Segment Fault!

Was continuing on my terrain generation today and it kept crashing a lot more often...

...i took that opportunity to hunt it down with gdb but weren't able to locate the error inside my code. But at least it hinted me a method were the error occured.

Normally i was writing some color values in the sf::Image i had there so i commented it out and discovered some weird blue and green pixels that shouldn't be there as the image was initialised fully transparent.

I changed the initial color to solid blue and watched my image.... it turned out that only 1/2 of the image were actually blue! Now i init the image with double the size and haven't run into another segment fault so far. I hope i get one of the developers of SFML on there IRC. It looks like a serious issue! If already an image of a size 30x30 has such issues i don't want to know what my 1024x1024 terrain image is currently doing!

Current status of the terrain generation.

Back to terrain i got with erosion something that almost looks like it could become terrain. Sadly the scale is a kinda to tiny for a character to wander through. Maybe i'm going to scale and blurr it up or i will need to improve the algorithm a lot more to handle terrain of bigger size!

I guess thats it already... i was going to tell you a lot about biologic neural networks compared to simulated ones... but it seems kinda off topic and now i already got enough to show you. So it will have to wait until i run out of stuff to tell or you asking me directly.

Cheers! - Arc.

P.S: Okay! I'm sorry guys! Indeed i was being pretty dumb and childish there! SFML is working all right! It was me making the mistake there. I just forgot to reset the j counter for every i.... So yes SFML is still save to work with!

Hope i wasn't too rude.