rotations being funny

alrighty, please see the attached picture.

here's the script for spawning the red, blue, and green bits

      llRezObject("red 2", llGetPos() + <-0.0, -0.0, -0.0>, ZERO_VECTOR, llGetRot() + <0,0,0,0>, 0);
      
      llRezObject("blue 2", llGetPos() + <-0.0, -0.0, -0.0>, ZERO_VECTOR, llGetRot() + quat2, 0);
       
      llRezObject("green 2", llGetPos() + <-0.0, -0.0, -0.0>, ZERO_VECTOR, llGetRot() + quat, 0);
       	 
       

     vector eul; //x degrees around the z-axis, in Euler form
     vector eul2;
     rotation quat;
     rotation quat2;
     //other irrelivant code in here
    
        eul = <0,0,90> * DEG_TO_RAD;
      quat = llEuler2Rot(eul);
      eul2 = <0,0,180> * DEG_TO_RAD;
      quat2 = llEuler2Rot(eul2);  
       

I am certain the second block of code is being called.

green is spawned at 90degrees as you can see.

but the blue screen is certainly NOT 90degrees more rotated. (it should be at 180 as you can see). setting all the angle offsets to 0 spawns them all in the same place, so it's not that...

what on earth is screwing with my angles? they're not altered or even referenced anywhere else in the script....
(eventually when I get the angles right, they will all line up with the bottom most brick looking thingy)
mucho thanks for reading

I don't think that you can get a good result by adding rotation each other. Use multiplying.

And when you DO multiply the rotations (instead of adding them), be aware that <0,0,0,0> is NOT the ZERO_ROTATION.

thanks for the tips guys. I've not had any problems using llGetRot() + someconstant with in the other things I've built... it's a very odd!


does adding <a, b, c, d> + <e, f, g, h> not work well in SL then?


edit: changed the + for a * and it works now thanks guys!

still not sure quite why though

Because Quadernions are weird!

I had thought that might be the case!

I'm sure you already have, but if not, be sure to read the Wiki page on rotations / quaternions: http://www.lslwiki.net/lslwiki/wakka.php?wakka=rotation

I've read that page over and over, for months now. It's finally starting to sink in.

Pay special attention to the part that talks about the ORDER of quaternion multiplication, which results in apparent rotations based on WORLD vs. LOCAL axes. That can be very helpful.

To combine rotations you have to multiply them (or divide). If you add them it just combines the fields, not calculate the combined rotation.

You may also want to use llRezAtRoot instead.

thanks for that, I'll have to give it a look over again when I get confused! seems like a mighty odd way to do things though... perhaps it's standard practice in 3D graphics, but I'd not heard that it was!

also, I've used llRezAtRoot for some of the other bits, but the the parent of the rez object isn't in a helpful place for this one

's all working and in position now