POV-Ray : Newsgroups : povray.documentation.inbuilt : looks_like suggestion : Re: looks_like suggestion Server Time
23 May 2024 18:11:51 EDT (-0400)
  Re: looks_like suggestion  
From: Mike Horvath
Date: 14 Sep 2016 21:32:58
Message: <57d9fa4a$1@news.povray.org>
On 9/12/2016 10:43 AM, Mike Horvath wrote:
> On 9/12/2016 7:56 AM, clipka wrote:
>> Am 12.09.2016 um 12:45 schrieb Jim Holsenback:
>>> On 9/11/2016 1:55 PM, clipka wrote:
>>>> Am 11.09.2016 um 16:23 schrieb Mike Horvath:
>>>>> Looks_like works fine with parallel lights as long as you position the
>>>>> light correctly to start off with instead of placing it at the origin
>>>>> and then translating it at some later time.
>>>>> Unless I made a mistake.
>>>> Judging from the code, I see no reason why that should make any
>>>> difference.
>>> Pulled directly from the previous version of this section:
>>> "For point lights and spotlights you could also place the light source
>>> and the looks_like object at the origin and translate them together
>>> later. But for parallel lights this will not work."
>>> Did I get it wrong? Just trying to honor what was previously mentioned!
>> I can't say for sure at the moment. All I can say is that I didn't find
>> anything in the code that would explain that wording, but I might have
>> overlooked something. I guess experiments may be needed.
>> Maybe the sentence was placed there not so much due to unexpected
>> effects on the looks_like object, but unexpected effects on the light
>> source itself. (For instance, whether or wheter not POV-Ray translates
>> the `point_at` location along with the parallel light source, either way
>> the effect may be surprising to some users.)
> I posted a test scene in p.t.s-f.
> Mike

Oops, I was wrong. But translation is *dangerous*. As per clipka's 
response to my post:

 > With a parallel light source, the direction of all light rays depends
 > only on the /relative/ location of two points: The light source's
 > nominal center and the point_at parameter. But transformations act on
 > both these points (I've just checked the code), so a translation
 > operation changes only the absolute location of the points, not their
 > relative location to one another. And thus the light direction
 > remains unimpressed.

I.e. translating a light source translates the location of the source of 
the light *and* the location the light is pointing at, which could 
produce unwanted effects. This should be mentioned in the article. Or at 
least state that you should explicitly configure point_at to keep this 
from happening.

You might want to update this page as well:


It recommends translating the light source. Quote:

 > It's a good idea to define both things, the light_source and the
 > looks_like object, at the origin, and then translate them to their
 > right place...


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.