On 05/03/2018 16:50, Stephen wrote:
> 
> Oo! shiny. I like.
> 
> You are on a roll, Norbert. :-)

It looks good as a negative, too (also).

-- 

Regards
     Stephen

Post a reply to this message

"Norbert Kern" <nor### [at] t-onlinede> wrote:
> Stephen <mca### [at] aolcom> wrote:
>
> > Has anyone tried to produce a solarisation effect?
>
> Hi,
> with a simple approach I got this image - here is the code
>
Brilliant! I actually didn't think it was possible.

Post a reply to this message

> On 4-3-2018 18:12, Norbert Kern wrote:
>> #declare p_o_x = function {p_original (x,y,z).x}
>> #declare p_o_y = function {p_original (x,y,z).y}
>> #declare p_o_z = function {p_original (x,y,z).z}
>>
> 
> I had thought to use:
> 
> #declare p_o_R = function {p_original (x,y,z).red}
> #declare p_o_G = function {p_original (x,y,z).green}
> #declare p_o_B = function {p_original (x,y,z).blue}
> 
> But that should yield the same thing I guess. I was going to code that...
> 

Yes, both forms will yield the same end result, but, the second form is 
recommended whenever you are dealing with colours.

Alain

Post a reply to this message

Thomas de Groot <tho### [at] degrootorg> wrote:

> >
> > I shall make a small test scene. A way for me to remember stuff.
> >
>
> Like this. Left, image gamma is 1.0; right, image gamma is 2.2.
>

Nice! And you picked a good pattern to show it off.

The real *fun* begins when you take a height_field image (made elsewhere) and
*combine* it with a procedural pigment or pattern-- both of which can be
gamma-bent. It's a bit more complicated to code, but I was working up an example
anyway, and will post in a new thread. (BTW, some other interesting effects can
be gotten by making the gamma *less than* 1.0.)

Post a reply to this message

"Kenneth" <kdw### [at] gmailcom> wrote:
> (There's also a HF difference
> between images SAVED at different gamma values, for the same reason-- 1.0 being
> a linear slope, 2.2 being more of a curve.)

Uh, I'm not so sure about that now. I made a test scene a couple of weeks ago
that showed this. (Or so I thought.) Now, a newer test scene shows no
difference-- producing the same 'linear' height_field slopes no matter what
gamma the images are saved at (with either JPEG *or* PNG image files.) But then
again, I'm using my older wonky Photoshop 5 to make and save the original
images, which can't be completely trusted as far as gamma goes.  Hmm, a mystery.

Post a reply to this message

On 5-3-2018 14:46, clipka wrote:
> Am 05.03.2018 um 14:27 schrieb Bald Eagle:
>> This is a great tool - one that is obvious in retrospect, but I had never
>> thought of.
>>
>> It's great that it's possible to save an image with a certain gamma and then use
>> it as a height field, but the obvious question that affects the real utility of
>> this is:
>>
>> Is it possible to READ the existing, non-gamma modified image in using a certain
>> gamma value for use in a height field?
>>
>> So, for example, use TdG's gamma 1.0 image to generate his gamma 2.2 height
>> field?
> 
> Well, yes, absolutely: In the height field, just specify "gamma FLOAT"
> after the image file name.
> 
> I think you need to specify "gamma 1/2.2" for the intended effect.
> 
> There may be slight differences due to different rounding, but other
> than that it should work just fine.
> 

And that is exactly what I did above; I see that I generated some 
confusion with BaldEagle :-) The - single - original height_field image 
has an unknown (to me) gamma. It is in POV-Ray that I declare the gamma 
value to use, like Clipka said [code snippet]:

   pigment {
     image_map {
       tga "Apophysis_06_hfisland.tga" gamma 2.2 //or 1.1
       map_type 0
       interpolate 2
     }
   }

-- 
Thomas

Post a reply to this message

On 5-3-2018 21:51, Kenneth wrote:
> Thomas de Groot <tho### [at] degrootorg> wrote:
> 
>>>
>>> I shall make a small test scene. A way for me to remember stuff.
>>>
>>
>> Like this. Left, image gamma is 1.0; right, image gamma is 2.2.
>>
> 
> Nice! And you picked a good pattern to show it off.
> 
> The real *fun* begins when you take a height_field image (made elsewhere) and
> *combine* it with a procedural pigment or pattern-- both of which can be
> gamma-bent. It's a bit more complicated to code, but I was working up an example
> anyway, and will post in a new thread. (BTW, some other interesting effects can
> be gotten by making the gamma *less than* 1.0.)
> 

Been there done that ;-)

Even better if you use the image and the pattern within functions and 
combine those.

-- 
Thomas

Post a reply to this message

On 5-3-2018 21:36, Kenneth wrote:
> "Norbert Kern" <nor### [at] t-onlinede> wrote:
>> Stephen <mca### [at] aolcom> wrote:
>>
>>> Has anyone tried to produce a solarisation effect?
>>
>> Hi,
>> with a simple approach I got this image - here is the code
>>
> Brilliant! I actually didn't think it was possible.
> 

*everything* is possible in POV-Ray! ;-)

-- 
Thomas

Post a reply to this message

Am 06.03.2018 um 00:02 schrieb Kenneth:
> "Kenneth" <kdw### [at] gmailcom> wrote:
>> (There's also a HF difference
>> between images SAVED at different gamma values, for the same reason-- 1.0 being
>> a linear slope, 2.2 being more of a curve.)
> 
> Uh, I'm not so sure about that now. I made a test scene a couple of weeks ago
> that showed this. (Or so I thought.) Now, a newer test scene shows no
> difference-- producing the same 'linear' height_field slopes no matter what
> gamma the images are saved at (with either JPEG *or* PNG image files.) But then
> again, I'm using my older wonky Photoshop 5 to make and save the original
> images, which can't be completely trusted as far as gamma goes.  Hmm, a mystery.

By default, when using images in a height_field, POV-Ray presumes that
the image data is to be taken linearly no matter what gamma the file
claims to match, e.g. in an 8-bit image a data value of 240 will always
be interpreted as a height of 240/255, even if the file claims to be
saved for a display gamma of 2.2.

So if you just create an image in Photoshop and /assign/ a particular
gamma to it later, there will be no difference as the image data will
not change. You'd need to /convert/ the image from one gamma to another
to see an effect.

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Am 06.03.2018 um 00:02 schrieb Kenneth:
> > Now, a newer test scene shows no
> > difference-- producing the same 'linear' height_field slopes no matter what
> > gamma the images are saved at...
>
> By default, when using images in a height_field, POV-Ray presumes that
> the image data is to be taken linearly no matter what gamma the file
> claims to match, e.g. in an 8-bit image a data value of 240 will always
> be interpreted as a height of 240/255, even if the file claims to be
> saved for a display gamma of 2.2.
>
> So if you just create an image in Photoshop and /assign/ a particular
> gamma to it later, there will be no difference as the image data will
> not change. You'd need to /convert/ the image from one gamma to another
> to see an effect.

Thanks. My older test scene was also more complicated-- using a HF-image gamma
*and* a pigment_map with a blend mode/blend gamma. That combination probably
confused the result for me.

Post a reply to this message