Autodesk Inventor

[#1]

i had something wierd happen just now...

often when constructing lofts i project the geometry from the previous
profile, then change it from reference to normal and dimension... this is to
make it easier to ensure i've got the right number of vertices so the loft
doesn't go silly...

well just then i projected a spline (for the first time in R6) which had
four control points, changed the type to normal and ended up with at least
fifty CP's...

that's kinda annoying...

[#2]

Hi, Colin.

I'd never notice that, but I don't ever change the projections.... I had
noticed that it happens if a spline is offset and the offset curve is then
changed to a spline (what are they before being changed?).

Can you post an example of a problem caused by having an unequal number of fit
points in spline profiles? Not saying it doesn't happen, just that I've never
seen one.

Have a good one,
Jeff
==========================



"AutoCOL" wrote in
message news:E06ECF54053D6643BE5CF6A16E2F06D1@in.WebX.maYIadrTaRb...
i had something wierd happen just now...

often when constructing lofts i project the geometry from the previous
profile, then change it from reference to normal and dimension... this is to
make it easier to ensure i've got the right number of vertices so the loft
doesn't go silly...

well just then i projected a spline (for the first time in R6) which had
four control points, changed the type to normal and ended up with at least
fifty CP's...

that's kinda annoying...

[#3]

Projected splines changed to normal do show many control points. I have used
3 point arcs to trace over the spline, place any CPs I want then delete the
spline. Makes it normal lines then. Don't know if the projected spline
changed to normal would cause any problems but all the current problems with
lofts don't need the potential of an increase, IMWO.
~Larry

"AutoCOL" wrote in
message news:E06ECF54053D6643BE5CF6A16E2F06D1@in.WebX.maYIadrTaRb...
> i had something wierd happen just now...
>
> often when constructing lofts i project the geometry from the previous
> profile, then change it from reference to normal and dimension... this is
to
> make it easier to ensure i've got the right number of vertices so the loft
> doesn't go silly...
>
> well just then i projected a spline (for the first time in R6) which had
> four control points, changed the type to normal and ended up with at least
> fifty CP's...
>
> that's kinda annoying...
>
>

[#4]

What sort of problems are you seeing?

Jeff
===================

"Larry Caldwell" wrote in message
news:CAA50B73A0BCECF008CA1837C7FD0BF1@in.WebX.maYIadrTaRb...
............
but all the current problems with lofts don't need the potential of an
increase, IMWO.
~Larry

[#5]

Well ... quite a few, actually. Simple lofts work pretty well but slightly
more complex lofts (like the one in CF) fail if I try to make it with one
loft and get Access Violations trying with two lofts. Using two lofts
results in visual lines where the lofts are joined instead of a smooth
interface. Fillets won't hide the joint lines and fail most of the time.
Make a cut on a loft it won't fillet in some cases (at least on curved
edges). Have two lofts at zero offset causes failure. Lofting from inside a
loft fails. Circles lofted to other circles leave surfaces anomalies that
won't take a large enough fillet to hide the imperfection. Lofts won't join
properly with other geometry. Basically many of the lofts I've tried just
plain fail or aren't what I'm trying to achieve without adding 50 splines to
smooth things out. Loft a shape with sharp edges they won't fillet
correctly. Guess that's a few to start with. Think maybe I just can't model
lofts very well? Maybe that's it, eh?
~Larry



"Jeff Howard" wrote in message
news:4BA94C80592ABABF349EACBB01B3BBF2@in.WebX.maYIadrTaRb...
> What sort of problems are you seeing?
>
> Jeff
> ===================
>
> "Larry Caldwell" wrote in message
> news:CAA50B73A0BCECF008CA1837C7FD0BF1@in.WebX.maYIadrTaRb...
> ............
> but all the current problems with lofts don't need the potential of an
> increase, IMWO.
> ~Larry
>
>
>

[#6]

Yup, that's it. You're just not holding your mouth right 8~)

I get the picture, though. I just wasn't considering most of the things you
mention to be loft specific problems, but more general problems associated
with a modeling engine that's designed for modeling cubes and cylinders. My
initial impression (limited time using) is that IV6 has brought some
considerable improvement, though, and thus my curiosity. I didn't see a file
in CF (a new one or a previous one?), but if you have any of the problem files
tucked away, I'd be interested in seeing them if for no other reason than to
get a leg up on the learning curve.

But, back to my original thought.... If I've ever seen an instance of a
problem caused by having an unequal number of fit points in spline profiles, I
can't remember it and would be interested in seeing one for curiosity's sake.

Have a good one,
Jeff
====================



"Larry Caldwell" wrote in message
news:B8A81051BF87533240073F1763D66B38@in.WebX.maYIadrTaRb...
Well ... quite a few, actually. Simple lofts work pretty well but slightly
more complex lofts (like the one in CF) fail if I try to make it with one
loft and get Access Violations trying with two lofts. Using two lofts
results in visual lines where the lofts are joined instead of a smooth
interface. Fillets won't hide the joint lines and fail most of the time.
Make a cut on a loft it won't fillet in some cases (at least on curved
edges). Have two lofts at zero offset causes failure. Lofting from inside a
loft fails. Circles lofted to other circles leave surfaces anomalies that
won't take a large enough fillet to hide the imperfection. Lofts won't join
properly with other geometry. Basically many of the lofts I've tried just
plain fail or aren't what I'm trying to achieve without adding 50 splines to
smooth things out. Loft a shape with sharp edges they won't fillet
correctly. Guess that's a few to start with. Think maybe I just can't model
lofts very well? Maybe that's it, eh?
~Larry

[#7]

Hey ... that gives me an idea ... a jaw jig! We could design one! When you
get something right, it could sense the position your mouth was in and
remember it for next time you are doing that. We'll be rich!

I'm not that good at saving failures. I usually keep on chugg'n till I find
something that's close enough. I will start saving some of them and pass
them along though.

I haven't seen any problems with uneven numbers of control points either;
just don't like having more than I need (get confused easily I guess).

Another problem I run into with lofts is getting the rail geometry in the
right place on rotated profiles. I've been using three-point workplanes but
when you have several profiles sometimes it's a real pain to get the rails
in the right place.Sure would be nice if the 3D sketch command had a
three-point arc tool to make rails with though. Dream on, no?
~Larry


"Jeff Howard" wrote in message
news:3E14ED0685994E56EFA06C486109598D@in.WebX.maYIadrTaRb...
> Yup, that's it. You're just not holding your mouth right 8~)
>
> I get the picture, though. I just wasn't considering most of the things
you
> mention to be loft specific problems, but more general problems associated
> with a modeling engine that's designed for modeling cubes and cylinders.
My
> initial impression (limited time using) is that IV6 has brought some
> considerable improvement, though, and thus my curiosity. I didn't see a
file
> in CF (a new one or a previous one?), but if you have any of the problem
files
> tucked away, I'd be interested in seeing them if for no other reason than
to
> get a leg up on the learning curve.
>
> But, back to my original thought.... If I've ever seen an instance of a
> problem caused by having an unequal number of fit points in spline
profiles, I
> can't remember it and would be interested in seeing one for curiosity's
sake.
>
> Have a good one,
> Jeff
> ====================
>
>
>
> "Larry Caldwell" wrote in message
> news:B8A81051BF87533240073F1763D66B38@in.WebX.maYIadrTaRb...
> Well ... quite a few, actually. Simple lofts work pretty well but slightly
> more complex lofts (like the one in CF) fail if I try to make it with one
> loft and get Access Violations trying with two lofts. Using two lofts
> results in visual lines where the lofts are joined instead of a smooth
> interface. Fillets won't hide the joint lines and fail most of the time.
> Make a cut on a loft it won't fillet in some cases (at least on curved
> edges). Have two lofts at zero offset causes failure. Lofting from inside
a
> loft fails. Circles lofted to other circles leave surfaces anomalies that
> won't take a large enough fillet to hide the imperfection. Lofts won't
join
> properly with other geometry. Basically many of the lofts I've tried just
> plain fail or aren't what I'm trying to achieve without adding 50 splines
to
> smooth things out. Loft a shape with sharp edges they won't fillet
> correctly. Guess that's a few to start with. Think maybe I just can't
model
> lofts very well? Maybe that's it, eh?
> ~Larry
>
>
>
>