 0, > 3.71578989828504, False, 0, "") > Application.SIAddPointOnCurveAtEnd("crvlist", -5.01338013942281, 0, > 1.09038848867011, False, 0, "") > Application.ApplyTopoOp("CrvOpenClose", "crvlist", 3, > "siPersistentOperation", "") > Application.ActivateObjectSelTool("") > Application.ActivateObjectSelTool("") > Application.GetPrim("Null", "", "", "") > Application.ApplyCns("Path", "null", "crvlist", "") > Application.SetValue("null.kine.pathcns.perc", 3.646, "") # on linux the > objects rotation will be nan when this is 0 and tangency gets activated > Application.SetValue("null.kine.pathcns.tangent", True, "") > Application.SetValue("null.kine.pathcns.dirx", 0, "") > Application.SetValue("null.kine.pathcns.diry", 0, "") > Application.SetValue("null.kine.pathcns.dirz", 1, "") > Application.SetValue("null.kine.pathcns.upvct_active", True, "") > > > > When you now move the path age slider you will notice that the null > doesn't align to the curve like it should. > Some observations: > + was introduced either with 2014SP1 or SP2 (works in 2014 but not in SP2 > or 2015, don't have SP1 available atm) > + only closed handmade curves are affected (but not > getPrimitve->curve->circle) > + null rotation becomes 'nan' on linux when path-age is 0 and tangency is > activated (no problem on windows) > > Please please fix it for for the absolute final Softimage version! > > cheers, > Thomas > > > > > -- > > > > > -=T=- > > > -- -=T=-
