Darwinbots Forum
Code center => Suggestions => Topic started by: Numsgil on October 13, 2005, 06:12:39 PM
-
Just because I'm masochistic:
This would be incredibly diffiuclt, but it might be fun to model a relativistic simuklation with the bots such that general relativity rules hold true.
Some effects might be bots' DNA executing slower (over several cycles) compared to other bots the faster they go.
Shots could be said to be light particles, and so would travel at a uniform, large speed.
Maybe the speed of shots/light limit is, say, 200 twips/cycle.
Some other effects, it's been too long since I even cared enough to crack open a physics book.
The hardest thing is basically getting into the mindset to figure out how the heck to model it. The math is always really easy.
-
But then we would have something like infinite mass, and length etc!
It will but extremely hard to program, and what would the fun be, if a bot fills the entire screen, because it moves at the speed of shots... ;)
Do whatever you want, but make it optional.
-
What about the spatial transformations that occur when a bot reaches the speed-of-shot 's'?
Could a bot travel faster than this and end up further back in the sim amongst less-evolved bots?
-
what would this accomplish as far as helping bots evolve?
iow ...what would be the advantages/disadvantages to
any particular organism? what difference is it going to
make as far as increasing the ability to survive?
or is it just some more math to play with? ;)
at any rate ...
no need to make it complicated ...
we only need to simulate/approximate some change ...
the 'actual' change due to the differences in the relative velocities
we are using here would be infinitesimal, eh?
it might be interesting to simply have a 'time' slowdown inversely
proportional [or whatever] to the velocity of the bot making it's
commands take longer to process/execute ...
from 'our' inertial frame of reference ...
as the observer of the entire sim universe.
I don't know that there would be any advantage in altering
the velocity or duration of the shots. ...
but who knows?
just don't know how/if it would contribute to the evolution
of the bots or the simulation ...
seems to me we already have way too many variables as is
to allow much control of the directing things take.
which brings up another point ...
how is the initial seed to begin the randomization selected?
where is the seed? do we have control of it?
iow ... if we don't begin a simulation with the same seed ...
then any variations we might be experimenting with ...
tryng to determine what changes result in a particular outcome,
are bogus as each run will have a variable that we have no
control over. is there some way that we can select the seed
at the beginning of a simulation ... one we can repeat?
just wondering.
~griz~
-
I am not understanding relativity theory completely, but I can say you DEFINITELY
NOT understand it.
There is NO absolute speed in the universe, thats why when an object moves RELATIVE to another object both objects can say I am resting right now and the other object moves and is therefore slowed.
(darwinbot only uses an absolute speed relative to the grid)
Also I think an fast object does not get wider, it shortens (distance is reduced)
Well I probably should know more details :/, cause I am studying phsics, but I am not that good, soI so not know anymore right now.
-
I am not understanding relativity theory completely, but I can say you DEFINITELY NOT understand it.
if you say so ... :D
~griz~
-
Also I think an fast object does not get wider, it shortens (distance is reduced)
nothing gets shorter ...
or more massive ...
or slower ...
within ones own intertial frame of reference. ever.
these things only 'appear' to be altered from 'other' inertial frames of reference ...
and why one has to employ the lsuch things as Lorentz transformations to understand
what takes place in another reference frame from within one's own frame of reference.
that's why it's called relativity my friend. ;)
here is a decent spiel on such things:
http://cmtw.harvard.edu/Courses/Phys16/l1_...x/l1_latex.html (http://cmtw.harvard.edu/Courses/Phys16/l1_latex/l1_latex.html)
~griz~
-
Hehe, okay, I'm going to go back and reread alot of those posts. In the mean time:
1. This would be quite optional, and quite pointless. An excercise in idleness really. I have no idea what it would entail or even look like in practice. I have a hard time keeping all of relativity in my head at once. It's really quite confusing from a Newtonian perspective. If nothing else, it would give me a chance to refresh myself in relativity.
The effect on evolution? I wouldn't even venture a guess. I wouldn't even venture a guess that there would be an effect.
2. Since it takes infinite energy to approach the cosmological speed limit, and bots at best can output a finite amount of energy, bots will never achieve infinite mass/length/or anything else like that. Neither can they travel faster than the cosmological speed limit, and thus travel back in time or otherwise do some really weird things that way.
3. Since all inertial (non-accelerating) frames of reference are valid, setting the computer and user as the "stationary observer" is perfectly acceptable. If you don't understand how the twin "paradox" is resolved, this might not make any sense to you.
More when I actually read through all of what you guys think...
-
which brings up another point ...
how is the initial seed to begin the randomization selected?
where is the seed? do we have control of it?
iow ... if we don't begin a simulation with the same seed ...
then any variations we might be experimenting with ...
tryng to determine what changes result in a particular outcome,
are bogus as each run will have a variable that we have no
control over. is there some way that we can select the seed
at the beginning of a simulation ... one we can repeat?
Not too long ago I set an option to allow user set random seeds. This is in the second panel of the options page.
You are absolutely right, without it simulations become much less controlable.
Excellent article BTW. Everyone who is slightly fuzzy on relativity should read it as a refresher.
-
There is NO absolute speed in the universe
Quite wrong. The speed of light (electromagnetic propogation) is constant for all inertial frames. Thus the speed of light is always absolute.
(darwinbot only uses an absolute speed relative to the grid)
Also true, and since the grid isn't moving, we can say the grid acts as valid inertial frame of reference.
I think, there might be some issues with distance and non-Euclidean space, but now I'm getting into theoretical geometry, which is a horse of an entirely different color.
Also I think an fast object does not get wider, it shortens (distance is reduced)
Depends upon which frame of reference we are considering. Objects that are traveling close to the Speed of light relative to us become forshortened in the direction of motion.
Basically squashed.
Well I probably should know more details :/, cause I am studying phsics, but I am not that good, soI so not know anymore right now.
It's way interesting and mind blowing stuff. On the bright side the math is all really, really, easy.
-
Well Numsgil, why not begin on the E-grid instead coming up with new ideas again...
I mean, after every new release you come up with another wild idea, which have no real effect for the progress for DB going into the next phase.
Some are okay, others?? Are a little less thought out.
We should really get going on the grid and other things which are more related to evolution and the well being of our bots.
Okay, it will be funny to play with, but wait to add it after the grid, it will be much easier to add shot-speed and that sort, after the grid is finished, because if you first add this, and then the grid, you will properbly have to change the code for the former, and thereby increasing the possiblity of bug.
-
Haha, yeah, I keep putting off the grid. It's like some giant stone wall, and this Jericho refuses to come tumbling down.
I'm not super eager to add this Relativity thing right right now. The "Planet Eaters" thing I suggested took me like 4 months to program. I generally have two projects I work on:
1. Needs to be done
2. I want to be done
I usually work on the first type for a while, then work on the second type to keep me entertained. The second type are usually not more than 10-20 hours of combined research and programming (compared to like 30 to 70 for major new features like the physics overhaul, the splitting of the VB code from one form into 12 different modules, etc.)
What features do you feel are less thought out? I generally try to post major features here so we can all discuss them, and then usually refine the ideas in the process.
-
Really what will it give evolution if we add this?
More fun, when getting tried of the usual DB, for us, but the bot!?
Why do they need to be tortured so much, just because we want fun!?
Do you poke the eyes out of a mouse when you want fun?
I mean ;)
-
I don't think you can argue that a relativistic environment would be easy to adapt to. I can barely understand it half the time. I have absolutely no idea what sort of bots would evolve in it, but I would love to find out :D
Basically I'm slowly turning DB into not just a Life simulator, but a reality simulator. An ideal reality that obeys all the math laws, but still.
-
I just thought about something...
What about making DB multi-dimensional!!!
Maybe xD!!! Were x can range from 0 to 10!
That could be funny! ;)
-
Think about it!
A 2D bot meets a 10D bot! Wow! The 2D bot wouldnt even be able to se the 10D bot... The 10D bots shots will be in 9D! Haha I really like this...
I think I must go to bed now, or else I am going insane!
-
Actually, I suggested a while ago (4 or 5 months ago I think) that if we changed the graphics to openGL, we could implement a 3D version.
Most of the functions would work very similarly if you extend them to the 3D case. If nothing else it would look cool, and we could have some slightly more interesting screen shots to show people.
-
we have relative size, relative speed, relative mass, relative dobbler effect
-
Hmm Greven, explain in detail what you mean with Multidimensional?
Oh and Griz do not lecture me about thinks I know ;P
In a way you lectured me with the same fact that made me think you did not understand the theory.(I meanr it gets shorter from an outer point of reference
in direction of movement)
And Num, stop tricking me with my formulations.
The speed of light can only be reached by particles without mass, so there is no absolute speed" for partcles with mass". Although that was quite a clever trick Num XD.
-
See, you always get confused when you talk about relativity :wacko:
It'd basically work thusly: Each bot would need to be given a "clock rate", that can decrease as they approach the maximum speed.
Each bot would become more and more ovular as it goes faster and faster. But only in the direction of movement.
Each bot may or may not increase in mass as it goes faster and faster. I'm having trouble finding more than passing references to this.
Everything would be from the point of view of a "stationary" observer, namely us, the computer and user. This is a valid reference frame since we aren't accelerating.
I see nothing intrinsically wrong with any of the above, but relativity is not my forte.
-
What about a random chance for a bot to "quantum tunnel" to a new part of the sim? :D
-
I don't think so :D
Let's pretend bots are macroscopic in terms of relativity mode.
-
Oh and Griz do not lecture me about thinks I know ;P
In a way you lectured me with the same fact that made me think you did not understand the theory.(I meanr it gets shorter from an outer point of reference
in direction of movement)
I wasn't I lecturing you ...
only responding to your assumption/accusation/statement that I did
not understand relativity ... in a manner which I found to be quite unfriendly ...
seeing as you have no idea of my own background or experience.
remember ... when you point a finger at someone ...
three of your own fingers point back at you. :P
you are, of course, quite welcome to your own interpretation/perception
of reality ... [it's ALL relative]
all I ask is that you own it and not project it onto me/others ...
you know ... apply the transformations first. ;)
And Num, stop tricking me with my formulations.
The speed of light can only be reached by particles without mass, so there is no absolute speed" for partcles with mass". Although that was quite a clever trick Num XD.
close ... but not quite correct my friend.
theory, speculation, concepts, ideas, formulations ...
are only what they are ...
theory, speculation, concepts, ideas, formulations ...
approximations ... tools we build to help us in our understanding.
but they are only the tools we use to measure/decipher reality ...
not reality itself ...
let us not confuse the map with the territory ...
or the menu for the meal.
so ...
if I tell you that there are indeed particles that do move faster then light ...
will you again accuse me of not knowing what I am speaking about?
there are indeed particles that do travel faster than light.
I have pictures. ;)
~griz~
-
But even for a macroscopic object there is still a very tiny chance for quantum tunneling.
Did you know that since matter and energy are completely interchangable in the laws of physics, it is theoretically possible to get a diffraction pattern by firing a bunch of Airbusses through a narrow slit? :D
(Don't try this at home boys and girls! The mess would be very difficult to clean up)
-
(Don't try this at home boys and girls! The mess would be very difficult to clean up)
Boy, now he tells me :wacko:
-
The only thing that really gets me would be collisions between bots.
That's where my head implodes, because from my frame of reference, if two bots collide head on each going .99c...
-
The only thing that really gets me would be collisions between bots.
That's where my head implodes, because from my frame of reference, if two bots collide head on each going .99c...
guess you could set up an experiment ...
gather some empirical evidence.
you and Bots drive, I'll be the independent observer. :lol: hehehe
~griz~
-
Aww, I never get to be the independant observer :ph43r:
-
Welwordion
the only thing I mean with multi-dimensional is ... multi-dimensional.
It were meant as a joke, but well we wouldnt even be able to see how an actual 10D bot looks like.
-
Sure we would, we could see whatever bits of it passed through our spatial 3D world. If time were one of its dimensions, we could see it move about and such.
Of course, time operates slightly different than spatial dimensions. Being 1 way is a prime example.
Any higher order bot in a 2D bots' world would probably look like a circle of varying radius.
-
HEHE!!
From a bot's eye perspective.
1 See a nice little circle, attack it then watch as it moves more into your 2d world and find that the thing is 3000 times your own size and packed with more firepower than a Warthog anti-tank plane.
2 Run like the proverbial wind before it mashes your ass!
-
Might be fun if a 3D mode ever gets created to have a 2D plane and a 3D world sort of coexisting. The 3D bots could move above or below the plane, and the 2D bots could eat the 3D bots as they passed through the plane.
Probably only let 3D bost feed on 2D bots if they fire a shot in the same plane as the bots, or otherwise prevent them from massacring 2D bots that can't fight back.
-
with VHs, cant we also have VWH(veggie worm holes)? like uf tge gravity is strong enough some creates a random contection with another veggie that might be one way or 2 way but also relativity is to hard, a bot shall have different mass and time slowing (this I belive is possible) depended on from what bot we look. But a quantum mode instant would be funnier. a mode where the impossible can happen with a probebility. like its 0,1% that 2 bots get fused toghater and become one. or that nthey fall through a micro wormhole. if we want alot of modes then we can also add 3D mode. not that we see they move in 3D but we see them pop up in a new place by "passing" through the 3rd dimension. im brainstorming here.
-
Worm Holes are more theoretical than actual. The fact that veggy singularities form at all is primarily because the integration method is Euler. That is, velocity = velocity + acceleration.
Gives wacky values when things get too extreme.
-
maybe, but its fun with wacky modes. but this aint to wacky, simple tell the program to move the bot when its X wicks away from the veg to the other veg that is the end.
-
Any higher order bot in a 2D bots' world would probably look like a circle of varying radius.
you guyz familiar with Flatland?
http://www.geom.uiuc.edu/~banchoff/Flatland/ (http://www.geom.uiuc.edu/~banchoff/Flatland/)
~griz~
-
I was but I'd never managed to find a copy to read.
Thnx Griz.
-
ive heard of it in a book about higher dimensions. and yes it would look like it, I´ll read it griz, thank you.
btw griz, do you know that if you change ur Z to a S it becomes the swedish word for pig?
-
ive heard of it in a book about higher dimensions. and yes it would look like it, I´ll read it griz, thank you.
btw griz, do you know that if you change ur Z to a S it becomes the swedish word for pig?
that's why it's a Z and not an S ...
griz is short for Grizzly ... ursus horibilis ...
literally, "bear horrible" ...
very a large n.american omnivore ...
that eats just about anything that crosses it's path.
hmmmm ...
may have to name my first killer bot ursus horibilis. ;) hehehe
~griz~
-
oh well, I´ll still call you gris, or pig, or maybe even piggy, nha kidding with you ;)
-
I always think of "Grizzly Adams", which I only really know from the cartoon family guy when they make fun of it.
-
oh well, I´ll still call you gris, or pig, or maybe even piggy, nha kidding with you ;)
;)
pigs get a bad rap ...
they are very intellegent beings ...
and only wallow around in their own s*%t 'cause we confine them to it.
I've always thought that humans contaminate their own environment
more than any other animal ...
and that calling a someone a pig IS indeed an insult ...
to pigs! :lol:
Oink!
~griz~
-
I dont think the pics complain about living there, ive not yet heard a pig say "I dont wanna leave like this, comon fellows, lets revolt and destroy the humans"
and im glad for that.
-
Try reading 'Animal Farm' by George Orwell
-
its a bok, it have nothing to do with reality, but might be fun. maybe in a alternetive universe the animals rule the world and we humans are used as pets
-
"Animal Farm" has alot to do about reality. Which is why I never read it :lol: