Cloth And Collison Problem

[djfluid]

hey guys i have a very odd problem (well it is odd to me).

upfront i'm fairly new to cloth-- just played around with it in the past (standard tablecloth stuff) but now i'm trying to use it in an actual animation project.

i have my character and she's wearing boots. all of it is fairly low poly (the boots have alot more polys than the character mesh- but still not smoothed)

when i make the pants using the : panel, garmet, stitch method. not using a poly mesh converted.

after adding the objects to the solver as collison objects the pants drape fairly fine (still have to work out the small kinks but that will come later)

however the one boot- no matter what i do: removing then readding, adding it in a different order to the solver, trying different methods of stitching, etc. - the boot just doesn't get added to the actual collison equation.

it shows up in the hypergraph as being connected just like the "correct" boot, but nothing happens!

i have duplicated the "problem" boot on the -X axis from the "good" boot but have deleted history and such before adding it to the solver.

this is really bugging me guys

any help would be great.

also i've read through the manual over and over but nothing seems to be mentioned

thanks guys.

bryan

[drblacksnake]

duplicated as a copy or as an instance?
are both boots doubleSided?

[djfluid]

it is duplicated as a "copy", they are both double sided and the normals are pointing in the correct direction.

the only thing that i can come up with is that it is duplicated on the -X axis.....if i make 'dummy' objects to stand in as the boots without duplicating they both work fine, but if i duplicate one on the -X axis then the one that is duplicated doesn't get "solved" properly.

to me this is strange and there has to be something that i'm doing because i wouldn't think that cloth would care about what axis and object is duplicated on, but like i stated i've only just "played" and the manual doesn't seem to mention anything about conflicts with axis.

if it comes down to it i will probably have to use a lattice deformer and curve the other boot but i really don't want to.

thanks guys
Bryan

[drblacksnake]

try to untick doubleSided and freezeTransforms for -X duplicated copy

[djfluid]

well guys, i'm an idiot.

after looking at and double checking everything it ended up being that the normals were or had gotten reversed.

my boots are grouped objects, so naturally when i checked the normals it was with the group selected, hence that were pointing the correct way.

after removing the boot from the group i proceded to add it to the solver by itself....and no luck. so i went back and double checked the normals and sure enough, they were flipped.

i really hate posting questions when the solution was that simple, especially since in all the other postings the first thing that always came up was " did you check the normals ?"

well now that the normals are pointing out again, everything works just like they should.

thanks guys,

Bryan