Qbasicnews.com

is on-line. It provides the source codes for the Perlin Noise, both the original Ken Perlin approach and a fast, QB oriented, new algorithm.



Those who may be discouraged by the theory in Part 1 should feel fine with this Part 2, much more practical.





This is the last tutorial of a series of 7:

http://mandelbrot.dazibao.free.fr/Tutos/Tutos.htm

Lesson 1 - True Colour bitmaps: How to use the bitmaps as unlimited virtual screens
Lesson 2 - Vesa/SVGA True Colour modes: No limit in terms of screen size and resolution
Lesson 3 - Mouse programming in SVGA modes: Drive your programs with the polling mouse technique
Lesson 4 - Bresenham's algorithms: Fast lines and circles programming - A reference in terms of optimisation
Lesson 5 - The HSV colorspace: The key to professional true-colour graphics
Lesson 6 - Perlin Noise Part 1: Get into the details of Ken Perlin's magic formulas
Lesson 7 - Perlin Noise Part 2: Programs, applications and references

I hope these tutos will help the QB community, and most of all, prove once and for all that Quick-Basic is alive :lol:

it's just amazing good work

You never cease to amaze me jark =P.

Yea, your work is amazing. I'll have to take a look at the tutorials

Wow, you can do that in QB?!?!?

@Hellbound: Of course you can. There's not much Qbasic cant do (but what it cant do is the stuff you want it to do :-?)

Quote:Wow, you can do that in QB?!?!?

Heh heh.... i had a hard time believing it myself when I first saw it... =P

This and much more is possible in QB =P. I think jark thought of porting it to C but then it would require a lot of code changes and the speed gains wouldnt be significant.

I cannot benchmark QB versus C myself, simply because I don't "speak" C.

In the last tutorial I have published, I managed to adapt the original Ken Perlin's algorithm for QB. The C code by Ken Perlin can be downloaded from a couple of webpages (I have quoted them at the bottom of the tuto).

I would find it interesting if someone could run the two codes on the same machine and with the same application (a cloud cover for example). None of these two codes is really optimised, and it's a mix of integer and floating points variable in both cases.

I've always been said the speed gain id between 3 and 6 times, depending on the study case... As TBBQ said, it's not worth reconding everything: when you can wait 2 minutes for a picture, then you can wait 10 minutes. But when you read Ken Perlin Code, then you understand why Quick-Basic is so programmer-friendly :lol:

Jark and i remember that you optimised ur qb prog so nicely that the time to render the pic came down from few days to mere hours ( or minutes).