per-pixel random problems?

[Cyrille Berger]

On Thursday 11 December 2008, Matthew Woehlke wrote:
> Krita is  supposed to be released in 2.0? If it can't even open a file it
> saved  without data loss, um...
guess what ? that's one of the reason it's not released yet...

> 64-bit. Hmm... not sure why that would make a difference. I don't know
> the distribution properties of erand48, otherwise I wonder if that might
> work better (and be faster, as well).
>
> (Obviously we would want to copy an /implementation/ of erand48, since
> it is a Linux-only function. However, it is a pure-integer function of
> the form '((A * seed) + B) mod C'.)

it's not erand... We need a non recursive function to be able to compute it's 
value at every pixel independentely of other values. Being integer based 
(unlike the current one) would be a huge bonus, since I discovered since that 
time that floating point units in CPU are buggy and give very different 
results from one arch to an other...

> The png is substantial (500+kb), so I dumped it at
> http://img124.imageshack.us/img124/9986/noisetesthq8.png. (Um. And
> ImageShack no longer works without scripting or View Source; guess I
> won't be using them any more.)
Here is what I get: http://cyrille.diwi.org/tmp/noise.png

-- 
Cyrille Berger