|
Seifert Surface
Mathematician
Join date: 14 Jun 2005
Posts: 912
|
07-14-2005 02:10
This is a building/scripting question, but I think it fits better in scripting.
So I've been playing around with taking a few prims arranged in some precise way (so that the edges all link up nicely), linking the lot, and using the build tools to rescale the whole link (which as I understand it is not possible to do using scripting? (at least without doing some repositioning of all the elements in the set relative to the center etc. - is there such a script available anywhere?)). Anyway, using the build tools it's possible to scale everything down until the smallest prim is at 0.01 meters.
Now (quite wonderfully), the various prims in my set are still arranged in that precise way I set it up when they were big, despite the fact that using the build tools' precision of 0.001 meters, there's no way I could arrange them how I want if I started out with them that small.
So clearly there is extra precision in the game that I can't get to via the building tools. Can I also get to that precision using scripting? Are there limits to the precision one can get out of llSetPrimitiveParams, other than the precision limit for floats?
Cheers,
Seifert
|
|
Lex Neva
wears dorky glasses
Join date: 27 Nov 2004
Posts: 1,361
|
07-14-2005 07:56
Interesting... this'll require some more testing. I have to wonder... is it possible that when you position something to a precision of .001 meters, then scale it down by a factor of 10, that the positions get rounded off? It may look like it's lined up right, but maybe things are wrong... or maybe there's some "hidden" precision. This'll take some testing. Has anyone already looked into this kind of thing?
|
|
Strife Onizuka
Moonchild
Join date: 3 Mar 2004
Posts: 5,887
|
07-14-2005 07:58
when things are linked the 0.001 pericion limit disapears.
_____________________
Truth is a river that is always splitting up into arms that reunite. Islanded between the arms, the inhabitants argue for a lifetime as to which is the main river.
- Cyril Connolly
Without the political will to find common ground, the continual friction of tactic and counter tactic, only creates suspicion and hatred and vengeance, and perpetuates the cycle of violence.
- James Nachtwey
|
|
Seifert Surface
Mathematician
Join date: 14 Jun 2005
Posts: 912
|
07-14-2005 13:41
From: Lex Neva Interesting... this'll require some more testing. I have to wonder... is it possible that when you position something to a precision of .001 meters, then scale it down by a factor of 10, that the positions get rounded off? It may look like it's lined up right, but maybe things are wrong... or maybe there's some "hidden" precision. This'll take some testing. Has anyone already looked into this kind of thing? Positions possibly, I don't have a good example that tests that yet. Scaling though I (was, see below) positive there is much hidden precision. There are currently (but probably wont be for long) some examples up at Montmarte (83,120). The seams between different prims in the things I've called "wild knot flower" are hard to find but are just to the left or right of vertically down. They fit together because the scaling and taper on the tori they're made of are exact at the largest scale. But then smaller ones are just scaled down when linked... and the smallest one is (I think) a 0.01 meter torus, which means the next biggest is (0.01)/0.95 = (something with lots of precision). Yeah its strange. Hmm. I just tried unlinking it, moving it around, even rescaling things and moving them back to what they should be... nothing seems to break it. I even tried copying down all the params of the smallest part and recreating it from scratch... and it still fit... so it can't be hidden precision because I just typed in the precision. Ok, I'm very confused  Anyone any idea what's going on? |
|
Seifert Surface
Mathematician
Join date: 14 Jun 2005
Posts: 912
|
07-14-2005 13:45
From: Strife Onizuka when things are linked the 0.001 pericion limit disapears. I have had cases in which I unlink some prims and they 'jump' to new positions. But apparently this doesn't always happen. |
|
Csven Concord
*
Join date: 19 Mar 2005
Posts: 1,015
|
07-27-2005 15:24
I'm finding that scaling down linked sets built with the "Copy" feature results in some mismatches on the smaller scale object. Irritating. And I too have found that unlinking afterward can sometimes cause a snap-back reaction.
|