Jump to content

  • Log In with Google      Sign In   
  • Create Account


Eralp

Member Since 08 Jun 2007
Offline Last Active Oct 03 2013 11:32 AM

Posts I've Made

In Topic: Aurum Game

14 September 2013 - 06:17 AM

I liked this http://img5.imageshack.us/img5/9692/zrqb.jpg over this http://imageshack.us/a/img109/7428/i5ys.jpg

 

I'm not an artist but I feel like the second-type started to look old and 'cheaper'


In Topic: Game based on real-world time

14 September 2012 - 10:19 AM

I wouldn't think simulating something in-game will take so much time that the user thinks the game crashed. I am not talking about 3d graphics simulations etc, I am only talking about the decision AI for example in sims.

Your simulation method shouldn't play sounds or show any graphics. You need to seperate those things based on which type of simulation you are making. If you extract user interface like sounds and graphics from your simulation method, I doubt it will take any longer than an average load time to simulate years.

In Topic: Chess AI with Neural Networks

02 September 2012 - 08:34 PM

It was this link:

http://www.chesscirc...RY-FOR-OCTAVIUS


I also found an extra link which claims that his NN may/can learn valid moves, but I couldn't compile it yet because it is written in vb6 back in 2001.

http://www.planet-source-code.com/vb/scripts/ShowCode.asp?txtCodeId=22202&lngWId=1

I mean, were people trying this in early 2000s but they gave up?

In Topic: Chess AI with Neural Networks

02 September 2012 - 07:44 PM

So are you pointing that the man I read about was not telling the truth?

Since my experience with AI is probably nowhere near yours I thought a pure NN AI could play chess with right training. It feels so intuitive. There has to be a set-up which makes this possible but I am not sure if that is feasible or not.

In Topic: Theory - ultimate AI, at atomic level

16 September 2011 - 04:42 AM

I can't understand why the thread creator referred as a troll here. I exactly thought the same thing, maybe we will play car racing games where the fuel goes to the engine and burns there due to chemical laws everything is calculated in atomic size, or maybe FPS games where enemies have circularity systems, lungs etc. and when you shoot them everything is realistic.

Of course they don't seem reasonable right now since Moore's law won't be valid for too long, maybe we need quantum computers or a change in the silicon material to achieve that.

Concerning the AI thing, if we COULD compute ALL atoms' movements. It is the same thing as AI. Your neural network, brain, and its decisions may look like VERY COMPLEX to you but we have been kinda training them for years and years without many people trying to understand the basics behind our decisions and movements.

But before going up to the most complex neural network, lets think of a one cellular living. When you are aiming to understand 5th dimension it is better to start with understanding the transaction between 2nd and 3rd dimension. WHAT is exactly giving life to that cell, can anybody explain that? In my opinion all the things it makes are due to physical forces between atoms & molecules. The DNA can be treated as brain of the cell but as far as I know they created an artificial DNA and put it inside a cell and it continued "living". What if we replaced all the molecules in one cell? Or samely what if we copied all the molecules in one cell to another place, will it be cloned or won't they live due to lack of "soul" :) I think it will live. And then it goes up like what if we copied all the molecules in a human, what do you think it will be?

To sum it up in my understanding of the world and universe, EVERYTHING is based on physics, and physical laws. I am not talking about the ones we modelled right now, our laws can be wrong but afterall there is something according to which particles are "moving". We know that Newton's approximation was false, maybe Einstein's will not work under some conditions too.

When the forces are about atoms & molecules, we created chemistry to make approximations and create straight-forward rules, sometimes not so straight-forward but compared to simulating and calculating everything in sub-atomic particles they are very well straight-forward.When we taught chemistry we generally hear things like, under normal conditions this is true but else this law will fail. This is because as I said they are approximations but that is not something bad because we need to do them to actually achieve something and make use of chemistry. Nothing in chemistry can contradict with physical laws, and every chemistry law can be derived from physical laws, I think everybody agrees with me here.

Quick example: Law: Polar and Apolar liquids do not resolve in each other. When a question is asked you can use this law to answer and to back your thoughts. But you can very well derive it from physics. Since polar molecules have + and - poles, because an atom's protons is closer they pull the electrons to one side, they pull themselves like magnets and they stick with each other this makes the other liquid's molecules go up or down.

When we needed to examine the interaction of very big numbers of molecules and atoms we created biology.
One of its basic laws: The cell membrane is waterproof, of course not fully but in a reasonable way. This is due to molecular interaction between the molecules of the cell membrane and water molecules. You can go up the hierarchy till physics to explain this.

PARTNERS