POV-Ray : Newsgroups : povray.programming : render differences in github/head and 3.7 : Re: render differences in github/head and 3.7 Server Time
24 Apr 2024 20:28:49 EDT (-0400)
  Re: render differences in github/head and 3.7  
From: Dick Balaska
Date: 10 Dec 2015 01:04:26
Message: <MPG.30d2dfe8fa9e8fa698968a@news.povray.org>
In article <5668c3c7$1@news.povray.org>, ano### [at] anonymousorg says...
> 
> If you're worried about the quality of your build, you may want to
> compare it to the latest semi-official development build of the master
> branch, which can be found at https://github.com/c-lipka/povray/releases.
> 
Qu'est-ce que c'est "quality of the build"?

> Also, Linux builds frequently use different mathematical precision than
> Windows builds (and there may also be differences between the Windows
> plain 32 bit builds on one hand, and the 32 bit SSE2 and 64 bit builds
> on the other hand), which in some rare cases might produce noticeable
> differences in the output.

Ah poo.

That Apocolypse pigment was a tiny background object and I wasn't 
*really* concerned [1], but I just noticed one of my foreground objects
	texture {
		pigment {Jade}
		finish {
			ambient 0.3
			reflection 0.05
			diffuse 0.7
			roughness 0.3
		}
		scale 5*inches
	}
looks completely different.  It smells like the povdos version is 
ignoring the scale 5*inches.  I definitely have to RCA this.

linux http://www.buckosoft.com/~dick/t/tteo2021.png
vs.
povdos http://www.buckosoft.com/~dick/t/tteo2033.png

> 
> 
> That said, there is also a possibility that you've found a bug, and
> therefore I do indeed want to know more about this.

[1] you can't even really tell in the video. But I use an ancient ACDSee 
to flip lossless png files at 18 fps during development, and there it 
was noticable.


Post a reply to this message

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