![](/i/fill.gif) |
![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
From: Thorsten Froehlich
Subject: Re: JPG-output with povray for windows v3.5
Date: 19 Feb 2004 10:59:37
Message: <4034dd69@news.povray.org>
|
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
In article <4034c5ff@news.povray.org> , Warp <war### [at] tag povray org> wrote:
> Then every existing image format is lossy because POV-Ray calculates
> the color components as 32-bit floats. Converting them eg. to
> 8 bits-per-component or even 16 bits-per-component image (eg. PNG)
> is lossy.
I would not be surprised if floating-point colors were part of some TIFF
format specification ;-)
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trf de
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Darren New <dne### [at] san rr com> wrote:
> Technically true, but since your eyes have a very difficult (if not
> impossible) time distinguishing more than 8 bits of color, one could
> argue differently too. :-)
Actually for B/W images 8 bits can sometimes be too little, specially
when dealing with large areas of very slowly changing shades of gray.
--
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}// - Warp -
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
> > Technically true, but since your eyes have a very difficult (if not
> > impossible) time distinguishing more than 8 bits of color, one could
> > argue differently too. :-)
>
> Actually for B/W images 8 bits can sometimes be too little, specially
> when dealing with large areas of very slowly changing shades of gray.
To add to that: I'm on 32bit for my Desktop, because I'm seeing bands of
colors when using only 16bit with my raytraced images (which is hence the
reason why I output to 24bit BMP most of the time). Since Windows offers me
only 16 or 32, I'm on 32.
Regards,
Tim
--
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>
Email: tim.nikias (@) nolights.de
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
> To add to that: I'm on 32bit for my Desktop, because I'm seeing bands of
> colors when using only 16bit with my raytraced images (which is hence the
> reason why I output to 24bit BMP most of the time). Since Windows offers
me
> only 16 or 32, I'm on 32.
And eventhough you are using 32-bit mode, you see only 24-colors, unless you
have some specialiced video card. I also agree that 8-bits/channel is way
too little with color gradations. And for editing purposes 16-bit/channel is
sometimes mandatory. Sadly there are not many (consumer grade) video cards
showing 48-bit colors.
Severi
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
On Wed, 18 Feb 2004 21:33:09 -0800, Darren New <dne### [at] san rr com>
wrote:
>Christopher James Huff wrote:
>
>> it isn't lossy, it's just really limited in the number
>> of colors it can represent.
>
>True, it depends on how many colors you're trying to represent. Given
>this is a ray-tracing newsgroup and GIF only supports 256 colors, I
>think it's safe to say it's lossy. ;-)
It's close to black & white, isn't it?
--
Andreas
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
In article <4035211f$1@news.povray.org>,
"Tim Nikias v2.0" <tim.nikias (@) nolights.de> wrote:
> To add to that: I'm on 32bit for my Desktop, because I'm seeing bands of
> colors when using only 16bit with my raytraced images (which is hence the
> reason why I output to 24bit BMP most of the time). Since Windows offers me
> only 16 or 32, I'm on 32.
Why BMP? You should use PNG...aside from the compression, it stores the
gamma value in the file, and lets you save images in 48 bit...16 bits
per channel, or 65536 levels per channel instead of 256.
--
Christopher James Huff <cja### [at] earthlink net>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tag povray org>
http://tag.povray.org/
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
In article <40352395$1@news.povray.org>,
"Severi Salminen" <sev### [at] NOT_THISsiba fi> wrote:
> And eventhough you are using 32-bit mode, you see only 24-colors, unless you
> have some specialiced video card.
"32 bit color" is just 24 bit color with an extra byte every 3 bytes.
This keeps memory nicely aligned, but contributes nothing to the color
precision. It's just a memory/speed tradeoff, though the fourth value is
often used for things like alpha.
> I also agree that 8-bits/channel is way
> too little with color gradations. And for editing purposes 16-bit/channel is
> sometimes mandatory. Sadly there are not many (consumer grade) video cards
> showing 48-bit colors.
You don't necessarily need to view it to get a benefit from it.
--
Christopher James Huff <cja### [at] earthlink net>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tag povray org>
http://tag.povray.org/
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
In article <4034dd69@news.povray.org>,
"Thorsten Froehlich" <tho### [at] trf de> wrote:
> I would not be surprised if floating-point colors were part of some TIFF
> format specification ;-)
There is...96 bits per pixel. Not often used for this reason...a
1600x1200 image would take 21MB. There are other formats that use about
half as much. They do so with less precision, but 32 bits per component
is overkill for most things.
--
Christopher James Huff <cja### [at] earthlink net>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tag povray org>
http://tag.povray.org/
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
> Why BMP? You should use PNG...aside from the compression, it stores the
> gamma value in the file, and lets you save images in 48 bit...16 bits
> per channel, or 65536 levels per channel instead of 256.
Well, for one, I can use BMP directly as wallpaper on my Windows OS. I don't
need more than the 24bit it comes with, that's another. And, most of the
time, the image gets converted to JPG anyway, either for posting on these
very newsgroups that I've got open and running almost every hour my PC is
running, or for my website. Only seldomly do I need other formats, like for
a book publishing. And even then, 24bit was sufficient so far.
So, though your points are valid and true, I'm too lazy. :-)
But I've never really understood that gamma issue with PNGs... How exactly
does it work, and how is it supposed to work? I've heard that it stores the
gamma value it has been made with, and a viewing app would adjust the gamma
to compensate for different screens, but does it really work?
--
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>
Email: tim.nikias (@) nolights.de
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Thorsten Froehlich wrote:
>
>> Then every existing image format is lossy because POV-Ray calculates
>>the color components as 32-bit floats. Converting them eg. to
>>8 bits-per-component or even 16 bits-per-component image (eg. PNG)
>>is lossy.
>
> I would not be surprised if floating-point colors were part of some TIFF
> format specification ;-)
>
Or maybe FITS standard for astronomical pictures ?
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |