Superscope...any help out there?

I'm rather new to AVS and I've been wondering, is there any sort of simple explanation posted somewhere about Superscope? I've seen it used to great effect, but I have no idea how to get started on making my own. Suggestions?

Hmm...
You know the syntax don't you?
= to assign a var to either
x, y, i, v, ect...
; ends a statement

x is the line on X, try writing " x=i-0.5; " in a superscope.
y is the vertical line, try changing " x=i-0.5; " to " y=i-0.5: ".
v is used to read the sound, try writing "x=i*2-1; y=v;" to get an example of this.
I is what creates the line, without it you'll just get a dot. try " x=0; " for an example.
to make a moving line use "x=sin(t)-i", in "frame" write "t=t+0.02", this will add 0.02 to "T" on each frame. sin/cos/others closes it into the box so it won't go out of it. I assume you know math here. 😉
x and y both go from -1 to 1, where -1 on both is the top left corner and one is bottom right.
That's the basics for ya, try continuing on this yourself. 😉


Linus

Different explanation

I found Linus' explanation to be a bit unclear so here's my way of explaining the superscope:

A superscope is basically a renderer that draws either dots, or the lines between these dots. You have to write code that, at the end, outputs a set of x,y coordinates for each point to be drawn. X ranges from -1 to 1 and is the horizontal coordinate, Y ranges from -1 to 1 and is the vertical coordinate. Some examples:

(0,0) - the center of the AVS screen
(-1,-1) - the top-left of the AVS screen
(0,1) - the center of the bottom border of the AVS screen
(0.8,0.9) - a point near the bottom right

The variable n is used to set the number of points to calculate each frame. Higher n means slower presets, so don't use huge numbers if it's not needed.

The variable i is different for each point and is a percentage that tells you which point you're drawing. For example:
- first point: i=0 (0%)
- last point: i=1 (100%)
- the 50th point of a 150 point superscope: i=0.33333... (33%)

The variable v is also different for each point and contains the current sound value for this point (either oscilloscope- or spectrumdata, depending on your choice).

If you click the 'help' button, you'll see all the functions you can use. Experiment with them if you don't know what they do exactly.

So now we'll make a basic superscope:
Init: n=300;t=0;tpi=acos(-1)*2;
Beat: t=t+rand(200)/50
Frame: t=t-0.05;rad=sin(t)/2+0.5
Point: x=cos(i*tpi)*(rad+v/5);y=sin(i*tpi)*(rad+v/5)

This will seem very complex at first, but let's look at it step by step:

Init - First we set n to 300, so that the superscope draws 300 different points. We also set the variable t to 0. Then, we set the variable tpi to twice the arc-cosine of -1. If you do the math, that means 2 times Pi (6.28....). Don't worry, it's just a trick to prevent you from having to type the number pi yourself, which is a useful number.

On Beat - Every beat, the variable t will be increased by a random integer number from 0-200, divided by 50. So that means a random decimal number from 0.00 to 4.00.

Per Frame - Every frame we decrease the t value slightly. We also calculate rad by taking the sine of t and scaling it a bit. If you know that a sine is a repetive wave-shape and that t is decreased slightly each frame, then you'll understand that the rad value will slowly pulse from 0 to 1 and back, except every beat. Then t gets modified drastically and the rad value jumps.

Per Point - Here we do the actual points. In our equation x and y are coordinates of a point on a circle. The circle has radius rad plus the current sound-value divided by 5. To make sure we traverse a full circle, we multiply i (range 0-1) with 2 times pi, so we get a range of 0-6.28...

Now you have a superscope that draws a spectrum or oscilloscope circle with a jumpy radius.

Another aspect of the superscope is colour. You can either use the (boring) color selector at the bottom, or you can write your own equations for the variables red, green and blue. They range from 0-1 and contain the value for their color. Let's spice up our superscope by adding this to the "on beat" equation:

cr=rand(100)/100;cg=rand(100)/100;cb=rand(100)/100;

And this to "per frame":

red=cr;green=cg;blue=cb;

What's going on here? Every beat we set cr, cg and cb to a random value in the range 0-1. Every frame, we assign these three to red, green and blue. Couldn't we just assign them directly 'on beat'? Nope... AVS resets them every frame with the color defined by the color-selector at the bottom.

So there you have your own groovy, color-changing superscope. It looks neat if you remove the t-changing on beat and combine it with a Trans / Water filter.

WHOA!

That was absolutely magnificent. Thanks!

trans...

...didnt any of the stuff i sent u help any?

Oh yeah, definitely 🙂

But you hafta admit,

Unconed was just alittle more detailed.. 🙂

Thanks

Thanks for the help guys. Though i guesss this means more work for me. 😁

I posting this reply so people can read Unconed's explaination of scopes....


NoSage|Jason

Just wondering, is it possible to specify the exact coordinates for each dot (so you could create images and shapes out of dots)?

yes, it is possible but a whole lot of work, especially(sp) when using 3D scopes.. here's an example that siddharta_one gave me once, but try to get Deamon's newest pack (hypernation) too, it has a 3D rendered cross..

OK, had a look at both the box example and Deamon's cross. It's for the Christian vis thing, so both of them help.

EDIT: I'm still unsure of how to determine the coordinates for each pixel, if ayone can help by creating a vis with just 2 pixels in different postitions, it would be much appreciated.

It's easy to do. All you have to is use a custom point count then use equal( to set the locations.
In Init

Xloc1=-.5;//sets the X coordinate of our first point to -.5

Xloc2=.5;//sets the X coordinate of our second point to .5

Yloc1=-.5;//sets the Y coordinate of our first  point to -.5

Yloc2=.5;//sets the Y coordinate of our second point to .5

In Perframe

p=0;//reset perpoint counter to 0

In PerPoint

p=p+1;//adds one to out perpoimt counter for every point being drawn

x=equal(p,1)*Yloc1+equal(p,2)*sloc2;//When P equals 1, sets the X coordinate to our Xloc1, and when P equals 2, sets our X coordinate to our Xloc2

y=equal(p,1)*Yloc1+equal(p,2)*Xloc2;//When P equals 1, set our Y coordinate to our Yloc1, and when P equals 2, sets our Y coordinate to our Yloc

You can copy and paste this into a superscope and fiddle with it if you want to. If you are sill confused just tell us what you are having a problem with or you could search the fourms for Superscopes or Per-Point, those might bring up some helpful things.

//edit-Fixed a few little mess-ups and "fixed" format

Somewhat simpler (but the theory is the same):

per frame:



p=0; (this is to reset the counter p you use for each point)



per point:



p=p+1; (meaning for every point p is increased by one).

x=0.1*equal(p,1)-0.5*equal(p,2);

y=0.4*equal(p,1)-0.35*equal(p,2);

In this case (and in all cases) you use a counter variable for each point. This counter is called p in this example.

Since we know n is the number of points to render, we set n to 2 in this case, otherwise the rest of the points have the coords (0,0). The equal statement is 1 or 0. It's 1 when p=1 for the first point, and p=2 for the second point. The rest of the points have coords (0,0), but since we don't have any other points, we don't have to bother about that. The reason why it works:

It works Per Point. It's really important to realise this. The code is run for every single dot. P is increased every dot, so we can number them. Point 1 equals p=1 etc. After that, you only have to enter coordinates for each point, and use the counter. Mind that N has to equal the last number of P or the last dots won't be drawn.

The statement:

If you multiply (*) any number 0, the result will be 0. If you multply by 1, the result will remain the same. Knowing this, we can do things like x=0.3*equal(p,1). In normal language: X equal 0.3 if p=1, otherwise it'll be 0.

Anyways... its easy to store points to megabuf(), like tell a scope run only once. wile doing it, store points to gmegabuf() and every other frame just load the points from buffer.
Im too lazy to write an example now :P

Thanks for the help.

EDIT: S-uper_T-oast, your code has a few mistakes in it, but I got the general idea. Thanks.

EDIT 2: Still having trouble. I'm trying to get a line from the center of the screen to the top-right corner, but I keep getting a line from the center straight down. My code:
In Init

Xloc1=0;   //sets the X coordinate of our first point

Yloc1=1;   //sets the Y coordinate of our first point

Xloc2=0;   //sets the X coordinate of our second point

Yloc2=1;   //sets the Y coordinate of our second point

In Frame

p=0;  //reset perpoint counter to 0

In Perpoint

p=p+1;  //adds one to out perpoint counter for every point being drawn

x=equal(p,1)*Xloc1+equal(p,2)*Xloc2;  //When P equals 1, sets the X coordinate to our Xloc1, and when P equals 2, sets our X coordinate to our Xloc2

y=equal(p,1)*Yloc1+equal(p,2)*Yloc2;  //When P equals 1, set our Y coordinate to our Yloc1, and when P equals 2, sets our Y coordinate to our Yloc2

Set n to the exact number of points, or else AVS sets all other numbers to 0

Edit: Instead of using Xloc1, Yloc2 and stuff, you can also use the numbers directly, since you'll only need them once
simplest example:

---init----

n=1

---frame---

p=0

---pixel---

p=p+1;

x=equal(p,1)*.5;

y=equal(p,1)*-.25

This sets a dot somewhere in the upper right of the screen.

You forgot to change the X coord, Braininator. Try:

Per frame:

p=0;



per point:



p=p+1;

x=0*equal(p,1)+1*equal(p,2);

y=0*equal(p,1)-1*equal(p,2);

I have used a minus in the Y coords because AVS has the Y coords from top to down from -1 to 1 for some strange reason. Somehow it's faster for your computer, though I don't know why.


To Jaak: Please write an example, I don't understand it as well, and I really like to know how to use the buffers.

The y starting at -1 instead of 1 is quite simple to understand: A PC 'thinks' from the upper left of the screen to the lower right. It's more that we as humans are strange, by starting at a high number and lowering our way down along the Y-axis 🧟

And I don't think it has something to do with the X-coörds

Deamon here ya go:

note that bla is set to 0 in end of frame, so that all loops will be run only 1 time after loading the preset
😉

Originally posted by Deamon
You forgot to change the X coord, Braininator.

I have used a minus in the Y coords because AVS has the Y coords from top to down from -1 to 1 for some strange reason.

Yeah, worked it all out. Already knew about the whole minus thing, just changed the wrong vars and forgot the minus. Thanks for looking at it.

Originally posted by ;-c ,rattaplan
Instead of using Xloc1, Yloc2 and stuff, you can also use the numbers directly, since you'll only need them once

Good point, but I wish to keep the variables so it makes it easier to change them. (I plan to have a few of them!)

@Jaak

...And here is where coding becomes programming
-Guess what? I love it!!!

By the way if you want to interrupt your shape somewhere, use:

skip=equal(p,4)+equal(p,6);

This will hide the 4th and 6th line segment (or 3rd and 5th, depending on how your counter works).

First bit of success!

Basically, the variables thing means you could set up a template for creating images in the superscope. All you'd need to do is create a number of different points (20 perhaps), set their coordinates to 0, then when wanting to create a new image just change the amount of points required and change their coordinates. So simple now!

Another way of using p2p scoping (point-to-point, which is what you're doing now...) is to let AVS figure out the coords instead of giving them. Here's a nice example I just made. I'm quite proud of it 🙂. Take a good look at the Texer II code. Don't bother the 3D stuff around it, the part that matters is:

init

pd=0.4                          //Distance between points



Per frame

xp=-1;                          //starting positions every frame

yp=-1;



Per point

xp=xp+pd;                       //gridforming

yp=if(above(xp,1),yp+pd,yp);



xp=if(above(xp,1),-1,xp);       //grid reset if N is set too high

yp=if(above(yp,1),-1,yp);

It took me some time to figure this out, though it's quite simple now I've got it. I was playing with extra counters first, but after some puzzling, I realised I didn't need any. Let's see if you can figure out what's happening yourself, it's better that way than if I say it right now 😉.

Other dudes around here, tell me what you think. Me personally likes it a lot.

[edit]
Read the comment for code explanation
[/edit]

Next job with my stuff, trying to get points to change on beat.

by the by, next time please don't revive three year old threads....

lol, I didn't even notice that 😛

Neither did I, sorry about that!

Originally posted by Atero
by the by, next time please don't revive three year old threads....

rofl.

i think that i too have been guilty of this once so thats all i'll say right there. it is strange how old threads get resurrected.. i think its probably the result of the search feature being used then replying to a post without thinking. or do people really search through page after page of winamp forum archives by hand (mouse i suppose)?

🤪

of course it's the search option, no-one would read about 50 pages of forum topics if maybe their question is already answered. Certainly not new people.