|
Fenix Eldritch
Mostly harmless
Join date: 30 Jan 2005
Posts: 201
|
04-09-2005 08:38
Since 1.6 came out, the method in which multi primed objects snap to the grid has changed. Originally, as I’m sure you all remember, the object would snap in relation to the center of its root prim. Now however, SL calculates an object’s center of mass and snaps it to the grid based on that.
The problem with this (in my opinion) is that it becomes incredibly difficult to work with objects that are not perfectly and equally proportioned in each direction. Because SL now uses the center of mass, almost all things won’t snap to the grid at the intervals defined in the grid options. So if your project is not complete and has been saved as multiple parts, repositioning them is a nightmare. The only way to counter this is to unlink the whole thing, manually reposition each prim, and then re-link the object. And that is only effective if you don’t have to move or re-rezz the object again.
All of my projects are now unworkable because of this change. Particularly my airship, which falls under the “Super Ship” category created by players. It’s size far surpasses the normal link distance, so it’s saved as a total of three group selections (the full ship is a bit under half a sim). All three groups were created in a way that their root object (the last object selected in the group selection) snaps to the original .5 grid and allows me to rezz the complete ship perfectly. Now of course, that is impossible. Manually entering the .5 coordinates is of little use because that only affects the root object of the group selection. And since all objects now use the center-o-mass snap, I must enter the coordinates for every single object if I want to align all parts correctly. And again, even if I do spend the time required to reposition everything, it’ll only work on that single rezz and for that single location .
I won’t go so far as to say that this new snap method is completely unusable, but is does seem less efficient. Think about it: an object’s center is now dynamic, depending on its construction of prims. Both the size and location of each prim now affects the center of mass of the whole object. There is no solid anchor to work off of anymore. Working with a small linkable object is one thing, but when you’re dealing with something that isn’t linked, (either because it’s too big or not complete) it’s almost hopeless.
So after my ranting, I present you with my suggestion. Please consider adding an option to enable all objects (and group selections) snap to the grid according to their root prim - the way they used to. Thanks for reading.
|
|
Lex Neva
wears dorky glasses
Join date: 27 Nov 2004
Posts: 1,361
|
04-09-2005 08:54
So that explains why I couldn't get snapping by reference to actually line things up together yesterday! This center-of-mass business means that if you choose one object as a reference, you can't line the others up with it in any way that makes any sense. You snap them to the "0x" gridline and check the coordinates, and the numbers will end up not being equal. From: someone I won’t go so far as to say that this new snap method is completely unusable, ... Well, I would. I ended up having to punch numbers in by hand. At the very least, it makes the reference mode practically unusable for anything but lining up individual prims.
|
|
Tiger Crossing
The Prim Maker
Join date: 18 Aug 2003
Posts: 1,560
|
04-10-2005 12:06
It's not the "center of mass" but it's not the correct center either.
This has been an age-old problem with many aspects of Second Life, the unreliability of chosen "center" calculations. Depending on what you are doing and what version you are running, you may end up using any combination of these:
Center of Mass - This is supposed to only be for physics calculations, and usually is. Two linked default cubes will have this center between them, but if one is hollow, this center will be closer to the other.
Average Center - This sums all of the prims in the link's centers and divides by the number of them. So it is usually somewhere around the middle, but adjustments to the position of any prim in the link can cause this one to move.
Bounding Box Center - This is the center of an imaginary box that is just large enough to contain the entire link. In this case, only movement of outlying prims in the link will change it. [ This is where the edit tool red/green/blue handles sit. ]
Root Center - By far the most useful, the last prim added to the link (the root) is used for the center. Moving any other prim in the link does nothing to this.
Second Center - NOT normally used, but right now (as of 1.6 and 1.6.1) the llRezAtRoot function seems to be using the next-to-last prim added to the link as the center. Rather broken.
Personally, I wish we could use the Root Center for EVERYTHING and be done with this mess.
_____________________
~ Tiger Crossing ~ (Nonsanity)
|
|
SuezanneC Baskerville
Forums Rock!
Join date: 22 Dec 2003
Posts: 14,229
|
Maybe a user controllable choice of the useful ones?
04-10-2005 13:06
Perhaps a drop down box with a choice of centering options, defaulting to root, with choices of root, bounding box, and center of mass?
_____________________
-
So long to these forums, the vBulletin forums that used to be at forums.secondlife.com. I will miss them.
I can be found on the web by searching for "SuezanneC Baskerville", or go to
http://www.google.com/profiles/suezanne
-
http://lindenlab.tribe.net/ created on 11/19/03.
Members: Ben, Catherine, Colin, Cory, Dan, Doug, Jim, Philip, Phoenix, Richard, Robin, and Ryan
-
|
|
Jeffrey Gomez
Cubed™
Join date: 11 Jun 2004
Posts: 3,522
|
04-10-2005 13:07
I agree with Tiger on this one. Center calculations should be standardized - jumping through the hoops to do so presently are a bit convoluted and unneccesary.
Edit: Standardized by default, that is.
_____________________
---
|
|
Fenix Eldritch
Mostly harmless
Join date: 30 Jan 2005
Posts: 201
|
04-11-2005 04:13
Ah, thanks for the more indepth explanation, Tiger. ^^ From: Tiger Crossing Personally, I wish we could use the Root Center for EVERYTHING and be done with this mess. My thoughts exactly.
|
|
McWheelie Baldwin
Registered User
Join date: 9 Apr 2004
Posts: 154
|
04-11-2005 09:21
From: Tiger Crossing Personally, I wish we could use the Root Center for EVERYTHING and be done with this mess.
From: Jeffrey Gomez I agree with Tiger on this one. Center calculations should be standardized - jumping through the hoops to do so presently are a bit convoluted and unneccesary.
Edit: Standardized by default, that is. I completely agree. McW
|
|
Damanios Thetan
looking in
Join date: 6 Mar 2004
Posts: 992
|
04-12-2005 10:48
Completely supported. (And bump, as this breaks about all larger building projects.)
|