You can completely simulate or model ''A'' world within a computer, but not ''THE'' world - the one we percieve to be real. (For ''world'' here read ''universe'')
Also, Anon. Poster (May 15th, 4:20:26), there are programs that write themselves or each other. Try researching ''polymorphic software'' or REDCODE and the related world of Tierra.

quote:

---

Original post by Mark_T

You can completely simulate or model ''A'' world within a computer, but not ''THE'' world - the one we percieve to be real. (For ''world'' here read ''universe'')

---

Yes, and a computer which is a machine inside a world A (that could be our "real" world or a computer simulated world) and which follows the laws of A''s physics, can''t simulate A correctly if A is no fractal world (fractal means that parts of A is exactly the same as A if you scale it to the size of A).

Visit our homepage: www.rarebyte.de.st

GA

Interesting thread that got started here, and although I know next to nothing about either Evolutionary Programming, or Evolution at all really (apart from the basics) I feel that two points have been overlooked here.

First off, the only change mentioned here is between generations. That is, there has so far been no mention of one specific individual learning new behaviour, only passing behaviour to offspring. Not a very interesting world to live in, is it? So there should be some possibility that the "behaviour matrix" of an individual changes over the course of its lifetime based on experiences. For instance, if a creature starts giving away food when it has in abundance, if it never gets anything back for it, and consequently goes through a period of low food-supply, it might learn that giving away food maybe isn''t such a hot idea after all. Thus you could add another piece of information to a creature, namely it''s ability to change it''s own behavioural patterns. Which, come to think of it, sounds suspiciously like low-level intelligence, but I don''t really know what I''m talking about here...

Second off, there should be .. ahem.. sex.
Now, I''m clearly not talking of dirty pictures here, but sex is an extremely important matter in evolution. Basically it means that the genetic material of an offspring is a mix between that of its two (or more? Could be interesting...) parents, thus creating a greater diversity in the species.

-Neophyte

~Death awaits you all, with nasty, big, pointy teeth~

Oops. Imagine my embarassment when I realized I had posted a reply to a 5-page thread after reading only 1 page, and that the 4 last pages strayed off from the initial subject so that my post is now largely off-topic.
Oh well. Since it was on-topic for the start of this thread, and I don''t think I''m completely way off base with what I wrote I''ll let it stay, and not delete it.
Hope you''ll forgive me

-Neophyte

~Death awaits you all, with nasty, bit, pointy teeth~

I like the idea of a world created by the quake engine,
then adding in the creature to learn.

Anyway in order to have the creature/model/thing learn
wouldn''t it need to have a few basic rules like self
preservation (like eating to stay alive along with hunting and staying away from BAD food)? Wouldn''t it need something that would force it to a new stimulus? Another thing would the thing try using assimilation then accommodation to the point of actually learning something new?

CEO of existance

thats force it to act on a new stimulus?


CEO of existance

I think the way to do this is to create atoms, and design a level editor that let''s you paste em together. This and the set of physics to go with it, yep all of em.
Now it could take years to build something incredibly simple, but once it has been done, we are one step closer to simulating this universe.

The_Minister
1C3-D3M0N Interactive

Neophyte: Don''t forget that also our evolution started with creatures which were hardly able to learn. The ability to learn has developed in the evolution process. And also experience can''t be inherited by the children.

Visit our homepage: www.rarebyte.de.st

GA

Right, you cannot code something accurately without all of the information.
If you were told to make a program that found the area of a triangle given the length and breadth as input by the user. However, unless you have the correct formula [(l*b)/2] then you are scuppered. So, you are given (l*b/k) where k is an unknown constant, you can''t find it out (not using that formula alone)!
Now! My point! You can make a world, with simple physics (keep it 2d, so there goes gravity and a whole load of other crap). Now what... There''s nothing in this world, loosely based on aspects of our world. When do things start appearing? You have an empty world, nothing is just going to simply appear, is it! Something has to change eventually, but that change shouldn''t be hard-coded. This is where I start to get lost I''m afraid. It''s just like, you cannot start a world and then call a function which creates an organism because it is inaccurate... Referring to my original point, we don''t know how life started, not properly, so how can we code the birth of life? Well, mail me if anyone doesn''t have a clue what I am on about, CertificateX@Hotmail.com
Also, I am interested in joining a team for a project on this topic. Mail me if you are interested in that too...

quote:

---

Original post by The_Minister

I think the way to do this is to create atoms, and design a level editor that let''s you paste em together. This and the set of physics to go with it, yep all of em.
Now it could take years to build something incredibly simple, but once it has been done, we are one step closer to simulating this universe.

The_Minister
1C3-D3M0N Interactive

---

Yes, you''re right about the ''atomic sim'' aspect but building a world wouldn''t actually take that long... Between one and ten lines of code per physical law, perhaps - and at the most basic level there aren''t many of those.
But can you imagine the processing power and memory you''d need to actually RUN a program like that?
Let''s see, 16,254,000,000,000,000,000,000,000,000,000,000,000,000 atoms in my glass of lemonade here... (believe me, if it''s a day this hot in England, you should be outside enjoying it, and not on the message boards)
I don''t know how much computer power we have at the moment, but I''m fairly sure it''s not enough. Back to the drawing board...