Re: [glob2-devel] main page announcement of bounty system

[Leo Wandersleb]

Kai Antweiler wrote:
> > First of all: I strongly believe oss will be the future in many if not all 
> > areas of software development.
> > Having said that I insist in comparing glob2 with commercial products.
>
> I don't, so I skip most parts of the mail.

My examples were all oss.

> We were lucky that Bradley wrote so much. By the way: he barely
> finished his last rewrite before he left.

Left? I didn't know he left us.

> If the timing had been a bit worse,
> another net-rewrite would have been left useless.

True. But maybe a bit more testing would have motivated Bradley to not leave the 
net code in this unfinished state.

> > I will look into it later.
>
> OK.

As far as i can see Perlin noise has not been replaced by simplex noise. Eric 
did only replace the function call that is called the most. The calls for cloud 
movement are still perlin-based.
Also the simplex noise is not initialized using random numbers. You can see that 
by the fact that the clouds shape is initialized the same way at every start of 
glob2 while the movement is different with every start. Choosing to replace all 
by simplex would allow to let the cloud movement (wind) affect particle 
movement. But as the cloud layer is rather infinite than torrodial the actual 
cloud cover depends on how many round trips you scrolled and would thus not be 
usable to for example determin wheat growth.

Are there any serious requests to kick out perlin, make the clouds torrodial or 
make cloud movement/cover available in game.cpp?

Please mention it at COfundOS.org ;)

Greetings,

Leo Wandersleb