> > > ..<snip>..

> > > > > apply the transform to each mesh? :rolleyes:

> > > > there is no need for that because the vertice
> positions
> > > > are given after a transform is already done for you.
> in
> > > > other words all the vertice positions are given
> relative
> > > > to the whole model and not relative to the meshes.
> also
> > > > duplicate meshes are copied and saved as though they
> are
> > > > each unique meshes; they are not saved once and then
> > > > used repeatedly over a model. lightwave format is
> more
> > > > like what you are thinking i believe.

> > > it is quite a while since I programmed any .3ds, but I
> am
> > > sure there is (an optional?) transform block that
> should
> > > be applied to a mesh. I know I've had to use it in
> the
> > > past to make things line up..

> > > ..<snip>..


> > yes i think there is and i'm wondering why. i tried it
> > once and it made things worse however.

> read the matrix {column then row} or {row then column} ?


i just tried out the data that i know had to do with positions rather than orientation before anything else and decided that it was not right. i think the chunk i was working with was called local coordinate system. that was the only one i could find that looked like a transform matrix.


> > i'm hoping i
> > might solve the problem by looking at the animation part
> > of the file finally. what do you think?

> not sure you will.


ok well you've probably saved me lots of time. i can concentrate on other things then but i really don't see where else it could be.


> > i don't use other people's code but i use their
> > published spec/guides. my main way of giving is to
> > release free programs.

> bug-free?

no not bug free. is any program entirely bug free? what's your point? i don't see how releasing my entire source code is going to lead to a better 3ds2vxl because nobody knows the westwood normals correctly, and if they did they could just tell me so i could put it into my program and then credit them.