Thorsten stated that the blob bugs are amongst those that will not be fixed 
in 3.5, because either 
-they would need fundamental changes in the code, or are inherited from 
very old versions
-they can not be fixed with current PC architectures
-are not very significant, since they had existed for quite some time, and 
have only recently been reported

(Excuse me if I don't remember the reasons exactly.)

But this post (Blob : Toward a solution
http://news.povray.org/3B991C90.D715B7A3[at]free.fr) suggests that at least 
some of the blob bugs did not exist in PoV-Ray 2.0, and were "introduced" 
somewhere in between 2.0 and 3.1. (Is it a naive assumption that this means 
that it should be relatively easy to fix them?)

Could someone please clarify this? Has anyone done extensive comparison of 
different versions concerning the blob bugs?

---
Gergely

Post a reply to this message

Gergely Vandor wrote:
[SKIP]
> But this post (Blob : Toward a solution
> http://news.povray.org/3B991C90.D715B7A3[at]free.fr) suggests that at least
> some of the blob bugs did not exist in PoV-Ray 2.0, and were "introduced"
> somewhere in between 2.0 and 3.1. (Is it a naive assumption that this means
> that it should be relatively easy to fix them?)

According, to my experiments, it's easy to fix, but there is a price:
 + sturm become highly recommended (I would say mandatory).
 + And of course, there is also a price in computation time
 (due to sturm but also to the removal of the optimisation).

If someone could come with a justification of the formulaes used for
 the hull optimisation, it might be better to debug the optimisation rather
 than deleting it.

And it does not explain the media/blob strangeness...

-- 
Non Sine Numine
http://grimbert.cjb.net/
Puis, s'il advient d'un peu triompher, par hasard,

Post a reply to this message

news:3CCD182B.F823895C@atosorigin.com: 

> Gergely Vandor wrote:
> [SKIP]
>> But this post (Blob : Toward a solution
>> http://news.povray.org/3B991C90.D715B7A3[at]free.fr) suggests that at
>> least some of the blob bugs did not exist in PoV-Ray 2.0, and were
>> "introduced" somewhere in between 2.0 and 3.1. (Is it a naive
>> assumption that this means that it should be relatively easy to fix
>> them?) 
> 
> According, to my experiments, it's easy to fix, but there is a price:
>  + sturm become highly recommended (I would say mandatory).
>  + And of course, there is also a price in computation time
>  (due to sturm but also to the removal of the optimisation).
> 
> If someone could come with a justification of the formulaes used for
>  the hull optimisation, it might be better to debug the optimisation
>  rather than deleting it.
> 
> And it does not explain the media/blob strangeness...
> 

At least and at last I understand. :) Thank you.

Post a reply to this message