A forum for topics that don't fit elsewhere. Introduce yourselves to other members of the forums, discuss how your name evolves when written out in the Game of Life, or just tell us how you found it. Forum rules still apply.
BokaBB wrote: October 27th, 2020, 7:25 am
I made an agreement with Hunting where he investigates my rules if he can and if he wish in exchange for lots of fun time in Sandbox.
Have a good day!
BokaBB
I saw. Can we talk about the CA now? Do you want more properties?
Oxygen is used in the body and carbon-dioxide appears,it must be removed.
Have a good day!
BokaBB
777
I CAN APGSEARCH NOW!
Sure, I was a bad person, but I have changed myself.
I'd love to befriend anybody who's interested.
Have a good day!
BokaBB wrote: October 27th, 2020, 4:44 am
We can make it to some degree,but the cellular degree will be too hard and everything would be big.
Have a good day!
BokaBB
This is quite an ambitious project. We should start way smaller. Please look at my chemistry CA proposal.
Agree
The idea of building a human in CA is
uh
Ridiculously impossible with today's knowledge
Observe that we can't even grasp how human brains work-- they're millions of times more complex than the most complex neural networks we've designed-- which nobody really understands the workings of very well either.
I agree that we should first merely try a chemistry CA
κ is measurable iff there is a nontrivial elementary embedding j:V→M (M transitive) with critical point κ
BokaBB wrote: October 27th, 2020, 4:44 am
We can make it to some degree,but the cellular degree will be too hard and everything would be big.
Have a good day!
BokaBB
This is quite an ambitious project. We should start way smaller. Please look at my chemistry CA proposal.
Agree
The idea of building a human in CA is
uh
Ridiculously impossible with today's knowledge
Observe that we can't even grasp how human brains work-- they're millions of times more complex than the most complex neural networks we've designed-- which nobody really understands the workings of very well either.
I agree that we should first merely try a chemistry CA
Ok, what proposals do you have for the chemistry CA?
Well first off, we should use a very simplified chemistry; even irl chemistry is far too complex to emulate in a Moore neighborhood CA with a reasonable amount of states. That should be a constant limit: keep the chemistry simple. Don't try to model real chemistry, just keep bare essentials. Bonding is essentially the only requirement, but I'm not sure whether bonds even require states; just make it so that atoms touching each other in their VN neighborhood are bonded (Moore neighborhood just helps for edge cases etc). Then, make it so only some quantities kinds of bonds are supported by each element; an analogue to carbon for instance supports either two double bonds at 180 degree angles, a single double bond and two single bonds, or four single bonds. An analogue to oxygen might support either a single double bond or two single bonds. Then just cover edge cases in the rules so that when an atom exceeds its capacity of bonds-- ex if the carbon analogue has three single bonds and a double bond, say, or two double bonds at 90 degree angles, it changes its state to, for instance, "carbon analogue + dropping the atom to my left", after determining it should drop that bond, which causes said atom to move as a lonely atom. This is a reasonable amount of simplicity, as with only a few types of atoms it should work all right (note however that you might need multiple states for each element: "double bonded oxygen", "single bonded oxygen" for instance. Finally, make unbonded atoms move, and make them capable of moving in all directions, moving in the direction they were initially kicked by their molecules. Don't try to make it so atoms try to keep their "valence shell" full; atoms with incomplete bonding will naturally fill their "valence shells" when they are bombarded by passing "free" atoms.
I'd like to see the carbon/oxygen analogue CA, it should be possible to create in less than 256 states and would give an idea as to whether this idea could be interesting to increase in element count.
κ is measurable iff there is a nontrivial elementary embedding j:V→M (M transitive) with critical point κ
Moosey wrote: October 27th, 2020, 8:26 am
Well first off, we should use a very simplified chemistry; even irl chemistry is far too complex to emulate in a Moore neighborhood CA with a reasonable amount of states. That should be a constant limit: keep the chemistry simple. Don't try to model real chemistry, just keep bare essentials. Bonding is essentially the only requirement, but I'm not sure whether bonds even require states; just make it so that atoms touching each other in their VN neighborhood are bonded (Moore neighborhood just helps for edge cases etc). Then, make it so only some quantities kinds of bonds are supported by each element; an analogue to carbon for instance supports either two double bonds at 180 degree angles, a single double bond and two single bonds, or four single bonds. An analogue to oxygen might support either a single double bond or two single bonds. Then just cover edge cases in the rules so that when an atom exceeds its capacity of bonds-- ex if the carbon analogue has three single bonds and a double bond, say, or two double bonds at 90 degree angles, it changes its state to, for instance, "carbon analogue + dropping the atom to my left", after determining it should drop that bond, which causes said atom to move to a free space nearby. This is a reasonable amount of simplicity, as with only a few types of atoms it should work all right (note however that you might need multiple states for each element: "double bonded oxygen", "single bonded oxygen" for instance.
I Was thinking six states should be enough. I think heat/energy could be possible. Atoms would radiate "photons" through empty space and
would also exchange heat through contact.
Moosey wrote: October 27th, 2020, 8:26 am
Well first off, we should use a very simplified chemistry; even irl chemistry is far too complex to emulate in a Moore neighborhood CA with a reasonable amount of states. That should be a constant limit: keep the chemistry simple. Don't try to model real chemistry, just keep bare essentials. Bonding is essentially the only requirement, but I'm not sure whether bonds even require states; just make it so that atoms touching each other in their VN neighborhood are bonded (Moore neighborhood just helps for edge cases etc). Then, make it so only some quantities kinds of bonds are supported by each element; an analogue to carbon for instance supports either two double bonds at 180 degree angles, a single double bond and two single bonds, or four single bonds. An analogue to oxygen might support either a single double bond or two single bonds. Then just cover edge cases in the rules so that when an atom exceeds its capacity of bonds-- ex if the carbon analogue has three single bonds and a double bond, say, or two double bonds at 90 degree angles, it changes its state to, for instance, "carbon analogue + dropping the atom to my left", after determining it should drop that bond, which causes said atom to move to a free space nearby. This is a reasonable amount of simplicity, as with only a few types of atoms it should work all right (note however that you might need multiple states for each element: "double bonded oxygen", "single bonded oxygen" for instance.
I Was thinking six states should be enough. I think heat/energy could be possible. Atoms would radiate "photons" through empty space and
would also exchange heat through contact.
Please do not give up on this project.
Have a good day!
BokaBB
777
I CAN APGSEARCH NOW!
Sure, I was a bad person, but I have changed myself.
I'd love to befriend anybody who's interested.
Have a good day!
BokaBB wrote: October 30th, 2020, 7:51 am
Just not to forget this.
Have a good day!
BokaBB
Well we I don't know what to post here exactly. This thread is a bit hard to progress in. If we do stuff in the chemical CA, maybe we can eventually be able to make something here.
BokaBB wrote: October 30th, 2020, 7:51 am
Just not to forget this.
Have a good day!
BokaBB
Well we I don't know what to post here exactly. This thread is a bit hard to progress in. If we do stuff in the chemical CA, maybe we can eventually be able to make something here.
I also think so. I just felt like you forgot this ambitious project. Please visit meteorological CA.
Have a good day!
BokaBB
777
I CAN APGSEARCH NOW!
Sure, I was a bad person, but I have changed myself.
I'd love to befriend anybody who's interested.
Have a good day!