 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here is the image exhibiting the unexpected (but not unwelcome) focal
blur effect. Normally, I would expect a circular region of clarity with
blurring around it. In this case, there is a horizontal swath of "in
focus" area, with the blurring above and below:
Post a reply to this message
Attachments:
Download 'oddblur.jpg' (44 KB)
Preview of image 'oddblur.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
in news:3D64EA58.40E092FC@aol.com Dave Dunn wrote:
> Normally, I would expect a circular region of clarity with
> blurring around it.
The plane of focus is always parallel to the "camera-plane" (unless you
sheared the camera). So you will never have a circular "spot" of focus,
but always something linear / planar.
Ingo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
You have a "line" of focus because this represents the locus of equidistant
objects (surfaces?) from the camera when you have a plane slanting away from
you.
If instead you had say a uniform field of particles (say an isosurface of
f_noise3d), then you would be right to expect a circular area to be in
focus, as these would be equidistant from camera.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Greg M. Johnson" wrote:
> If instead you had say a uniform field of particles (say an isosurface of
> f_noise3d), then you would be right to expect a circular area to be in
> focus, as these would be equidistant from camera.
Quite acceptable if there were a plane tilting away from me. The object covered
with sand particles (440k, count 'em) is an <ahem> f_noise3d isosurface ; }
Dave
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I've seen other examples of this same behavior in the past. The
documentation (6.4.3) indicates the blur should radiate from a focal point.
It says nothing about a "plane" of focus or it being linear. I think it is a
bug in Povray's focal blur.
"Dave Dunn" <poi### [at] aol com> wrote in message
news:3D64EA58.40E092FC@aol.com...
> Here is the image exhibiting the unexpected (but not unwelcome) focal
> blur effect. Normally, I would expect a circular region of clarity with
> blurring around it. In this case, there is a horizontal swath of "in
> focus" area, with the blurring above and below:
>
>
>
----------------------------------------------------------------------------
----
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> I've seen other examples of this same behavior in the past. The
> documentation (6.4.3) indicates the blur should radiate from a focal
point.
> It says nothing about a "plane" of focus or it being linear. I think it is
a
> bug in Povray's focal blur.
No. The area in focus should be the space near the surface of a sphere
around the camera, with a radius of the distance between the camera location
and the focal point. That is, all points that have the same distance from
the camera as the focal point does should be in focus.
The docs may be misleading here.
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Slime wrote:
> No. The area in focus should be the space near the surface of a sphere
> around the camera, with a radius of the distance between the camera location
> and the focal point. That is, all points that have the same distance from
> the camera as the focal point does should be in focus.
Well, that makes sense then, and reinforces my original idea that the cause of
my particular odd blurring was that my look_at and my focal_point were two
different vectors. If they are the same, this imaginary "sphere" around the
camera will intersect at one point, i.e. where look_at/focal_point coincide. If
they are different, there should be an arc shaped ring between the f_p, and the
l_a, which is exactly what I experienced.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
in news:3d6521b6$1@news.povray.org Patrick Dugan wrote:
> The
> documentation (6.4.3) indicates the blur should radiate from a focal
> point.
"The center of the zone of sharpness is specified by the focal_point
vector. Objects close to this point are in focus and those farther from
that point are more blurred. The default value is focal_point<0,0,0>."
Is this the part that makes you think that the area of focus is
circular? If so it probably needs to be re-phrased.
>It says nothing about a "plane" of focus or it being linear. I
> think it is a bug in Povray's focal blur.
>
The secene below tries to illustrate that it is a plane of focus, just
like in the real world.
---%<------%<---
#version 3.5;
global_settings {assumed_gamma 1.0}
camera {
location <3.0, 0.0, -3.0>
look_at <0.0, 0.0, 0.0>
right x*image_width/image_height
aperture 20.0
blur_samples 50
focal_point <0,0,0>
confidence 0.99
variance 0.001
}
light_source {<500,500,-500>, rgb 1}
plane {y, -1 pigment { checker}}
sphere {
0.0, 0.5
pigment {rgb 1}
}
---%<------%<---
Ingo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Dave Dunn" <poi### [at] aolcom> skrev i melding
news:3D64EA58.40E092FC@aol.com...
> Here is the image exhibiting the unexpected (but not unwelcome) focal
> blur effect. Normally, I would expect a circular region of clarity with
> blurring around it. In this case, there is a horizontal swath of "in
> focus" area, with the blurring above and below:
I thought it worked that way too, and asked same question a few months back.
Although you already got some answers, I suggest reading the tread here:
http://news.povray.org/povray.binaries.images/23677/
I also suggest (as Peter Popov suggested me in the tread above) you try out
DoF.inc, it really helps a lot when setting focal point and aperture. You
can find it here:
http://www.wozzeck.net/images/DoF.html
Good luck!
--
_.-=^=-._.-=[ Peter ]=-._.-=^=-._
_.-=[ http://hertel.no/peter ]=-._
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Well, that makes sense then, and reinforces my original idea that the
cause of
> my particular odd blurring was that my look_at and my focal_point were two
> different vectors. If they are the same, this imaginary "sphere" around
the
> camera will intersect at one point, i.e. where look_at/focal_point
coincide. If
> they are different, there should be an arc shaped ring between the f_p,
and the
> l_a, which is exactly what I experienced.
Hmm? No, the look_at point has *absolutely nothing* to do with focal blur.
If you have a focal point 2 units away from the camera, and a look_at point
2000 units away from the camera, then everything 2 units away from the
camera will be in focus.
The arc is caused by the intersection of the surface of the sphere with the
surface of the plane.
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
