Prev: 9/11 was an inside job. NASA Flight Director Confirms 9/11 Aircraft Speed As The "Elephant In The Room"
Next: AI Programation was re Programation Atoms and the Void was Re: SculptingLinguistics was Re: Networking Veldts
From: rick_s on 23 Jun 2010 09:04

Lets take a quick look at the inner workings of a computer game in 3D.

x horizontal y vertical z depth

you iterate through the 3D dot matrix as shown previously
(the order is not important but what is important is that you want to be
able to identify and change the value of every dot in that 3D cube.

Which we will say is 720x720x720
x is assigned from 1 to 720 do
y is assigned from 1 to 720 do
z is assigned from 1 to 720 do

If that was just a 2D bitmap image we would omit z, and x would tell y
to repeat 720 times.

720 rows of 720 dots and the computer can set a color value of all those
dots in a second or less.
25 frames per second in a video.

In 3D you now have to do that 720 times.

so really what you want is for x to tell y to repeat 720 times, and y to
tell z to repeat 720 times.

And to do ALL of that 25 times per second.

And on top of that to be interactive, you have to set values of those
dots or only those dots within a small geographic location within the
screen. Where a character is animated and the scenery is static.

So if you feel like doing some math, you can find out how fast a modern
quad duo can iterate through that dot matrix.

See if the computers we have now have the speed you need which is 25
frames per second. 25 cubes per second.

Then you need to check values and set values for each dot and test the
computational speed.

If you can access and set the values for each dot then you need to try
some complex functions and see how using real numbers as a for instance
slows down the machine.

If you have all that together you are good to go man.

Not only that but map bitmaps onto 3D surfaces, mesh drawings that sort
of thing and then since you have dots, you can have smoke clouds
fireworks anything that scale of size can imitate in terms of atoms or
anything else in the known universe since what you have to work with is
atoms and the void between them.

With sufficient speed you can take 3D images of people by surrounding
them with cameras, or wire them up with sensors to imitate human
movement since trying to fake human movement will always look like Bruce
Lee vrs Iron Man in youtube.
http://www.youtube.com/watch?v=Ywu1DeqXTg4

Here the choreography is fantastic but not enough frames in between to
be fluent so a morphing program would help that by inserting in between
frames.

It calculates the area each of two frames its xyz coordinates and as it
iterates through the matrix, it compares the image on the right with the
image on the left and creates as many in between frames as you need
automatically so you don't have to draw them and so that your animation
is fluent.

You see you could just scan those objects, and then using wire frame
diagrams, give them the movements. That is how they did Avatar.
Bitmaps mapped over wire frame characters.

The best animation though is still from Japan. Best artists.

Now we have the ability to display the screen. The actual physical
memory would have to mirror that dot matrix. Plus each dot would have to
have its own set of switches. Enough to set a string of values, a set of
values, that are common to all dots. That would be graphic memory, plus
cache memory and whatever.

So we can throw 25 3D cubes of dots to the screen/projector

Then you have to be following a programmatic theme which in a game is
some storyline and then you are checking for input and adjusting the
values of the dots accordingly.

That's enough to do animation in proper 3D so you could walk around it.






 | 
Pages: 1
Prev: 9/11 was an inside job. NASA Flight Director Confirms 9/11 Aircraft Speed As The "Elephant In The Room"
Next: AI Programation was re Programation Atoms and the Void was Re: SculptingLinguistics was Re: Networking Veldts