Archive: Some help plz


17th August 2003 05:34 UTC

Some help plz
Click here to goto download page


Remote Sensor is just my second pack of avs's and I think they're just ok (so do other people). I was just wondering what I could do to make it a better pack. Plz don't suggest any big coding or stuff like that because I can't do it.


Also, remixes would be nice.


17th August 2003 05:48 UTC

Re: Some help plz
It's quite simple: make better presets. Get that down and you'll be on your way to AVS stardom.

Originally posted by pure-kryptonPlz don't suggest any big coding or stuff like that because I can't do it.
...Then learn how? :rolleyes:

17th August 2003 05:57 UTC

You can't do big coding because you think you can't.

The key to a successful avs pack seems to be a mix of originality, technicality and plain old aesthetic beauty. This is why I think that EL-VIS, Tuggummi and UnConeD are probably the most popular avs artists around, most of their work is a good balance between these three things. All three of these things are also linked together, you can't be original without technical skill to some degree because more people have less technical skill and so there are fewer original ideas to be made with moving particles and blitter feedbacks, also a lot of the most attractive effects require some technical skill.. 3D for instance. Its also a pretty good idea to steer clear of filling a pack with lots of really similar presets or sequels.. I think this is why my eighth pack failed so spectacularly to have the effect of the two previous ones (and why my next pack will be great :p). There are a lot of different things that make a good avs pack and different people have different ideas on this, the best way would be to come up with your own conclusions by looking at the avs of others and deciding which *you* like best... then you know what sort of stuff you should make. I don't know about you but I tend to watch my own AVS much more than the AVS of others.. I think this is the best way to judge your own stlye.. by how much you like your presets. Even looking back at my more noobish presets I still really enjoy a lot of them.


17th August 2003 06:04 UTC

The thing is, I don't understand dm's and scopes.
Does it just take time?


17th August 2003 07:27 UTC

Code is just really simple logic.

Here... think about it for a little bit.

x=2*i-1;
y=0;

An SSC.. by reading the help that comes with it you should be able to figure out what this is without making it for real right? 'i is the position of the scope (0.. 1)' so by multiplying it by two and subtracting one the new range is -1 to 1 so by setting x=2*i-1 we are making the scope draw from -1 to 1 in the x axis.

I really can not get what is so hard about it.. its not like you need to actually know anything more than the expression help, and that text crammed onto the SSC page, to code it. To code something specific however will just be a thing that comes with practise and mathematical knowledge. To start its probably best to experiment with the varying mathematical functions and remember the laws of code:

variable=variable(operator)variable(operator)function(variable,variable);

Avoid crazy invalid statements like x=y=0; that just confuses AVS. Or x+y=1;, it will not draw a straight line because AVS is not a math interpreter, it will do nothing because you are not assigning a value to a variable you are trying to assign it to x+y which you can not do.

If you understand the basic rules then the only limit is your mathematical and programming knowledge.. expand both.


17th August 2003 07:41 UTC

Okay, this is the first section of one your scopes(jheriko) and I don't get where you got u&v from?


u=tpi*i;
v=200*u;
x1=0.2*cos(v)*abs(sin(2*u))+2*cos(u)*sin(u);
y1=0.2*sin(v);
z1=-0.2*cos(v)*abs(cos(2*u))+2*sin(u);

17th August 2003 13:35 UTC

don't do stuff that complicated at first. Look through this. I haven't finished it, but it should help you.


17th August 2003 15:55 UTC

try atero's primer ffs.


17th August 2003 22:14 UTC

I didn't get u and v from anywhere, I made them, thats the whole point of scope code, you define values of variables and use them to set the x and y co-ords of the scope.

u ranges from 0 to 2pi and v from 0 to 400pi, i then use these as parameters in that equation to define x1,y1,z1.. which is a set of coordinates on that shape.. the per point code is called repeatedly with a different value for i each time (going from 0 to 1) which in turn moves u and v through their ranges creating a sequence of points which lie on the shape. i then convert x1,y1,z1 into and x and y by doing 3 rotation matrices and a perspective calculation later on in the code.


17th August 2003 22:16 UTC

Atero told me that AVS Primer v2.0 will not all be released this year, and it will be an ongoing project after the resource pack is released. The first part (v1.75), however, should be out within a week, and after he gets all the bugs worked out he will release another version (v1.85).

But it's going to be a really helpful resource once it finally comes :up:


17th August 2003 23:34 UTC

Hopefully he'll find some way of explaining how code works on a really fundamental level so that we can avoid this sort of thing. I find it hard to explain because I understood code from the first time I ever used it in qbasic.. a simple set of instructions that execute in order. I really can't see how anyone can get confused with the basic concept of code because it really is so really really *REALLY* simple. I can understand someone not being able to so some specific thing in code, because they can't see how, though. Perhaps you non-coders just don't understand what '=' does, I really don't know.


17th August 2003 23:56 UTC

I didn't understand when i first played with it. Before AVS i was completely ignorant of code of all sorts.

I find it very irritating that every time we make an explanation of anything, as soon as it goes off the first page it disappears into the void of threads, never to be looked upon by human eyes again. SEARCH FEATUTRE PEOPLE.


18th August 2003 03:02 UTC

yea well this is a pretty general query. try searching the avs forums for "help" and see what comes up, raz.

pixelcraft, you bastard, that was a *private* message, nobody was supposed to know! :p ;)


18th August 2003 03:40 UTC

Oh man, I'm sorry! :eek: You should have told me it was a secret....

/me kicks myself :(

Originally posted by jheriko
I really can't see how anyone can get confused with the basic concept of code because it really is so really really *REALLY* simple. I can understand someone not being able to so some specific thing in code, because they can't see how, though. Perhaps you non-coders just don't understand what '=' does, I really don't know.
Huh? Jeez, come on Jheriko, not everyone is smart like you, give us a break.

18th August 2003 09:18 UTC

Learnt C language first.


18th August 2003 09:33 UTC

You don't have to be smart to understand code, if you understand what all of the symbols mean then you are able to understand how code works. All the code does is what it is told, in the exact sequence it was told to do it.. what could be simpler?


18th August 2003 09:35 UTC

Learning how to code takes time, of course it depends on the person and his/her will to learn new things. Also if you don't have any resourcers where you could learn things (like the forums) it probably takes longer. I have avsed since 1999 and i didn't learn how to code even the simpliest ssc until my 4th pack and by then i haved avsed for about a year or maybe a 1½ years, don't remember so well :p But the reason that i learned so slowly is that i didn't have any resources for information (nor i even didn't know where to look for) so i just basicly had to test&try and try again and again and again... well you get the picture :)

So it takes time, but don't be discourage by it. Only think hard before you decide how envolved you want to get in avs, because it really grabs you thight and won't let go :D And there are more important things in life than avs.


18th August 2003 22:40 UTC

I spend a few hours a week doing avs and i'm slowly working on a pack. I don't say i'm leaving because i can't avs 90 hours a week, i merely continue with what i'm doing and work on it when i can. I am enveloped in it but i don't have to focus my entire life around it.

Atero if they search for help on the search feature you would come across the many many threads in which people have asked for general help and we have pointed them in the right direction. This isn't a freak occurence as you know.


20th August 2003 07:33 UTC

Originally posted by Tuggummi
Learning how to code takes time, of course it depends on the person and his/her will to learn new things.

It's gonna take forever then.

20th August 2003 07:59 UTC

Actually try and learn instead of sitting there like a child saying can't do it won't do it without even taking a look at it. If that's your attitude towards learning things then don't expect praise and/or respect. From anyone. Ever.


20th August 2003 08:20 UTC

I've already made a good dm (I myself think it is), but I'm just making it better. And one other one that moves a "Trans / Interleave" down the y axis. Then i'll release the it at devArt and here to see what othe people think of it.


20th August 2003 20:30 UTC

I thought I could never learn it until I actually sat down and started to learn and experiment. At first I just typed random code, not understanding it, but now I can do more than I ever thought I could. :D And still there's lot of time ahead of me to learn... ;)


21st August 2003 13:35 UTC

Yeh, same here. Thought I could never code, and I'm now even capable to define a rotating cube in a 3d space. Not that that is original, but hey, you've got to learn somehow.

Here is another simple DM for you:



Pixel:

d=d*0.75;
r=r+rt;

frame:

rt=rt+rts;

on beat:

rts=getosc(0,0,0)*0.1



This will cause the whole screen to rotate slowly in a random direction. "d=d*0.75" is used to zoom in a little on the preset so you won't get any ugly sides. "rt=r+rt" will cause to add rt always to r, and not only once. "rt+rts" will cause to add rts always to rt, and not only once. "rts=getosc(0,0,0)*0.1" means that every beat there will be taken a number from the oscilope data between -1 and 1, and than smallened a bit, cause otherwise it'll go to fast. For now, remember getosc as a sort of rand(). The only profit in this movement is that it'll go left and right, instead of only one of the two.

This is my most used DM, and also the one I learned first :).

Play around with it a bit, slowly it'll all become clear for you.

21st August 2003 16:40 UTC

I still can't 3D at all, but I will try to experiment with that when I return from my vacation...


21st August 2003 20:34 UTC

Quote:


21st August 2003 22:16 UTC

Pixelcraft, you should download my newest pack "HyperNation". I've just put it up, and maybe there are some things in it you might be able to figure out. It's all low-tech so you can always give it a shot. Good luck, mate.


21st August 2003 22:38 UTC

Um, I already did that, and wrote a whole review too...


23rd August 2003 13:05 UTC

I've made a ssc moving particle but I need to know how to make it react on beat (radius gets bigger)


Init:
n=4
On Beat:
dt=(rand(100)/50)/10 ; bx=(rand(100)/50)/25 ; by=(rand(100)/50)/25 ;
Per Frame:
t=t+dt ; fx=fx+bx ; fy=fy+by ; rc=rc+0.2*getspec(0.2,0.1,0) ; bc=bc+0.2*getspec(0.1,0.1,0) ; gc=gc+0.2*getspec(0.3,0.1,0);
Per Point:
x=sin(t+xf*2)/1.1;
y=cos(t+fy)/1.1;
red=max(sin(y+rc),0.1);
green=cos(y-bc);
blue=max(abs(sin(y+1+gc)),0.3);


I want it to be like this on beat :: but off beat like this .

23rd August 2003 13:56 UTC

Originally posted by Deamon
....Here is another simple DM for you:



Pixel:

d=d*0.75;
r=r+rt;

frame:

rt=rt+rts;

on beat:

rts=getosc(0,0,0)*0.1


Do you think i'm completly fucken disabled?

I know how to do the above.

23rd August 2003 20:51 UTC

For the fisrt thing, there are much better ways to, (in my opinion), make moving particles, but the only way I could think of making yours more beat reactive is by increasing the number by which you divide the sin( ans cos( by on beat using a variable that then decreses to a set point on frame. Second, you don't need to get so goddamn crabby when somebody gives you a helpful suggestion, and from your response, I do think you are completly disabled.


24th August 2003 01:59 UTC

I'm sorry, I should have told you that I could do them. I apoligise (spelling?)


24th August 2003 03:37 UTC

s-t, i made a real moving particle system a long time ago, you'll have to search the forums for it tho...and yours isn't nearly correct, though it's a common misconception. for a perfect, nondecaying orbit: an anchor point is randomized on beat, and the third derivative of the motion at any time is proportionate to the distance from that point. so you have (ax,ay)=anchor, (px,py)=particle, (mx,my)=motion vector (second derivative). at every step calculate the vector between (mx,my) and (ax,ay) on each axis, scale it down so you actually get a smooth movement, and add it to (mx,my). to make the orbit decay then just do a logarithmic decay on (mx,my) every step as well (e.g. mx=mx*0.95)


24th August 2003 03:52 UTC

I'll look that up, I made my own semi-real moving particle system not that long ago, but I mainly just made up a system based on particle position and particle motion with a decay.
Sure thing Krypton, it's no problem, you just got to keep your cool and not get angry at people no matter what they do, or how many times they ask the same questions without searching the fourms.


25th August 2003 14:21 UTC

Sorry Pixelcraft, watched your reply right after replied here.


26th August 2003 04:10 UTC

Heh, that's alright, it happens :)



Originally posted by Deamon
Play around with it a bit, slowly it'll all become clear for you. *inhales deeply*.........................*exhales* THANK GOD. :D