sit target height adj

[Xiija]

 

I cannot get my mind around the math.

i'm doing

  float x = ( avi.z/2 ) + (root.z / 2);

but it ends up either too low or too high.

should i be using a percent of the avi height or something?

[Innula Zenovka]

4 minutes ago, Xiija said:

 

I cannot get my mind around the math.

i'm doing

  float x = ( avi.z/2 ) + (root.z / 2);

but it ends up either too low or too high.

should i be using a percent of the avi height or something?

Depends what you're trying to calculate, of course, but if you want the position of the avatar's feet, I think the calculation is 

vector vAvSize = llGetAgentSize(kAv);

float f = -1*((vAvSize.z*1.1)/2);

(and then add f to the avatar's z coordinate, of course).    That's what I use for traps and things, anyway.

[Rolig Loon]

Pay close attention to the caveats in the wiki discussion of llGetAgentSize.  The returned value may not be what you expect.  I don't think it will make much difference how you calculate it.  You can hope that standard- size avatars will need to be offset by almost the same amount, but non-standard shapes may vary quite a bit.

EDIT: I have used a method like Innula's, but it's really a crap shoot. The 1.1 factor is one that I've seen in scripts for a very long time.  It seems to be an empirical fudge factor that approximates the difference between the value reported by llGetAgentSize (which does not include the head, for some reason I have forgotten -- the bounding box, maybe?) and the avatar's real height (which does include the head).  That's close for standard-sized avatars, but not for everyone.

Edited September 13, 2021 by Rolig Loon

[Innula Zenovka]

5 hours ago, Rolig Loon said:

Pay close attention to the caveats in the wiki discussion of llGetAgentSize.  The returned value may not be what you expect.  I don't think it will make much difference how you calculate it.  You can hope that standard- size avatars will need to be offset by almost the same amount, but non-standard shapes may vary quite a bit.

EDIT: I have used a method like Innula's, but it's really a crap shoot. The 1.1 factor is one that I've seen in scripts for a very long time.  It seems to be an empirical fudge factor that approximates the difference between the value reported by llGetAgentSize (which does not include the head, for some reason I have forgotten -- the bounding box, maybe?) and the avatar's real height (which does include the head).  That's close for standard-sized avatars, but not for everyone.

The other option, I guess, would be a ray cast to see what's directly underneath the avatar.    Hard to go wrong with that, I'd have thought.

[Fenix Eldritch]

7 hours ago, Rolig Loon said:

llGetAgentSize (which does not include the head, for some reason I have forgotten -- the bounding box, maybe?)

Yup, the function returns the size of the bounding box for the avatar's physics representation - which terminates at eye level and is where the camera is positioned in mouselook.

Edited September 14, 2021 by Fenix Eldritch
clarification