=?ISO-8859-1?Q?J=F6rg_=27Yadgar=27_Bleimann?= <yaz### [at] gmxde> wrote:
> I applied your suggestions - but the hole at the north pole (not at the
> south pole!) remained!
>
> Obviously, the last belt (#case (398)) doesn't connect - and how should
> it do, as its "baseline" still is a-1, i. e. 397?
>
> But when, in #case (398) I replace a-1 by a, I again get that "face
> index out of range error! Why? The absolute number of faces did not
> change...

Yadgar, I guess you need to approach this thing more systematically; right now,
it seems like you're still using the "tweak it until it works" approach, which
doesn't really seem to get you anywhere at the moment.

What you need is a clear picture of what "loop number" actually corresponds to
which vertex index numbers; and what "belt" should connect which pair of
"loops".

I'm not talking about a "mental picture" here. Do write it down. Take this
advice from a professional software developer: If at first you don't succeed,
*document*. To a painstaking detail. It forces to formalize the mental picture
of the intended solution, which is often fuzzier than one thinks.


For instance, the notion that "1+a*800" should be the correct vertex index
baseline for loop #a may seem intuitive - but as a matter of fact it's wrong.
For instance, loop #1 - the very first loop - is comprised of vertices
#1..#800, matching a loop baseline formula of 1+(a-1)*800.

So the baseline I suggested for the last belt - #398 - matches the baseline for
loop #398 - the last full loop, which belt #398 should connect to the pole.

Trust me - sometimes I do know what I'm doing :P

So there still must be a different problem.

Post a reply to this message