window?

Post a reply to this message

Apache wrote:
>window?
>

Or maybe the door on the left (look carefully) ;)


John

Post a reply to this message

> Or maybe the door on the left (look carefully) ;)

Thanks for saying this ;)

-Ive

Post a reply to this message

> More interesting than this would be to have your LogLuv output patch.

More interesting than what?

> Do you use the libtiff library?

Sure, its already there. And thanks to Greg Ward who did contribute
the code for SGI tiff log-luv to libtif.

>Could you publish you patch?

It was made for this purpose. But it was only a one night quick and
brute hack. So when enabling the LogLuv file output, the continue
trace option and the render window does not work, maybe there are
more side effect. Nothing is yet documented and I did not even read
the guidelines for patching pov. I guess there are some.
So I do not want to become crucified for my dirty hacking style, but
have currently no time to work on this.
In fact, I guess the log-luv reading part is done very similar to the patch
by Mael (even more simple because most of the work is done by libtif)
and for the writing part only all color clipping had to be skipped, gamma
correction had to be disabled and a kind of exposure control (I have not
installed mega-pov, but have read about it, so I guess it works the same
way) For the file writing itself does again libtif most of the work.
Thats all.

-Ive

Post a reply to this message

Ive wrote:
> 
> > More interesting than this would be to have your LogLuv output patch.
> 
> More interesting than what?
> 

More interesting than the image :-)

Don't get me wrong, the image looks good but being able to generate one's
own HDR maps would be more useful.

> >Could you publish you patch?
> 
> It was made for this purpose. But it was only a one night quick and
> brute hack. So when enabling the LogLuv file output, the continue
> trace option and the render window does not work, maybe there are
> more side effect. Nothing is yet documented and I did not even read
> the guidelines for patching pov. I guess there are some.
> So I do not want to become crucified for my dirty hacking style, but
> have currently no time to work on this.

You can send it to me if you want - someone from the Megapov Team might
find the time to work on it so it can be included in Megapov.  

If continued trace does not work i assume you have not yet implemented the
Read_Line() method.

Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 28 Feb. 2003 _____./\/^>_*_<^\/\.______

Post a reply to this message

Ive and folks:

Here is what it looked like when I tried it.

Aaron

Aaron Gillies
New York City
x3rxes[&]yahoo.com




"Ive" <ive### [at] lilysoftcom> wrote in message news:3e5f5bfe@news.povray.org...
> Hi,
>
> as already mentioned some time ago, I have done a patched version
> of PoV that does not only read but also write high dynamic range
> images. This patch does not work with the radiance HDR images
> used by MLPov but uses the superior LogLuv Tiff format, that
> provides both, a 'higher' high dynamic range and does also cover the
> full range of the visible gamut (radiance files do not!).
>
> Anyway, here is the jpeg preview of  my simplified italian room scene -
> rendered as LogLuv HDRI and for your convenience I have already
> converted the original image to the Radiance file format to be easily
> used with MLPov as a global illumination map.
>
> I am definitely not sure if I have used good settings to render this hdr-
> image (there are some more values to adjust for this image type) and
> as you can see, the used radiosity settings where also quite poor.
> But this should not matter when used just as a lighting map (well hmm,
> I hope so). But for the moment, I do not have the time (and mood) to
> develope this hdr-thing any further - so here is the link to download this
> radiance HDR image:
>
> http://www.lilysoft.com/gallery/download/room-luv.hdr
>
> feel free to use it for yout own scene (in case you are bored by using
> the kitchen-probe again) and note thas this image is already rendered
> by using the spherical camera (there is definitely no need to simulate
> the chrome ball as in photographic light probe images) so you should
> use map_type 1 (and not 7 !!!) to map the image on your sphere.
>
> And well, just in case anybody uses it, I would be glad to have a look
> at the resulting image and would like to know if it works in fact similar
> to the photographed probes.
>
> -Ive
>
>
>

Post a reply to this message

Attachments:
Download 'room.JPG' (154 KB)

Preview of image 'room.JPG'

Ive:

Here is another one.  You know, your HDR comes through as being
really dark, much darker than the reference probes.  I had to turn the
ambience up in order to get just the illumination that you see here.  Any
idea why that might be?

Aaron

Aaron Gillies
New York City
x3rxes[^]yahoo.com

Post a reply to this message

Attachments:
Download 'glass_02.JPG' (142 KB)

Preview of image 'glass_02.JPG'

> Here is another one.  You know, your HDR comes through as being
> really dark, much darker than the reference probes.  I had to turn the
> ambience up in order to get just the illumination that you see here.  Any
> idea why that might be?
>

Well, this was my first experiment with a more complicated HDR scene
than just a box. And it seems that producing good HDR images with
PoV requires some experience than I already have.
The original LogLuv image did also store a so called STONITS information.
This can be used to calculated the physical brightness for each pixel in
candelas/squaremeter but this info was lost when transformating the image
to the radiance HDR format. But I'm not even sure if the stonits was
calculated correct. And as this image with the 'poor' radiosity settings
(samples 500, error_bound 0.3) took already about 3 days to render
on my old machine I am meanwhile a little bit bored in experimenting
with it. But for sure I will continue the work sooner or later (maybe
later, because I need a faster machine) so I have thought I do post this
little experiment to see if it somehow works for you.

And thanks a lot for trying it :)

-Ive

Post a reply to this message

> You can send it to me if you want - someone from the Megapov Team might
> find the time to work on it so it can be included in Megapov.

Great. I will try to at least do clearly indicate the parts I have changes and add
some comments. As soon this is done I will send it.
And I guess hdr output is not only useful for creating global illumination maps.
It should also work fine in combination with a program like your HCR-Edit
to have the possibility to apply gamma and exposure control after the render
process but with the full range of data.


> If continued trace does not work i assume you have not yet implemented the
> Read_Line() method.

Right. :)

-Ive

Post a reply to this message

3e63aaf2@news.povray.org...
> > Here is another one.  You know, your HDR comes through as being
> > really dark, much darker than the reference probes.  I had to turn the
> > ambience up in order to get just the illumination that you see here.
Any
> > idea why that might be?
> >
>
> Well, this was my first experiment with a more complicated HDR scene
> than just a box. And it seems that producing good HDR images with
> PoV requires some experience than I already have.

I've played with many HDR image now and it seems that there's no real
standardisation about their ambience value. Some are very light and some are
very dark, at the point that they must be corrected in HDRShop just to see
something else than pure white or pure black... Eventually, I always have to
tweak the ambient value to get nice results. I wouldn't worry to much about
this, as long as the image is usable (which is the case here).

G.


--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message