 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
hello pov.monde
In the doc:
http://www.povray.org/working-docs/id000158.html#6_4_1_3
was it possible to add the default angle value used if no
angle given ?
--
Thierry, (42++)++ ::: http://la.buvette.org/exPOV/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Thierry B." <oul### [at] chez com> wrote in message
news:3BCBF089.CBB50DBB@chez.com...
> hello pov.monde
>
> In the doc:
> http://www.povray.org/working-docs/id000158.html#6_4_1_3
>
> was it possible to add the default angle value used if no
> angle given ?
If you are asking about what the default is or if it should be added to the
documentation I have only found it to be approximately angle 67 for the
perspective projection camera. When I try to find that according to the
formula mentioned in that section, "direction_length = 0.5 * right_length /
tan(angle / 2) where right_length is the length of the right vector", I come
up with about 'direction 0.56' which is just more than half of what it ought
to be. I just don't know if I'm figuring that correctly. Although I have
compared a direction 1 to angle 67 and managed to see nearly an equivalence.
I haven't checked again recently to see if it remains that way, but it might
be okay to add that angle defaults to approx. 67 in the documentation if it
is so. Either that or someone needs to provide an exact number for it I
guess.
Bob H.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> When I try to find that according to the
> formula mentioned in that section, "direction_length = 0.5 * right_length
/
> tan(angle / 2) where right_length is the length of the right vector", I
come
> up with about 'direction 0.56' which is just more than half of what it
ought
> to be.
Really? I come up with 1.00722346241. The exact value for angle is 2 atan
(2/3), or about 67.38013502.
Anders
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Speaking of this, in doc sections 6.4.1.5.1 and 6.4.1.5.2, it says that the
default value for right is <1.33, 0, 0>. Shouldn't this be <4/3, 0, 0>, as
stated in section 6.4.1.5?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Anders K." <and### [at] f2scom> wrote in message
news:3bcc5632$1@news.povray.org...
> Speaking of this, in doc sections 6.4.1.5.1 and 6.4.1.5.2, it says that the
> default value for right is <1.33, 0, 0>. Shouldn't this be <4/3, 0, 0>, as
> stated in section 6.4.1.5?
Hmm, AFAIK 4/3 cannot be expressed accurately in either dec. or bin. - both
therefore look like approximations - or did you just mean for consistency's
sake?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Tom Melly wrote:
>
> Hmm, AFAIK 4/3 cannot be expressed accurately in either dec. or bin. - both
> therefore look like approximations - or did you just mean for consistency's
> sake?
To clarify things, the default right vector is 1.33*x so this is what the
docs should say too IMO.
Christoph
--
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Anders K." <and### [at] f2scom> wrote in message
news:3bcc53ab$1@news.povray.org...
> > When I try to find that according to the
> > formula mentioned in that section, "direction_length = 0.5 *
right_length
> /
> > tan(angle / 2) where right_length is the length of the right vector", I
> come
> > up with about 'direction 0.56' which is just more than half of what it
> ought
> > to be.
>
> Really? I come up with 1.00722346241. The exact value for angle is 2 atan
> (2/3), or about 67.38013502.
Right. Well, apparently tan 67/2 isn't a good way to go about calculating
it. I left off the parentheses in my quick attempt :-)
Bob H.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> To clarify things, the default right vector is 1.33*x so this is what the
> docs should say too IMO.
If the default right vector is really 1.33*x and not 4/3*x then that is a
program bug (or design flaw, or whatever), since the aspect ratio of the
images it normally produces is *exactly* 4 to 3 and not 1.33 to 1.
Of course, I realize that 1.33 is an "approximation" to 4/3, but IMO it's a
pretty bad approximation, especially for large images. And in this case
there is no need to approximate at all (other than the internal floating
point rounding error, which is going to be much, much less than .01). So I
think that both the program and the docs should use 4/3.
Anders
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
in news:3BCC633C.FFF910D0@gmx.de Christoph Hormann wrote:
> To clarify things, the default right vector is 1.33*x so this is
> what the docs should say too IMO.
If this value comes from the source, I'll match the documentation.
Ingo
--
Photography: http://members.home.nl/ingoogni/
Pov-Ray : http://members.home.nl/seed7/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
ingo wrote:
>
> If this value comes from the source, I'll match the documentation.
>
It does (in Povray 3.1)
Christoph
--
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
