Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

Fred Beckhusen
This looks perfectly normal to me.  Your avatar always sits on the 'top'
of the prim, no matter where the prim is rotated, as the sit target is
relative to the prim, not the world..

Perhaps you intended the avatar to sit on the 'top' of the rotated prim
from the worldz reference, no matter which way it is rotated?

if so, you subtract the books rotation from the sit target position and
rotation. In rotations, a subtract is a divide.

vector rotacion = <0,0,0> * DEG_TO_RA;
vector posicion = <0.50, 0.0, 0.50> ;

default
{
     state_entry()
     {
         llSitTarget( posicion / llGetRot() , llEuler2Rot(rotacion) /
llGetRot() );
    }
}

Fred Beckhusen/Ferd Frederix

>     1. Strange behavior of the llsittarget when the primitive is
>        rotated 90? in X (Luisillo Contepomi)
>
> Test this please.
> A prim rotate in x axe 90 and then try to modify rotation of sit target.
>
> Use this script in a Cube
> //
>
> vector rotacion = <0,0,0> * DEG_TO_RAD;
> vector posicion = <0.50, 0.0, 0.50> ;
>
> default
> {
>      state_entry()
>      {
>          llSitTarget( posicion, llEuler2Rot(rotacion) );
>     }
> }
>
> Then rotate the cube 90 ? in x axe and edit for try to sit avatar up
> on the cube.
>
> Now Y rotations are Z and Z are Y and positions are bad.
>
> My English is not enough to explain it clearly. Please use the script
> and edit modify in the scripts rotations for sitting on top to see
> what I am saying
>
> Version 0.0.9.1 and all devs.
>
> If have confirmation I will report to mantis.
>
> Best regards,
>
>

_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
Reply | Threaded
Open this post in threaded view
|

Re: Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

Luisillo Contepomi
"as the sit target is relative to the prim, not the world.." This is
the problem. When prim is rotated in X (red axe). Rotation on Y or Z
are bad . Are relative to the world and I think this is not correct
behavior. llSitTarget must be local rotations no world rotations.
When prim is rotated 90 or others degrees in X then Y and Z are world
relative.  I think this is not correct. (But is a big problem because
when solve this I have many thousands of targets that I will have to
relocate.)
El sáb., 1 dic. 2018 a las 23:07, Fred Beckhusen (<[hidden email]>) escribió:

>
> This looks perfectly normal to me.  Your avatar always sits on the 'top'
> of the prim, no matter where the prim is rotated, as the sit target is
> relative to the prim, not the world..
>
> Perhaps you intended the avatar to sit on the 'top' of the rotated prim
> from the worldz reference, no matter which way it is rotated?
>
> if so, you subtract the books rotation from the sit target position and
> rotation. In rotations, a subtract is a divide.
>
> vector rotacion = <0,0,0> * DEG_TO_RA;
> vector posicion = <0.50, 0.0, 0.50> ;
>
> default
> {
>      state_entry()
>      {
>          llSitTarget( posicion / llGetRot() , llEuler2Rot(rotacion) /
> llGetRot() );
>     }
> }
>
> Fred Beckhusen/Ferd Frederix
> >     1. Strange behavior of the llsittarget when the primitive is
> >        rotated 90? in X (Luisillo Contepomi)
> >
> > Test this please.
> > A prim rotate in x axe 90 and then try to modify rotation of sit target.
> >
> > Use this script in a Cube
> > //
> >
> > vector rotacion = <0,0,0> * DEG_TO_RAD;
> > vector posicion = <0.50, 0.0, 0.50> ;
> >
> > default
> > {
> >      state_entry()
> >      {
> >          llSitTarget( posicion, llEuler2Rot(rotacion) );
> >     }
> > }
> >
> > Then rotate the cube 90 ? in x axe and edit for try to sit avatar up
> > on the cube.
> >
> > Now Y rotations are Z and Z are Y and positions are bad.
> >
> > My English is not enough to explain it clearly. Please use the script
> > and edit modify in the scripts rotations for sitting on top to see
> > what I am saying
> >
> > Version 0.0.9.1 and all devs.
> >
> > If have confirmation I will report to mantis.
> >
> > Best regards,
> >
> >
>
> _______________________________________________
> Opensim-users mailing list
> [hidden email]
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
Reply | Threaded
Open this post in threaded view
|

Re: Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

Ethan Gardener
On Sun, Dec 2, 2018, at 10:47 AM, Luisillo Contepomi wrote:
> "as the sit target is relative to the prim, not the world.." This is
> the problem. When prim is rotated in X (red axe). Rotation on Y or Z
> are bad . Are relative to the world and I think this is not correct
> behavior. llSitTarget must be local rotations no world rotations.
> When prim is rotated 90 or others degrees in X then Y and Z are world
> relative.  I think this is not correct. (But is a big problem because
> when solve this I have many thousands of targets that I will have to
> relocate.)

Looks local to me; see attachment.  Prim rotation here is, in degrees:
<45.00000, 0.00000, 0.00000>

_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
Reply | Threaded
Open this post in threaded view
|

Re: Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

Ethan Gardener
Great!  The list removes attachments (It's only 45KB!) and my idiot ISP MitMs imgur so badly it won't load at all.  And I don't have a website right now.

On Sun, Dec 2, 2018, at 11:09 AM, Ethan Gardener wrote:

> On Sun, Dec 2, 2018, at 10:47 AM, Luisillo Contepomi wrote:
> > "as the sit target is relative to the prim, not the world.." This is
> > the problem. When prim is rotated in X (red axe). Rotation on Y or Z
> > are bad . Are relative to the world and I think this is not correct
> > behavior. llSitTarget must be local rotations no world rotations.
> > When prim is rotated 90 or others degrees in X then Y and Z are world
> > relative.  I think this is not correct. (But is a big problem because
> > when solve this I have many thousands of targets that I will have to
> > relocate.)
>
> Looks local to me; see attachment.  Prim rotation here is, in degrees:
> <45.00000, 0.00000, 0.00000>
> _______________________________________________
> Opensim-users mailing list
> [hidden email]
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users


--
Progress might have been all right once, but it has gone on too long -- Ogden Nash
_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
Reply | Threaded
Open this post in threaded view
|

Re: Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

CiderJack
In reply to this post by Luisillo Contepomi
I may be missing something, but it reads to me like you have two
complaints relating to prim rotation?

1) When rotating a prim that has an avatar 'sitting' on it, the avatar
follows the rotation of the prim, rather than maintaining it's position
relative to the world.

2) When a prim is rotated, the XYZ axes rotate to remain faithful
relative to the prim, not the world.

Either way, what you describe sounds like normal/correct behavior. It
has been like this since the beginning, and is the same in Second Life.

Doing it like this is really the only way that makes sense. If the
avatar didn't follow the prim movement, things would get very difficult
very quickly (think vehicles) and others would be downright impossible.
The same with the prim maintaining it's own axes relative to it - prims
would be impossible to manipulate if the axes remained rigid relative to
the world.

It was planned to work this way intentionally and is not a problem.
Perhaps this might help:
http://wiki.secondlife.com/wiki/Rotation

Cheers

---
~!CJ



On 2018-12-02 23:47, Luisillo Contepomi wrote:

> "as the sit target is relative to the prim, not the world.." This is
> the problem. When prim is rotated in X (red axe). Rotation on Y or Z
> are bad . Are relative to the world and I think this is not correct
> behavior. llSitTarget must be local rotations no world rotations.
> When prim is rotated 90 or others degrees in X then Y and Z are world
> relative.  I think this is not correct. (But is a big problem because
> when solve this I have many thousands of targets that I will have to
> relocate.)
> El sáb., 1 dic. 2018 a las 23:07, Fred Beckhusen (<[hidden email]>)
> escribió:
>>
>> This looks perfectly normal to me.  Your avatar always sits on the
>> 'top'
>> of the prim, no matter where the prim is rotated, as the sit target is
>> relative to the prim, not the world..
>>
>> Perhaps you intended the avatar to sit on the 'top' of the rotated
>> prim
>> from the worldz reference, no matter which way it is rotated?
>>
>> if so, you subtract the books rotation from the sit target position
>> and
>> rotation. In rotations, a subtract is a divide.
>>
>> vector rotacion = <0,0,0> * DEG_TO_RA;
>> vector posicion = <0.50, 0.0, 0.50> ;
>>
>> default
>> {
>>      state_entry()
>>      {
>>          llSitTarget( posicion / llGetRot() , llEuler2Rot(rotacion) /
>> llGetRot() );
>>     }
>> }
>>
>> Fred Beckhusen/Ferd Frederix
>> >     1. Strange behavior of the llsittarget when the primitive is
>> >        rotated 90? in X (Luisillo Contepomi)
>> >
>> > Test this please.
>> > A prim rotate in x axe 90 and then try to modify rotation of sit target.
>> >
>> > Use this script in a Cube
>> > //
>> >
>> > vector rotacion = <0,0,0> * DEG_TO_RAD;
>> > vector posicion = <0.50, 0.0, 0.50> ;
>> >
>> > default
>> > {
>> >      state_entry()
>> >      {
>> >          llSitTarget( posicion, llEuler2Rot(rotacion) );
>> >     }
>> > }
>> >
>> > Then rotate the cube 90 ? in x axe and edit for try to sit avatar up
>> > on the cube.
>> >
>> > Now Y rotations are Z and Z are Y and positions are bad.
>> >
>> > My English is not enough to explain it clearly. Please use the script
>> > and edit modify in the scripts rotations for sitting on top to see
>> > what I am saying
>> >
>> > Version 0.0.9.1 and all devs.
>> >
>> > If have confirmation I will report to mantis.
>> >
>> > Best regards,
>> >
>> >
>>
>> _______________________________________________
>> Opensim-users mailing list
>> [hidden email]
>> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
> _______________________________________________
> Opensim-users mailing list
> [hidden email]
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
Reply | Threaded
Open this post in threaded view
|

Re: Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

Luisillo Contepomi
In reply to this post by Ethan Gardener
https://www.youtube.com/watch?v=8uEyPf8Gydk best with a video
El dom., 2 dic. 2018 a las 12:09, Ethan Gardener
(<[hidden email]>) escribió:

>
> On Sun, Dec 2, 2018, at 10:47 AM, Luisillo Contepomi wrote:
> > "as the sit target is relative to the prim, not the world.." This is
> > the problem. When prim is rotated in X (red axe). Rotation on Y or Z
> > are bad . Are relative to the world and I think this is not correct
> > behavior. llSitTarget must be local rotations no world rotations.
> > When prim is rotated 90 or others degrees in X then Y and Z are world
> > relative.  I think this is not correct. (But is a big problem because
> > when solve this I have many thousands of targets that I will have to
> > relocate.)
>
> Looks local to me; see attachment.  Prim rotation here is, in degrees:
> <45.00000, 0.00000, 0.00000>
> _______________________________________________
> Opensim-users mailing list
> [hidden email]
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
Reply | Threaded
Open this post in threaded view
|

Re: Strange behavior of the llsittarget when the primitive is rotated 90? in X (Ferd Frederix)

AJLDuarte
Euler angles (or rotations) can be confusing.

< a,b,c> on our and sl convention means

rotate around Z by c;

then rotate the result around Y by b

then rotate the result around X by a

applied exactly by this order.

if you look your numbers considering this, it may make more sense

Ubit


On 02-Dec-18 12:34, Luisillo Contepomi wrote:

> https://www.youtube.com/watch?v=8uEyPf8Gydk best with a video
> El dom., 2 dic. 2018 a las 12:09, Ethan Gardener
> (<[hidden email]>) escribió:
>> On Sun, Dec 2, 2018, at 10:47 AM, Luisillo Contepomi wrote:
>>> "as the sit target is relative to the prim, not the world.." This is
>>> the problem. When prim is rotated in X (red axe). Rotation on Y or Z
>>> are bad . Are relative to the world and I think this is not correct
>>> behavior. llSitTarget must be local rotations no world rotations.
>>> When prim is rotated 90 or others degrees in X then Y and Z are world
>>> relative.  I think this is not correct. (But is a big problem because
>>> when solve this I have many thousands of targets that I will have to
>>> relocate.)
>> Looks local to me; see attachment.  Prim rotation here is, in degrees:
>> <45.00000, 0.00000, 0.00000>
>> _______________________________________________
>> Opensim-users mailing list
>> [hidden email]
>> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
> _______________________________________________
> Opensim-users mailing list
> [hidden email]
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
_______________________________________________
Opensim-users mailing list
[hidden email]
http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users