Am 28.12.2013 18:43, schrieb MichaelJF:

>> In a sense yes - the crackle pattern has changed. The general properties
>> of the pattern haven't though, so the description in the docs is still
>> valid.
>
>> There are a few other random-determined patterns that have changed as
>> well in a similar manner. I can't tell off the top of my head though
>> which those are.
>
> I have no idea what he ment with "in a sense" really. The crackle pattern is
> based upon the calculation of Voronoi diagrams, a (very) well defined issue
> within computational geometry. May be the algorithm to calculate Voronoi
> diagrams was exchanged with a better one. But as Christoph continues "other
> random-determined patterns [...] have changed as well in a similar manner" (the
> random part with crackle is the positioning of the points in space to calculate
> the Voronoi diagram). That gave me the suspicion that not the pattern are
> changed but the noise generator again (last official change after 3.1). That
> would explain a lot of the issues you all report here. But it is only a guess
> finally.

I can confirm that it's not just a guess - it is indeed exactly what is 
happening.

(I must confess though that I'm not into details /why/ the random 
generator was changed. Supposedly it was necessary to make the whole 
crackle mechanism work with multithreading at reasonable speed.)

Post a reply to this message