megapw

29th December 2003 21:40 UTC

in need of help
Im new to making AVS (playing around since yesterday)and this is my first real attempt and ofcourse it doesnt get quite right...
If you look at my preset youll probably quite easy se what i was aiming for, some things got wrong tho...

I cant get the spectrum analyzer to get to the bottom, it starts a bit up (gets a jagged bottom) and it seems like 20% of the right side never gets used.
And then its the front of the whole landscape wich i cant figure out a way to hide. :(

ofcourse, all comments is welcome
[edit]
Doh seems like the wrong place to post, sorry.
i cant delete the post tho (it says i dont have permission even tho im logged in), could anyone move this to the right place?

hungryskull

30th December 2003 02:13 UTC

Since I am not completely sure what you want I can't help you much. But I do have a couple things to say about this preset.

The first is: you don't need to use a DM in this preset. Use a static movement, it's faster.

The second is: Thanks for using SSC's and a movement instead of Render/simple's and default movements.

Very n00bish preset.

sidd

30th December 2003 04:58 UTC

not that bad really. Alot better than most "first attempts". At least you are starting to get the hang of basic code.

The best way around avs diplaying those ultra-rarely-used frequencies (the flat bit at the end of your spectrum) is to use the getspec(a,b,c) command instead of the 'v' variable.

Using getspec(i,0,0) is pretty much the same as using v. So, try putting v=getspec(i*.6,0,0); at the start of your per point secton in the superscopes.

The only difference, is that getspec() returns values between 0 and 1, wheras v returns values between -1 and 1. So youll have to play around with it a bit so that it matches your other code.

Secondly, same as hungry skull said. The dynamic movement is both slower and less precise than the normal trans/movement tool. This is because it is dynamic; you can make it change every frame, so the code has to be recalculated every frame. Not only this, but it can only afford to calculate a certain numbre of points (the grid size), and it has to estimate the rest (bilinear filtering). The normal movement, however, only calculates your code once, and can afford to do it for every single pixel on the screen.

In other words, unless you put something usefull in the 'frame' or 'beat' section of the dynamic-movement, you might as well be useing the normal one.

im not sure what you meant by this:

I cant get the spectrum analyzer to get to the bottom, it starts a bit up (gets a jagged bottom)

Jaheckelsafar

30th December 2003 08:43 UTC

You could probably cover the jagged bottom somehow, and that might be your best option. If you make it start off the bottom of the screen, the movement won't pull it on, it only works with what's on the screen. If you want it to come up from the bottom of the window you'd have to stick all that you have in an effect list and stick a zooming movement after the effect list. Something simple like x=x*.8; y=y*.8;

The reason it's "starting up" is because that's where you tell it to start. Adding .19 to the y value in each scope will put it just above the bottom of the windows.

Remeber to put the movement before the scopes or it will be harder to place them precicely.

megapw

31st December 2003 17:26 UTC

Altho i dont have time for much testing things right now, newyears eve and all but, thanx for helping me out.
(i did try to ....stick all that you have in an effect list and stick a zooming movement after the effect list. Something simple like x=x*.8; y=y*.8;.... but it did nothing, must have done some silly mistake)

Happy New Year all!

fragmer

31st December 2003 20:47 UTC

to modify "x" and "y" you need to enable "rectangular coords" in movement/DM. You can also convert polar coordinates ("r" and "d") to rectangular ("x" and "y") by using these formulas:

 x=sin(r)*d;
 y=cos(r)*d;
 d=sqrt(x*x+y*y);
 r=atan2(x,y);

r=atan2(x,y)+.1;   // find rotation
d=sqrt(x*x+y*y);   // find distance
x=sin(r)*d-sin(y*2)*.1;   // create a swirl!
y=cos(r)*d;   // calculate x

Jaheckelsafar

1st January 2004 19:36 UTC

mess around. I nothing woeks, post the messed with version we might be able to help.