These Vectors constructs are not complex at all, but they are highly confusing. Glad you figured out.
Fabrice On Mar 24, 2011, at 6:52 PM, Dave wrote: > HI Fabrice, > > I was indeed doing UV's wrong. It turns out that ASCollada's > DaePrimitive reorders the UV's to be in face order, not in INDEX > order. Once I discovered this, I was able to fix it pretty easily. > > -Dave > > On Mar 23, 3:09 pm, Dave <[email protected]> wrote: >> Hi Fabrice, >> >> Thank you for replying. >> >> What you describe is exactly what I am doing, and I have plotted the >> values (outputted via a trace and imported into excel) that have been >> filled in the array, and the results are proper. I will look further. >> >> -Dave >> >> On Mar 23, 2:48 pm, Fabrice3D <[email protected]> wrote: >> >> >> >> >> >> >> >>> sounds like you do not fill your vector.<Number> of verts the propper way >>> trace the relation indice*3 (+1 and +2) to trace a vert and its indice*2 uv >>> (+1 for v) >> >>> fabrice >> >>> On Mar 23, 2011, at 10:40 PM, Dave wrote: >> >>>> I'm having a heck of a time getting my textures to map properly to my >>>> objects. I'm at my wits end. I cannot figure out what is wrong. >> >>>> I have written a collada parser for broomstick that takes care of >>>> geometry, textures and materials. For whatever reason, however, my UV >>>> mapping when rendered in away3D looks completely wrong. >> >>>> I've reduced my problem down to a cylinder with cylindrical mapping of >>>> a 512x512 texture. It's still not clear what is happening. >> >>>> [Code snippet (not very useful, I realize)] >> >>>> var uvData:Vector.<Number> = new Vector.<Number>(numVertexIndexes * 2, >>>> true); >>>> <snip - code to fill in uv's> >>>> subGeometry.updateUVData(uvData); >> >>>> I do properly dereference the texcoord chunk in collada to produce the >>>> proper UV list in indexOfVertex order. >> >>>> I have confirmed (by plotting in Excel) that my UV's in Collada are >>>> correct, and I've confirmed that my numbers are the same on read of >>>> the Collada file, and setting into the subGeometry object. For >>>> whatever reason, those values are not being properly associated with >>>> my vertices. >> >>>> I am using a simple BitmapMaterial. >> >>>> Is there some dumb mistake based on this description (which seems >>>> fairly simple) that sticks out, or is there a simple mistake that >>>> everyone makes with UV's? >> >>>> -Dave
