 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
(RC5, Redhat Linux, 64MB, 300MHz, why does this matter?)
The camera definition does not behave how I think it should.
The following is a minimal scene.
The two cameras should be equivalent, but are not.
They were back in 3.1g.
For once it is NOT a problem with #declared cameras.
Mark Weyer
light_source {
<-1,2,-1>*100
color rgb <1,1,1>
}
box {
<3,-1,-10> <4,1,10>
texture {pigment {rgb <1,1,1>}}
}
#declare Cam1 = camera {
location <0, 0, 0>
look_at <1, 0, 0>
right -y
up -z
}
#declare Cam2 = camera {
location <0, 0, 0>
look_at <0, 0, 1>
right x
up y
rotate -90*x
rotate -90*z
}
camera {Cam1}
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> (RC5, Redhat Linux, 64MB, 300MHz, why does this matter?)
(I suspect it rarely does, but i'm sure it's frustrating to the team to not
have that information when it *is* necessary.)
> The camera definition does not behave how I think it should.
> The following is a minimal scene.
> The two cameras should be equivalent, but are not.
> They were back in 3.1g.
> For once it is NOT a problem with #declared cameras.
I have to point out that you're putting look_at before specifying your right
and up vectors; I was under the impression that this was illegal in 3.5.
Then again, if it's illegal, then it's a bug that POV-Ray is letting you do
it without giving you an error, I'd think.
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3CF72E3C.7E90042A@frege.mathematik.uni-freiburg.de> , Mark Weyer
<wey### [at] frege mathematikuni-freiburgde> wrote:
> The two cameras should be equivalent, but are not.
No, they should not be equivalent. If you use look_at in the 3.5 camera (in
short) only their length is considered. This behavior is similar to placing
the look_at after the up and right in the 3.1 camera model. If you absolutely
want to change vectors look_at adjusted for you, switch the version to 3.1
when declaring the camera to get the old behavior.
Thorsten
____________________________________________________
Thorsten Froehlich
e-mail: mac### [at] povrayorg
I am a member of the POV-Ray Team.
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3cf79bd2@news.povray.org> , "Slime" <slm### [at] slimelandcom> wrote:
> I have to point out that you're putting look_at before specifying your right
> and up vectors; I was under the impression that this was illegal in 3.5.
There is nothing illegal here. The docs clearly explain that the order
dependencies have been removed in the 3.5 camera. Any oder will yield exactly
the same effect.
Thorsten
____________________________________________________
Thorsten Froehlich
e-mail: mac### [at] povrayorg
I am a member of the POV-Ray Team.
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> There is nothing illegal here. The docs clearly explain that the order
> dependencies have been removed in the 3.5 camera. Any oder will yield
exactly
> the same effect.
Oooooh, I get it. So, even though you can specify them in any order, it
applies them in the correct order?
I was thinking, that since order no longer changed anything, it was illegal
to specify them in an order different from which they were applied.
Hmm... order no longer matters. Then you might want to change the error for
non-perpendicular camera vectors, which says that "making look_at the last
statement might help."
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3cf7dfed@news.povray.org> , "Slime" <slm### [at] slimelandcom> wrote:
> Hmm... order no longer matters. Then you might want to change the error for
> non-perpendicular camera vectors, which says that "making look_at the last
> statement might help."
As look_at is always applied after up/right and look_at forces perpendicular
vectors there no longer is such a problem...
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> As look_at is always applied after up/right and look_at forces
perpendicular
> vectors there no longer is such a problem...
There is if you apply a matrix transformation after the look_at vector that
shears the camera. (I know of this since it's an issue with my zoomin.inc
file)
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3cf90440@news.povray.org> , "Slime" <slm### [at] slimelandcom> wrote:
> There is if you apply a matrix transformation after the look_at vector that
> shears the camera. (I know of this since it's an issue with my zoomin.inc
> file)
This was not my point. It is a very bad idea to apply look_at after the
transformations as it would undo them and thus most transformations would be
pointless to begin with...
Thorsten
____________________________________________________
Thorsten Froehlich
e-mail: mac### [at] povrayorg
I am a member of the POV-Ray Team.
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thorsten Froehlich wrote:
> "Slime" <slm### [at] slimelandcom> wrote:
>
>> There is if you apply a matrix transformation
>> after the look_at vector that shears the camera.
>
> This was not my point. It is a very bad idea to
> apply look_at after the transformations as it
> would undo them and thus most transformations
> would be pointless to begin with...
Thorsten, did you even read what Slime wrote?
He said that if a matrix transformation is applied *after* the look_at
vector, the error message says that it might help to make the look_at
the last statement. This error message does not make sense with the
POV-Ray 3.5 camera.
Minimal test scene that produces the error:
camera {
location 0
look_at z
matrix <1,0,0,0,1,0,1,0,1,0,0,0>
}
// insert 25 speres to turn on vista buffer
#declare C = 0;
#while (C<25)
sphere {z,0.1 pigment {rgb 1}}
#declare C = C+1;
#end
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated May 20)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3cf9180e@news.povray.org> , "Rune" <run### [at] mobilixnetdk>
wrote:
> Thorsten, did you even read what Slime wrote?
>
> He said that if a matrix transformation is applied *after* the look_at
> vector, the error message says that it might help to make the look_at
> the last statement. This error message does not make sense with the
> POV-Ray 3.5 camera.
No, but the warning itself does.
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
