Zevensoft

18th September 2003 13:14 UTC

Quick Contest
Zevensoft Media Network in conjuction Zevensoft Interactive present the:
Sonic 2 Contest

What you have to do, is design a "level" for sonic to run through. How you do this, is by making a 'Vector Equation'. Example:

R(t) = tx + sin(t*pi*10)y
Will make a sine wave level for sonic to run/jump/roll through.

The basic way it works, is it specifies where the ground is at time 't', which means you can make it loop if you want, however you'll have to provide a derivative for rotating sonic in that case.

The best 3 will be included into the Sonic 2 preset, which will be released 2 weeks from now. Contest closes in 1 week.

Tuggummi

18th September 2003 13:30 UTC

Sorry, would love to, if i would understand any of that.

Zevensoft

18th September 2003 13:36 UTC

Attached is an example.

Tuggummi

18th September 2003 13:44 UTC

Which is apprently done with avs2.7pre-release :igor:
Well i doubt i could do much out of it anyway. Sorry to waste your time.

anubis2003

18th September 2003 14:22 UTC

And it has to have loops in it. :p

Deamon

18th September 2003 14:28 UTC

I do not have the coding skills to do that for you. Sorry mate :(

[Ishan]

18th September 2003 17:09 UTC

me neither.:| altho i understand the equation but i'm too lazy and on top of that a n00b.:p

dirkdeftly

19th September 2003 00:21 UTC

i'd do it if I had the time :/
when I end up with some I'll try it out - then can I use it for the (never ending quest of) tff4? :D

jheriko

24th September 2003 16:03 UTC

you mean parametric? thats what that type of equation is called... and you can do beziers with it, so if you want a really cool static level you could make a huge 5000 control point bezier or something.

EDIT: I'll cook something up anyway. But I might not get round to it.

EDIT2: Why do you need a derivative to rotate sonic? why not calculate it roughly from the previous position and the current position. There are better ways to approximate implict derivatives but this is easy and works really well.

Zevensoft

25th September 2003 05:43 UTC

I suppose, I could just use atan2 to calculate the angle, but the main purpose of the dy/dx was for velocity reasons.

sidd

25th September 2003 11:04 UTC

by 'velocity reasons' i assume you mean that sonic will slow down (or actually remain the same speed) when going up and down hills...er..... i mean he wont just go at a static speed along the x axis.. er... his speed will be affected by hills :S bleh.. i cant describe... i wish i had the math-vocab of a true avser. :(


anyways.. why do i get the feeling that this 'contest' is just a lame excuse for you not to make your own terrain :D

.. hmm ive messed around abit.. i cant find a good way of creating true flats and slants, only curvs.... any suggests anyone?

[edit]
btw, no need for atan2-age, looksee unicycle thing: angle=(pos-oldpos)*pi or somthing ther'bouts

btw2, you guys seen the new edit post scheme?.. you can now delete post and even threads!!! w00t!
[/edit]

jheriko

26th September 2003 09:22 UTC

sidd: R(t)= atx + bty should be a straight line for different values of a and b.

zeven: can't you use (dy/dt)÷(dx/dt) = dy/dx and calculate dy/dt and dx/dt by simply dividing the difference in x or y per frame by whatever you increment t by every frame? Just an idea, I'm at work so I can't test it out. Though i think its not good enough for what you want to do with it.