plib-users

[Plib-users] Problem with rotations

[Ben W. <be...@ec...>]

Hello Everybody

I ran into a little problem, and I am unable to resolve it. Basically I a=
m=20
just trying to rotate something around its axes. At different times. For =
x=20
and z this works perfectly but for y it does not.=20
The model start with its feet on the origin and rotates around the x axes=
, and=20
it does that but not completely. Once it flat (z =3D 0) then it jumps so =
that=20
the x axes is now going though the middle of it, and its feet are on the =
y=20
axes. It then completes half a rotation and jumps back and completes the=20
orignal rotation with is feet on the x axes.

function rotate (float h, float p, float r) {
  sgMat4 moveMat;
  sgMat4 newMat;
  sgMat4 currentLocation;

  sgCoord newCoord;

  sgMakeRotMat4(moveMat,h,p,r);
  sgMakeCoordMat4(currentLocation, & this->location);
  sgMultMat4(newMat,moveMat, currentLocation);

  sgSetCoord ( &newCoord, newMat ) ;
  this->location.xyz[0] =3D newCoord.xyz[0];
  this->location.xyz[1] =3D newCoord.xyz[1];
  this->location.xyz[2] =3D newCoord.xyz[2];
  this->location.hpr[0] =3D newCoord.hpr[0];
  this->location.hpr[1] =3D newCoord.hpr[1];
  this->location.hpr[2] =3D newCoord.hpr[2];
  this->setTransform( & this->location );
}

Re: [Plib-users] Problem with rotations

[Ben W. <be...@ec...>]

Hello Everybody=20

I have moved the code into tux_example and removed the complexity and I h=
ave=20
the same problem non the less.. Is there a bug releated to pitch.=20

Here is the file
Ben

Re: [Plib-users] Problem with rotations

[Steve B. <sjb...@ai...>]

Ben Woodhead wrote:

> I have moved the code into tux_example and removed the complexity and I have 
> the same problem non the less.. Is there a bug releated to pitch.

Sounds like a classic case of gymbal lock to me.

Recommended reading:

     http://www.sjbaker.org/steve/omniv/eulers_are_evil.html
     http://www.sjbaker.org/steve/omniv/matrices_can_be_your_friends.html


----------------------------- Steve Baker -------------------------------
Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
URLs : http://www.sjbaker.org
        http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
        http://prettypoly.sf.net http://freeglut.sf.net
        http://toobular.sf.net   http://lodestone.sf.net

Re: [Plib-users] Problem with rotations

[Ben W. <za...@ec...>]

I did read these articles but I will read them again. But I did use a matrix
stack from the rotations and I am only rotating around 1 axe. If I change
the code to just rotate around x it works, and if I change it to rotate
around just z there is no problem, but just y causes some really weird
results. Did you have a look at the code..

Thanks, Ben

----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Monday, September 09, 2002 8:55 PM
Subject: Re: [Plib-users] Problem with rotations


> Ben Woodhead wrote:
>
> > I have moved the code into tux_example and removed the complexity and I
have
> > the same problem non the less.. Is there a bug releated to pitch.
>
> Sounds like a classic case of gymbal lock to me.
>
> Recommended reading:
>
>      http://www.sjbaker.org/steve/omniv/eulers_are_evil.html
>      http://www.sjbaker.org/steve/omniv/matrices_can_be_your_friends.html
>
>
> ----------------------------- Steve Baker -------------------------------
> Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
> URLs : http://www.sjbaker.org
>         http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
>         http://prettypoly.sf.net http://freeglut.sf.net
>         http://toobular.sf.net   http://lodestone.sf.net
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

Re: [Plib-users] Problem with rotations

[Steve B. <sjb...@ai...>]

Ben Woodhead wrote:
> I did read these articles but I will read them again. But I did use a matrix
> stack from the rotations and I am only rotating around 1 axe. If I change
> the code to just rotate around x it works, and if I change it to rotate
> around just z there is no problem, but just y causes some really weird
> results. Did you have a look at the code..

Sorry - no - I've been kinda busy.  I've seen *thousands* of cases of
objects rotating nicely around the Y axis - so I'm not inclined to suspect
a bug in PLIB...but anything's possible.

Just to be *sure*, I went into examples/src/ssg/tux/tux_example.cxx - and
changed the line:

   sgSetCoord ( & tuxpos, 0.0f,  0.0f, 0.0f, frameno, 0.0f, 0.0f ) ;

...and switched 'frameno' from Heading (Rotation about Z) to Roll (Rotation about Y):

   sgSetCoord ( & tuxpos, 0.0f,  0.0f, 0.0f, 0.0f, 0.0f, frameno ) ;

...and tux leans to the right, gradually disappears under the pedestal and
reappears on the left...exactly as I'd expect.

----------------------------- Steve Baker -------------------------------
Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
URLs : http://www.sjbaker.org
        http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
        http://prettypoly.sf.net http://freeglut.sf.net
        http://toobular.sf.net   http://lodestone.sf.net

Re: [Plib-users] Problem with rotations

[Ben W. <za...@ec...>]

Hello Steve..

Ya, I don't have a problem with that.. It seems to be sgMakeRotMat(mat,
h,p,r) where pitch is set to something other then 0. Then I converted it
back to sgCoord and used transform, although in my code, I also tried just
using the matrix and it did the same thing. Try coping the file that I sent
over tux_example.cxx and see that happens. If you go back to the code in the
update_motion function and user the header or the role things work fine, but
pitch causes problems. I will keep looking for the problem, but I can't find
it, when I seperated into tux_example (so its not something in my
applicaton) and it only happens with pitch (header and role are fine) then I
don't know what else it could be.

Ben

----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Monday, September 09, 2002 11:51 PM
Subject: Re: [Plib-users] Problem with rotations


> Ben Woodhead wrote:
> > I did read these articles but I will read them again. But I did use a
matrix
> > stack from the rotations and I am only rotating around 1 axe. If I
change
> > the code to just rotate around x it works, and if I change it to rotate
> > around just z there is no problem, but just y causes some really weird
> > results. Did you have a look at the code..
>
> Sorry - no - I've been kinda busy.  I've seen *thousands* of cases of
> objects rotating nicely around the Y axis - so I'm not inclined to suspect
> a bug in PLIB...but anything's possible.
>
> Just to be *sure*, I went into examples/src/ssg/tux/tux_example.cxx - and
> changed the line:
>
>    sgSetCoord ( & tuxpos, 0.0f,  0.0f, 0.0f, frameno, 0.0f, 0.0f ) ;
>
> ...and switched 'frameno' from Heading (Rotation about Z) to Roll
(Rotation about Y):
>
>    sgSetCoord ( & tuxpos, 0.0f,  0.0f, 0.0f, 0.0f, 0.0f, frameno ) ;
>
> ...and tux leans to the right, gradually disappears under the pedestal and
> reappears on the left...exactly as I'd expect.
>
> ----------------------------- Steve Baker -------------------------------
> Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
> URLs : http://www.sjbaker.org
>         http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
>         http://prettypoly.sf.net http://freeglut.sf.net
>         http://toobular.sf.net   http://lodestone.sf.net
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

Re: [Plib-users] Problem with rotations

[Ben W. <za...@ec...>]

Sorry Steve,  its not role I was talking about its pitch (rotating around
the x axes).

Ben

----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Monday, September 09, 2002 11:51 PM
Subject: Re: [Plib-users] Problem with rotations


> Ben Woodhead wrote:
> > I did read these articles but I will read them again. But I did use a
matrix
> > stack from the rotations and I am only rotating around 1 axe. If I
change
> > the code to just rotate around x it works, and if I change it to rotate
> > around just z there is no problem, but just y causes some really weird
> > results. Did you have a look at the code..
>
> Sorry - no - I've been kinda busy.  I've seen *thousands* of cases of
> objects rotating nicely around the Y axis - so I'm not inclined to suspect
> a bug in PLIB...but anything's possible.
>
> Just to be *sure*, I went into examples/src/ssg/tux/tux_example.cxx - and
> changed the line:
>
>    sgSetCoord ( & tuxpos, 0.0f,  0.0f, 0.0f, frameno, 0.0f, 0.0f ) ;
>
> ...and switched 'frameno' from Heading (Rotation about Z) to Roll
(Rotation about Y):
>
>    sgSetCoord ( & tuxpos, 0.0f,  0.0f, 0.0f, 0.0f, 0.0f, frameno ) ;
>
> ...and tux leans to the right, gradually disappears under the pedestal and
> reappears on the left...exactly as I'd expect.
>
> ----------------------------- Steve Baker -------------------------------
> Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
> URLs : http://www.sjbaker.org
>         http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
>         http://prettypoly.sf.net http://freeglut.sf.net
>         http://toobular.sf.net   http://lodestone.sf.net
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

Re: [Plib-users] Problem with rotations

[Steve B. <sjb...@ai...>]

Ben Woodhead wrote:

> Ya, I don't have a problem with that.. It seems to be sgMakeRotMat(mat,
> h,p,r) where pitch is set to something other then 0. Then I converted it
> back to sgCoord and used transform...

Could your confusion be this:

    * You take three angles (H,P,R) and convert them into a matrix M
      (using sgMakeRotMat...or whatever).

    * Then you convert M *BACK* into an sgCoord and extract (H',P',R').

Do you expect (or rely upon) H' == H,  P' == P and R' == R ???

If so, then that's not right.  There are an infinite number of values
for H,P,R that all generate the exact same rotation matrix.  Hence, when
you convert a matrix *BACK* into three Euler angles, you don't get the
same three angles back...although the three angles you get *DO* perform
the same effective rotation as the three you put in.

Hence, it's possible to say things like:

     H =   0
     P = 180
     R =   0

...then convert to a matrix...convert back again and get back:

     H' = 180
     P' =   0
     R' = 180

...totally different numbers - but having the same exact same effect.

If that is your problem, then it's not something that we can fix - it's
just the way math works.

----------------------------- Steve Baker -------------------------------
Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
URLs : http://www.sjbaker.org
        http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
        http://prettypoly.sf.net http://freeglut.sf.net
        http://toobular.sf.net   http://lodestone.sf.net

Re: [Plib-users] Problem with rotations

[Ben W. <za...@ec...>]

That is a good question, and I don't know the answer. But why would it work
for header and role.

h=1
p=0
r=0
this works fine and rotates around. same with
h=0
p=0
r=1
it also works fine. Its just
h=0
p=1
r=0
That causing the problem. And why would it jump to another axe and the come
back.
I did also try this with just a matrix, were I created the rotation matrix
with sgMakeRot4, then muliply it by the location and running the transform
without converting it back to euler, but I got the same result.
The thing I find weird is that its only when it hits one axe that it jumps
to anyother one.
Anyway I don't know, I can't find the problem, and I have a test problem
that does it, but I just don't understand why its doing it, so I can't
really explain.
Ben


----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Tuesday, September 10, 2002 1:47 AM
Subject: Re: [Plib-users] Problem with rotations


>
>
> Ben Woodhead wrote:
>
> > Ya, I don't have a problem with that.. It seems to be sgMakeRotMat(mat,
> > h,p,r) where pitch is set to something other then 0. Then I converted it
> > back to sgCoord and used transform...
>
> Could your confusion be this:
>
>     * You take three angles (H,P,R) and convert them into a matrix M
>       (using sgMakeRotMat...or whatever).
>
>     * Then you convert M *BACK* into an sgCoord and extract (H',P',R').
>
> Do you expect (or rely upon) H' == H,  P' == P and R' == R ???
>
> If so, then that's not right.  There are an infinite number of values
> for H,P,R that all generate the exact same rotation matrix.  Hence, when
> you convert a matrix *BACK* into three Euler angles, you don't get the
> same three angles back...although the three angles you get *DO* perform
> the same effective rotation as the three you put in.
>
> Hence, it's possible to say things like:
>
>      H =   0
>      P = 180
>      R =   0
>
> ...then convert to a matrix...convert back again and get back:
>
>      H' = 180
>      P' =   0
>      R' = 180
>
> ...totally different numbers - but having the same exact same effect.
>
> If that is your problem, then it's not something that we can fix - it's
> just the way math works.
>
> ----------------------------- Steve Baker -------------------------------
> Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
> URLs : http://www.sjbaker.org
>         http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
>         http://prettypoly.sf.net http://freeglut.sf.net
>         http://toobular.sf.net   http://lodestone.sf.net
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

Re: [Plib-users] Problem with rotations

[Ben W. <za...@ec...>]

Ok, I see what you are talking about and it makes sence. But why would this
only cause a problem with pitch. I know that is questionable to return back
from a matrix, so I tried it just using the matrix and not returning it back
and got the same results..

Ben
----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Tuesday, September 10, 2002 1:47 AM
Subject: Re: [Plib-users] Problem with rotations


>
>
> Ben Woodhead wrote:
>
> > Ya, I don't have a problem with that.. It seems to be sgMakeRotMat(mat,
> > h,p,r) where pitch is set to something other then 0. Then I converted it
> > back to sgCoord and used transform...
>
> Could your confusion be this:
>
>     * You take three angles (H,P,R) and convert them into a matrix M
>       (using sgMakeRotMat...or whatever).
>
>     * Then you convert M *BACK* into an sgCoord and extract (H',P',R').
>
> Do you expect (or rely upon) H' == H,  P' == P and R' == R ???
>
> If so, then that's not right.  There are an infinite number of values
> for H,P,R that all generate the exact same rotation matrix.  Hence, when
> you convert a matrix *BACK* into three Euler angles, you don't get the
> same three angles back...although the three angles you get *DO* perform
> the same effective rotation as the three you put in.
>
> Hence, it's possible to say things like:
>
>      H =   0
>      P = 180
>      R =   0
>
> ...then convert to a matrix...convert back again and get back:
>
>      H' = 180
>      P' =   0
>      R' = 180
>
> ...totally different numbers - but having the same exact same effect.
>
> If that is your problem, then it's not something that we can fix - it's
> just the way math works.
>
> ----------------------------- Steve Baker -------------------------------
> Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
> URLs : http://www.sjbaker.org
>         http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
>         http://prettypoly.sf.net http://freeglut.sf.net
>         http://toobular.sf.net   http://lodestone.sf.net
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

Re: [Plib-users] Problem with rotations

[Ben W. <be...@ec...>]

Ok, sorry my fault, it really was converting it back and forth between eu=
ler=20
and matrix. If I keep the matrix the hole time you just see a quick jump =
to a=20
different axe and then back. But this still does not explain why only pit=
ch=20
does this.=20

It seems to be only when pitch hits either 0 or 180 (and no others do thi=
s).=20
It will rotate with its feet around the x axes, but when x hits 0 or 180,=
 it=20
roles so that its feet are on the y axes rotated 90 degrees. If that make=
s=20
any sence.=20

Thanks for your help, has anybody actually tried the code.

Ben

Re: [Plib-users] Problem with rotations results

[Ben W. <be...@ec...>]

Wow, I had no idea the results returning from a matrix were that like tha=
t.. I=20
understood what you meant, when you said h <> h, and so on. But I didn't=20
realise just how off they would actually be.. Anyway back to the point, I=
 was=20
wrong about the 0 and 180, its at 90 degrees that cause the problem. Here=
 is=20
the header, pitch and role from the 3 different rotations. Take a close l=
ook=20
at the pitch.

Header
Header: 88.999992, Pitch: 0.000000, Role: -0.000000
Header: 89.999992, Pitch: 0.000000, Role: -0.000000
Header: 90.999992, Pitch: 0.000000, Role: -0.000000

Pitch
Header: -0.000000, Pitch: 88.002075, Role: -0.000000
Header: -0.000000, Pitch: 89.004166, Role: -0.000000
Header: 0.000000, Pitch: 90.000000, Role: 90.000000

Role
Header: -0.000000, Pitch: 0.000000, Role: 88.999992
Header: -0.000000, Pitch: 0.000000, Role: 89.999992
Header: -0.000000, Pitch: 0.000000, Role: 90.999992

As you can see in pitch when it hits 90 on pitch, role also goes to 90. T=
he=20
one thing that I just noticed is that non of the other rotations get to=20
exactly 90.0.=20

Any ideas, thanks for all your help steve.
Ben

Re: [Plib-users] Problem with rotations results

[Steve B. <sjb...@ai...>]

Ben Woodhead wrote:

> Wow, I had no idea the results returning from a matrix were that like that.. I 
> understood what you meant, when you said h <> h, and so on. But I didn't 
> realise just how off they would actually be.

Yes - well, it's a mathematical fact.  There are an infinite number of ways
to represent an arbitary rotation matrix in terms of Euler angles - and the
code has no way to know which one you wanted - so it just picks one.

It's hard to know what else we could do.

> Anyway back to the point, I was 
> wrong about the 0 and 180, its at 90 degrees that cause the problem. Here is 
> the header, pitch and role from the 3 different rotations. Take a close look 
> at the pitch.

(BTW: The word is "Roll" - not "Role" and "Heading" - not "Header")

> Pitch
> Header: -0.000000, Pitch: 88.002075, Role: -0.000000
> Header: -0.000000, Pitch: 89.004166, Role: -0.000000
> Header: 0.000000, Pitch: 90.000000, Role: 90.000000

Eh?  Are you sure the heading was zero?  I'd expect it to
be -90 or something.

But anyway - this *is* classic 'Gymbal Lock'.

As Norman suggested, you can visualise it in these
terms:

     "A hunter walks South one mile from his tent, then
      he is chased due East for one mile by a ferocious Bear.
      Having escaped, he walks North for one mile and finds
      himself back at his tent.  What colour was the Bear?"

     (White!)

> As you can see in pitch when it hits 90 on pitch, role also goes to 90. The 
> one thing that I just noticed is that non of the other rotations get to 
> exactly 90.0. 

It's not so much that Roll or Pitch is special, it actually comes
about that way because of the order of rotational operations.

The bottom line is that Mat-to-Coord is working OK - it's just
that you have to understand it's limitations.

----------------------------- Steve Baker -------------------------------
Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
URLs : http://www.sjbaker.org
        http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
        http://prettypoly.sf.net http://freeglut.sf.net
        http://toobular.sf.net   http://lodestone.sf.net

Re: [Plib-users] Problem with rotations results

[Curtis L. O. <cu...@me...>]

Steve Baker writes:
>      "A hunter walks South one mile from his tent, then
>       he is chased due East for one mile by a ferocious Bear.
>       Having escaped, he walks North for one mile and finds
>       himself back at his tent.  What colour was the Bear?"

Is this based on magnetic heading or true heading?

>      (White!)

Extra credit:  What colo[u]r were his pants.

Curt.
-- 
Curtis Olson   IVLab / HumanFIRST Program       FlightGear Project
Twin Cities    cu...@me...                  cu...@fl...
Minnesota      http://www.menet.umn.edu/~curt   http://www.flightgear.org

Re: [Plib-users] Problem with rotations results

[Norman V. <nh...@ca...>]

Curtis L. Olson

> Steve Baker writes:
> >      "A hunter walks South one mile from his tent, then
> >       he is chased due East for one mile by a ferocious Bear.
> >       Having escaped, he walks North for one mile and finds
> >       himself back at his tent.  What colour was the Bear?"
> 
> Is this based on magnetic heading or true heading?
> 
> >      (White!)
> 
> Extra credit:  What colo[u]r were his pants.

Same colo[u]r as the Bear !

need-open-water-nearby-to-bath-in'ly yr's

Norman

Re: [Plib-users] Problem with rotations results

[Ben W. <za...@ec...>]

----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Tuesday, September 10, 2002 11:46 PM
Subject: Re: [Plib-users] Problem with rotations results


> Ben Woodhead wrote:
>
> > Wow, I had no idea the results returning from a matrix were that like
that.. I
> > understood what you meant, when you said h <> h, and so on. But I didn't
> > realise just how off they would actually be.
>
> Yes - well, it's a mathematical fact.  There are an infinite number of
ways
> to represent an arbitary rotation matrix in terms of Euler angles - and
the
> code has no way to know which one you wanted - so it just picks one.
>
> It's hard to know what else we could do.
>
> > Anyway back to the point, I was
> > wrong about the 0 and 180, its at 90 degrees that cause the problem.
Here is
> > the header, pitch and role from the 3 different rotations. Take a close
look
> > at the pitch.
>
> (BTW: The word is "Roll" - not "Role" and "Heading" - not "Header")
>
> > Pitch
> > Header: -0.000000, Pitch: 88.002075, Role: -0.000000
> > Header: -0.000000, Pitch: 89.004166, Role: -0.000000
> > Header: 0.000000, Pitch: 90.000000, Role: 90.000000
>
> Eh?  Are you sure the heading was zero?  I'd expect it to
> be -90 or something.

No, that was actually the log from the program running, that I copied into
here.. That was what was causing me the greef.

> But anyway - this *is* classic 'Gymbal Lock'.

Ah, I always thought gymbal lock was only when rotating around all 3 axes at
the same time.

> As Norman suggested, you can visualise it in these
> terms:
>
>      "A hunter walks South one mile from his tent, then
>       he is chased due East for one mile by a ferocious Bear.
>       Having escaped, he walks North for one mile and finds
>       himself back at his tent.  What colour was the Bear?"
>
>      (White!)

What? I don't get it..

> > As you can see in pitch when it hits 90 on pitch, role also goes to 90.
The
> > one thing that I just noticed is that non of the other rotations get to
> > exactly 90.0.
>
> It's not so much that Roll or Pitch is special, it actually comes
> about that way because of the order of rotational operations.
>
> The bottom line is that Mat-to-Coord is working OK - it's just
> that you have to understand it's limitations.

Well, thanks everybody.. I rewrote all the code to have not use coord and
everything is good..
Thanks Ben

> ----------------------------- Steve Baker -------------------------------
> Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
> URLs : http://www.sjbaker.org
>         http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
>         http://prettypoly.sf.net http://freeglut.sf.net
>         http://toobular.sf.net   http://lodestone.sf.net
>
>
>
>
>
> -------------------------------------------------------
> In remembrance
> www.osdn.com/911/
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

Re: [Plib-users] Problem with rotations results

[Norman V. <nh...@ca...>]

Ben Woodhead writes:
>Steve Baker writes

> > But anyway - this *is* classic 'Gymbal Lock'.
>
> Ah, I always thought gymbal lock was only when rotating around all 3 axes
at
> the same time.
>
> > As Norman suggested, you can visualise it in these
> > terms:
> >
> >      "A hunter walks South one mile from his tent, then
> >       he is chased due East for one mile by a ferocious Bear.
> >       Having escaped, he walks North for one mile and finds
> >       himself back at his tent.  What colour was the Bear?"
> >
> >      (White!)
>
> What? I don't get it..

Going back to my original question:
"When EXACTLY at the North Pole how may directions can you move"

Answer:
Only one, the only direction from the North Pole is SOUTH

Norman

Re: [Plib-users] Problem with rotations results

[Steve B. <sjb...@ai...>]

Norman Vine wrote:

> Going back to my original question:
> "When EXACTLY at the North Pole how may directions can you move"
> 
> Answer:
> Only one, the only direction from the North Pole is SOUTH

Yep - if you are a zero-sized person.  Otherwise, you get the
interesting experience of your nose heading south at the same instant
as your backside heading north and one arm going east while the other
goes west...although after you've taken just one step, everything
is more or less sorted out and heading more or less South!

----------------------------- Steve Baker -------------------------------
Mail : <sjb...@ai...>   WorkMail: <sj...@li...>
URLs : http://www.sjbaker.org
        http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
        http://prettypoly.sf.net http://freeglut.sf.net
        http://toobular.sf.net   http://lodestone.sf.net

Re: [Plib-users] Problem with rotations results

[Norman V. <nh...@ca...>]

Steve Baker writes:

> Norman Vine wrote:
> 
> > Going back to my original question:
> > "When EXACTLY at the North Pole how may directions can you move"
> > 
> > Answer:
> > Only one, the only direction from the North Pole is SOUTH
> 
> Yep - if you are a zero-sized person.  

Almost  :-)  < see sig > 

> Otherwise, you get the
> interesting experience of your nose heading south at the same instant
> as your backside heading north and one arm going east while the other
> goes west...although after you've taken just one step, everything
> is more or less sorted out and heading more or less South!

Yep, and you have just 'nicely' described why the 'kludge' of offseting 
the pole position a 'smidge' works :-)

how-small-is-epsilon-anyway'ly yr's

Norman

Re: [Plib-users] Problem with rotations results

[Ben W. <be...@ec...>]

On September 11, 2002 08:19 am, Norman Vine wrote:
> Ben Woodhead writes:
> >Steve Baker writes
> >
> > > But anyway - this *is* classic 'Gymbal Lock'.
> >
> > Ah, I always thought gymbal lock was only when rotating around all 3 =
axes
>
> at
>
> > the same time.
> >
> > > As Norman suggested, you can visualise it in these
> > > terms:
> > >
> > >      "A hunter walks South one mile from his tent, then
> > >       he is chased due East for one mile by a ferocious Bear.
> > >       Having escaped, he walks North for one mile and finds
> > >       himself back at his tent.  What colour was the Bear?"
> > >
> > >      (White!)
> >
> > What? I don't get it..
>
> Going back to my original question:
> "When EXACTLY at the North Pole how may directions can you move"
>
> Answer:
> Only one, the only direction from the North Pole is SOUTH
>
> Norman
>
Hahaha, thats good, I like that.. I just want to let you all know that yo=
u=20
guys are crazy.. joking.. :)

Ben
>
>
>
> -------------------------------------------------------
> In remembrance
> www.osdn.com/911/
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users

Re: [Plib-users] Problem with rotations

[Norman V. <nh...@ca...>]

Ben Woodhead writes:


> Ok, sorry my fault, it really was converting it back and forth between
euler
> and matrix. If I keep the matrix the hole time you just see a quick jump
to a
> different axe and then back. But this still does not explain why only
pitch
> does this.
>
> It seems to be only when pitch hits either 0 or 180 (and no others do
this).
> It will rotate with its feet around the x axes, but when x hits 0 or 180,
it
> roles so that its feet are on the y axes rotated 90 degrees. If that makes
> any sence.

Ben

Another way to think about this is
What direction(s) can you walk in if you are
EXACTLY at one of the Poles ?

< not exacly the same problem but very close >

Norman

Re: [Plib-users] Problem with rotations results

[Christian M. <ma...@Ch...>]

"Curtis L. Olson" wrote:
> 
> Steve Baker writes:
> >      "A hunter walks South one mile from his tent, then
> >       he is chased due East for one mile by a ferocious Bear.
> >       Having escaped, he walks North for one mile and finds
> >       himself back at his tent.  What colour was the Bear?"
> 
> Is this based on magnetic heading or true heading?

Doesn't matter for the question (as long as you are allways refering to
the same type of heading). It only changes the absolute position of the
tent.

But as everybody around here wants a house/flat with a south side front
I should build a house on the same spot where the tent was. Then I can
get rich as all sides re facing south!

CU,
Christian

--
The idea is to die young as late as possible.        -- Ashley Montague

Re: [Plib-users] Problem with rotations results

[David M. <da...@me...>]

Christian Mayer writes:

 > > Is this based on magnetic heading or true heading?
 > 
 > Doesn't matter for the question (as long as you are allways refering to
 > the same type of heading). It only changes the absolute position of the
 > tent.

Sure it does.  You might find a polar bear at the magnetic pole, but I
don't know if it's as likely at the true pole.


All the best,


David

-- 
David Megginson, da...@me..., http://www.megginson.com/

Re: [Plib-users] Problem with rotations results

[Curtis L. O. <cu...@me...>]

David Megginson writes:
> Christian Mayer writes:
> 
>  > > Is this based on magnetic heading or true heading?
>  > 
>  > Doesn't matter for the question (as long as you are allways refering to
>  > the same type of heading). It only changes the absolute position of the
>  > tent.
> 
> Sure it does.  You might find a polar bear at the magnetic pole, but I
> don't know if it's as likely at the true pole.

Right, unless one of them is working for Santa, but in that case he
wouldn't be mean and nasty and wouldn't be chasing defensless hunters
around.

Curt.
-- 
Curtis Olson   IVLab / HumanFIRST Program       FlightGear Project
Twin Cities    cu...@me...                  cu...@fl...
Minnesota      http://www.menet.umn.edu/~curt   http://www.flightgear.org