In article <3A9220F5.F5E3D687@kivisalo.net>, Kari Kivisalo 
<kar### [at] kivisalonet> wrote:

> > Can you give any clues on what differences we might see
> > if POV-Ray becomes able to output 48 bit images ?
> 
> I plan to experiment with different post processing algorithms,
> dynamic range compression etc., which take images with high dynamic
> range and optimise them for monitor viewing. This requires the 48 bits.
> The images should look more like photographs than CG. That's the short
> version.

I don't think 48 bit output will do what you want...it doesn't affect 
the dynamic range of the image, just the number of levels in each color 
channel. You will probably need to use the post_process patch, which has 
access to the unclipped, full precision values.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

Chris Huff wrote:
> 
> In article <3A9220F5.F5E3D687@kivisalo.net>, Kari Kivisalo
> <kar### [at] kivisalonet> wrote:
>
> > I plan to experiment with different post processing algorithms,
> > dynamic range compression etc., which take images with high dynamic
> > range and optimise them for monitor viewing. This requires the 48 bits.
> > The images should look more like photographs than CG. That's the short
> > version.
> 
> I don't think 48 bit output will do what you want...it doesn't affect
> the dynamic range of the image, just the number of levels in each color
> channel.

Dynamic range/contrast ratio is directly proportional to number of levels,
when talking about linear samples.

Think about audio analogy, 8bit samples have a dynamic range of 48 dB
and 16 bit samples have 96 dB. ( dB=20*log(max_sample/min_sample) )

I know what I'm doing. Trust me :) I also know what you mean but for
hands on quick testing I have to get the images to Matlab and Photoshop.


______________________________________________________________________
Kari Kivisalo                                  http://www.kivisalo.net

Post a reply to this message

Kari Kivisalo wrote:
> 
> Tor Olav Kristensen wrote:
> >
> > Can you give any clues on what differences we might see
> > if POV-Ray becomes able to output 48 bit images ?
> 
> I plan to experiment with different post processing algorithms,
> dynamic range compression etc., which take images with high dynamic
> range and optimise them for monitor viewing. This requires the 48 bits.
> The images should look more like photographs than CG. That's the short
> version.

I see.

Thank you for explaining.

I also wonder if there are any non-linear properties 
related to the monitor's "presentations" of the binary 
coded light intensities (in images) that needs to be 
corrected.

Do you have any thoughts regarding this ?


-- 
Best regards,

Tor Olav

mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok

Post a reply to this message

I like the over-exposed bits - this makes it seem more real to me (in a
similar way to my comments on lens flare elsewhere). I know that this isn't
a real object (although it would be an interesting challenge for a sculptor)
but I enjoy seeing bits of the real world creeping in.

"Christoph Hormann" <chr### [at] gmxde> very nearly wrote in message
news:3A8F0261.3859B799@gmx.de...
>
> It looks good, the bright parts are a bit overexposed.  The radiosity
> settings seem quite good, I suppose you used the same values Gilles used
> in his picture.
>

Post a reply to this message

Tor Olav Kristensen wrote:
>
> I also wonder if there are any non-linear properties
> related to the monitor's "presentations" of the binary
> coded light intensities (in images) that needs to be
> corrected.

I'm just trying to find a way to present images on a limited contrast
ratio medium. Display device properties and viewing conditions are
a separate problem. 

______________________________________________________________________
Kari Kivisalo                                  http://www.kivisalo.net

Post a reply to this message