 | | | constraint updating in 4 and 4_2 | constraint updating in 4 and 4_2 2005-01-18 - By kim aldis
Back
>From your original mail:-
'First, a set of two orient constraints to keep the null oriented
relative to the cubes, should they be animated. The first has a blend
value of 1.0, meaning it takes full control'.
No, it doesn't take full control. The blend is just that, a blend *over* the
previous blends. The value of the second blend is added to the first so
neither has 'control' as such. Regardless of the value of the first, the
second will always have some control.
Remember also that with your position constraints you've a situation where
the sum of the blends to either of your cubes could be > 1. At this point
it'll probably be clamped at 1. If the sum is greater than this then
changing either will have no effect.
I'm still not convinced you have a correct understanding of how this works.
Either way, I thnk Matt's right, you're overcomplicating the issue. He knows
a great deal about how this works. I'd be inclined to take his advice and
move on.
K.
> -- --Original Message-- --
> From: owner-xsi@(protected)
> [mailto:owner-xsi@(protected)] On Behalf Of Brad Friedman
> Sent: 17 January 2005 14:34
> To: XSI@(protected)
> Subject: Re: constraint updating in 4 and 4_2
>
> kim aldis wrote:
>
> >That's not strictly speaking correct. Most of the discussion
> has been
> >about cycle dependancy, a condition that arises when two objects are
> >each dependant upon each other. There's no such thing as a correct
> >result under these circumstances. Xsi constraints generally
> behave well
> >but they do sometimes ally you to construct illegal
> conditions. Use the
> >cycledependany function to check for this.
> >
> >
> I'm aware of this. I ran cycle checks manuallly a number of
> times to make sure I was not creating them. And regarding
> the flakey constraints thing. I'm just relaying the
> impression I got from doing a search for the terms
> "constraint" and "update". And more importantly, I'm trying
> to deflect the "do a search of the archives before you ask a
> stupid question like that!" comment. I did not mean to
> disparage XSI in general.
>
> >Your condition doesn't contain cycle dependancies but it
> does behave as
> >indicated in the documentation. Simplifying your condition
> for clarity,
> >create only 2 position constraints on the null, one to each
> cube. Now,
> >it's easy to assume that a blend factor of 0.5 on each would
> result in
> >the null positioning midway between the two cubes.
> >
>
> I can see how newbies would want it to work that way. And
> thats how it works in Maya as well. However thats not how I
> have it set up. The first position constraint has a blend of
> 1.0. Full on. The second has a blend of 0.5 to put the null
> in the middle.
>
> The two constraints after that start with blends of 0.0.
> They don't do anything at first. If you turn either of them
> up to 1.0, the null should move to the target fully
> (overriding the first two constraints completely). If you
> turn it back down to 0.0, control should be turned back over
> to the result from the first two constraints. I understand
> the order of operations and what the blend value should do.
> I'm using it to get the results I want. And it works in
> almost all cases, except for interactive updates of the blend
> weight value, and animated updates of said value, unless I
> use the partial workaround I described in my original
> message. Please reread my original post for a description of
> my intended behavior vs. what is happening on my machine
> here. If you are getting the intended behavior instead of
> the errant behavior I describe (causing an assumption that
> I'm just doing it wrong, cause everything works as it should)
> I'd very much like to know because it indicates some odd
> condition over here in my world that I need to seek out.
>
> Thank you for your time.
>
> -brad
>
>
> --
> No virus found in this outgoing message.
> Checked by AVG Anti-Virus.
> Version: 7.0.300 / Virus Database: 265.6.13 - Release Date: 1/16/2005
>
> ---
> Unsubscribe? Mail Majordomo@(protected) with the following
> text in body:
> unsubscribe xsi
>
>
---
Unsubscribe? Mail Majordomo@(protected) with the following text in body:
unsubscribe xsi
|
|
|