How do they look - AVS Forums

Limpet

18th January 2003 01:01 UTC

How do they look
I am new to the avs presets and i came up with a zip to show off some stuff i made.

anubis2003

18th January 2003 01:35 UTC

No offense, but these really are bad. One of them only has fireworks and water. Some of them are not just pre-defined, but I think most of the code is somewhat random, which can generate some good effects, but is much harder. The best thing I can tell you to do is to figure out some more math and when you go to make something in AVS have an idea for what you want as an outcome (although this can, and most of time will, change before you are done). To figure out the math needed you can search the web, avs forums, etc. Once you do this try going back and remixing these presets using more sophisticated approaches, i.e. SSC's, DM's, APE's(Convolution Filters, Texers, etc.), buffer blending, etc. After you do this you can create presets that will look much better and are more beat reactive than these ones.

The preset that you have that is all black is black because the red, green, and blue variables are most likely 0. Try using a graphing calculator or computer program(Maple, Mathematica) to graph the functions first and make sure they fit the parameters (i.e. colors have to be in between 0 and 1, x and y have to be in between -1 and 1. Good luck to you and I hope this doesn't dishearten you.

For some help you can check out Atero's AVS primer (it's in his signature) which he will be updating soon.

Jaheckelsafar

18th January 2003 06:55 UTC

Everybody started as a noob. Don't take bad reviews to heart. They are meant to be constuctive criticism.

That said, if you're planning on putting these up anywhere, don't. Never release the first prestets you make, even if they're wicked awsome amazing. Why? 'Cause you're still learning. You think they're great because they're yours when infact, they're no great shakes.

I likeed flower, but most of the others were either too static or too chaotic for my taste.

Raz

18th January 2003 12:16 UTC

Yeah they arent as bad as annubis made out they look better than my first attempts. There are some half decent scopes in there. zenx - rush is an easy preset to remix isnt it :p. And that scope is probably broken because like annubis said.

Pi in avs can be calculated by "acos(-1)" in avs btw, according to atero anyway. I think you used fireworks just a little too often it really isnt that good an effect. Dont get disheartened and keep trying, learn the math and look at the avs primer and youl get better.

I might as well attatch my take on a kalleidoscope while im here :) (its just a slight mod of one of my presets i already posted here btw but it looks like a kalleidoscope)

Jaak

18th January 2003 12:22 UTC

Yea, Anubis right U should read Atero's AVS Primer. This is really good stuff, thing will be much clearer after reading/watshing(I know, spelling!) it. Keep up AVS-ing.

:winamp: AVS is LOVE :winamp:

anubis2003

18th January 2003 17:45 UTC

Limpet, sorry if I was too rough on your presets. I think I was wrong too when I told you that the one SSC's variables were set to 0. I think they might actually cause the x and y variables to be out of the range. As another pointer, don't use x and y variables anywhere but in the pixel editbox. Anywhere else they don't make sense. Try smoothing out your presets some, trade out your fireworks for some superscopes which can be much more dynamic. Most importantly search the forums for more avs coding knowledge. Once you gain more skills remix these and make them look better. Once again, sorry about yesterday, I had spent all day applying for scholarships for college. And to prove that we all started out as newbies I am going to attach my first preset.

dirkdeftly

18th January 2003 19:43 UTC

the cosine of pi is -1. the inverse cosine function (arccosine) of -1 will obviously give us pi.
in the same way, you can calculate pi as: pow(acos(-1)*asin(-1)*atan(-1)*8,1/3);

or, you can just use asin(1)*2 or atan(1)*4.

but the fastest is acos(-1) ;)

UnConeD

19th January 2003 05:14 UTC

the cosine of pi is -1. the inverse cosine function (arccosine) of -1 will obviously give us pi.

The cosine of (3*pi) is -1 as well ;).

Explanation:
A relationship is something that maps a set of values (a,b,c,d) onto their images (R(a),R(b),R(c),R(d)). The inverse relationship is the one that maps R(x) to x for all image-points.

If every value in the domain (the input, this case a,b,c,d) has only one image, then we call the relationship a function.

However, the inverse relationship of a function is not always (in fact most of the time) a function as well. For example the relationship f(x)=x^2 for all real points is a function. The inverse relationship (y where y^2=value) has to map for example '4' to '2' and '-2', so it is not a function.
On the other hand, you can split the inverse relationship up into two inverse functions sqrt(x) and -sqrt(x) and pick one of them to use (we normally use the positive one).

Similarly, look at the cosine function. For every image-value between -1 and 1, there are an infinite number of points in the domain that map to them, each offset by 2*pi. For example:

...
cos(-2*pi+pi) = -1
cos(pi) = -1
cos(2*pi+pi) = -1
cos(4*pi+pi) = -1
...

So the inverse relationship is definitely not a function, and you can't split it up into functions, because there are an infinite amount of them.

So what they do is take one of those functions (for acos it is the interval [0,pi]) and use that to define the inverse function ;)

dirkdeftly

20th January 2003 08:45 UTC

The cosine of (3*pi) is -1 as well

The cosine of (5*pi) is -1 as well :rolleyes:
i think that deserves a "well, duh" ...anyone that could understand my post in the first place is smart enough to figure that out (and probably stoned). i'm just too damn lazy to explain all the triggernomitry behind it....

anubis2003

20th January 2003 16:28 UTC

Alright, we've established that the arccos of -1 is pi in 3 posts. Time to move on.