19 Nov 12
Time marches relentlessly forward for you and me; watch a movie in reverse, and you'll quickly see something is amiss. But from the point of view of a single, isolated particle, the passage of time looks the same in either direction. For instance, a movie of two particles scattering off of each other would look just as sensible in reverse – a concept known as time reversal symmetry.
I thought this was untrue. After all, entropy goes one way and is largely responsible for the arrow of time. Surely this one directional entropy shows up even with small numbers of particles?
19 Nov 12
Originally posted by twhiteheadWould entropy affect something really really small?
[quote]Time marches relentlessly forward for you and me; watch a movie in reverse, and you'll quickly see something is amiss. But from the point of view of a single, isolated particle, the passage of time looks the same in either direction. For instance, a movie of two particles scattering off of each other would look just as sensible in reverse – a conce ...[text shortened]... row of time. Surely this one directional entropy shows up even with small numbers of particles?
Is it possible to be called a 'thing' and be so small as to deny entropy?
19 Nov 12
Originally posted by karoly aczelI would think 'small' would be relative term, since even by the sizes of sub particles, they are many orders of magnitude bigger than the quantum limits of planks constant, 1E-34 meters or so. From that perspective, those particles would be like planets.
Would entropy affect something really really small?
Is it possible to be called a 'thing' and be so small as to deny entropy?
- Joined
- 31 May 06
- Moves
- 1795
20 Nov 12
Originally posted by twhiteheadNope.
I thought this was untrue. After all, entropy goes one way and is largely responsible for the arrow of time. Surely this one directional entropy shows up even with small numbers of particles?
Entropy only appears with large numbers of particles.
The reason being that entropy is a function of the number of possible different states of a system.
There are many more disordered states than ordered ones.
As you increase the number of particles then the number of possible states increases and the ratio
of ordered to disordered states goes down.
So entropy is (hah, buzzword) an emergent phenomenon.
It's a bit like moving from quantum mechanics that applies to small scales to classical physics at large ones.
21 Nov 12
Originally posted by googlefudgeMy understanding was that it arose directly from basic mechanics and was a property of the direction of time.
Nope.
Entropy only appears with large numbers of particles.
The reason being that entropy is a function of the number of possible different states of a system.
There are many more disordered states than ordered ones.
As you increase the number of particles then the number of possible states increases and the ratio
of ordered to disordered sta ...[text shortened]... moving from quantum mechanics that applies to small scales to classical physics at large ones.
So explain to me why a large number of particles in an ordered system evolves into a disordered system, but not the reverse.
If time is seen as flowing backwards, why is disordered to ordered now possible? Surely there is some difference between the two time directions that causes this?
I need to give it some more thought.
21 Nov 12
4 edits
Originally posted by twhitehead
My understanding was that it arose directly from basic mechanics and was a property of the direction of time.
So explain to me why a large number of particles in an ordered system evolves into a disordered system, but not the reverse.
If time is seen as flowing backwards, why is disordered to ordered now possible? Surely there is some difference between the two time directions that causes this?
I need to give it some more thought.
So explain to me why a large number of particles in an ordered system evolves into a disordered system, but not the reverse.
I am not an expert on this so please would someone correct me if I am wrong but, if I understand this correctly, the answer is that it only generally does but the reverse can happen! It is just that, the larger number of particles involved, the less probable it is to do it in reverse during any arbitrary chosen period of time T. Consider there being a continuum from a very 'small' number of particles to being a very 'large' number of particles and along that continuum the probability of 'reverse entropy' during period of time T steadily smoothly and seamlessly shifts its value.
Lets consider this just for thermal movements of particles:
Imagine a closed system containing, say, nothing but a simple gas.
Now imagine two distinct areas of gas distinct from each other only by a difference of temperature so that the gas in one half of of this closed system is at 0C and the other at 100C.
Now, if we ignore thermal convection and radiative movement of heat and only consider the movement of heat through the collisions of randomly moving particles, you can imagine a simulation where it makes sense that, at the boundary between the two areas of gas, generally, the hotter faster-moving particles in the warm side would transfer more momentum to the colder slower-moving particles in the cold side than the other way around simply because the hotter particles are moving faster. So that explains why heat would generally tend to move from the hot area to the colder area in this case than the other way around.
BUT, this is only generally the case because of the way probabilities work out on a large scale.
Even if there are millions of gas molecules in this closed system, there must be a small probability ( albeit vanishingly small ) that, within a particular moment of time, the random motion of particles could for an instant be just exactly such that heat travels the other way causing a measurable greater temperature difference rather than a smaller temperature difference.
The larger the number of particles involved, the lower that probability. But, if you imagine, say, just four particles involved, and imagine the simulation of that, you should see it would happen very often. And, if you imagine just a few more particles involved, you should see it would happen less often. But, no matter how many more particles you add, you can never quite make it never happen.
21 Nov 12
My understanding is that for any given system of multiple particles interacting, it is easier to calculate the past than the future. This is why we 'remember'
the past but not the future. It is my understanding that it is a fundamental property of the arrow of time and that it is directly related to entropy.
21 Nov 12
8 edits
Originally posted by twhitehead
My understanding is that for any given system of multiple particles interacting, it is easier to calculate the past than the future. This is why we 'remember'
the past but not the future. It is my understanding that it is a fundamental property of the arrow of time and that it is directly related to entropy.
My understanding is that for any given system of multiple particles interacting, it is easier to calculate the past than the future.
if true, this is news to me!
This is why we 'remember' the past but not the future.
I am uncertain what you mean. Surely we don't remember the past because it is easier to 'calculate' it than the future but rather because we have experienced the past?
And do you imply here that if it was the other way around i.e. easier to calculate the future than the past, we would 'remember' the future? If so, I am not sure how that would work. I would think we can merely 'predict' but never 'remember' the future even if and when we have infinite confidence of our prediction.
- Joined
- 31 May 06
- Moves
- 1795
21 Nov 12
Originally posted by humyBingo.So explain to me why a large number of particles in an ordered system evolves into a disordered system, but not the reverse.
I am not an expert on this so please would someone correct me if I am wrong but, if I understand this correctly, the answer is that it only generally does but the reverse can happen! It is just that, the larger nu ...[text shortened]... o matter how many more particles you add, you can never quite make it never happen.
Exactly right, and well put.
22 Nov 12
Originally posted by humyThink of it like this:
I am uncertain what you mean. Surely we don't remember the past because it is easier to 'calculate' it than the future but rather because we have experienced the past?
And do you imply here that if it was the other way around i.e. easier to calculate the future than the past, we would 'remember' the future? If so, I am not sure how that would work. I wo ...[text shortened]... r 'remember' the future even if and when we have infinite confidence of our prediction.
The universe at point X is in a given state. The laws of physics determine the interaction of all the particles/energy in the system. The laws of physics can tell you what 'came before' in the negative time direction and what 'comes after' in the positive time direction. If there is practically no difference between the two directions then the past and future would be equally predictable. The illusion that you have experienced the past and not the future would disappear. You would remember both past and future with equal clarity.
However, the universe is not like that. We can predict the past much better than we can predict the future. We cannot however predict the past perfectly (due to quantum mechanics) and this results in the conclusion that multiple pasts exist (just as multiple futures exist). The difference being that there are far fewer possible pasts than futures, so the past seems more solid.
22 Nov 12
1 edit
Originally posted by twhiteheadI will have to spend some time thinking about that.
Think of it like this:
The universe at point X is in a given state. The laws of physics determine the interaction of all the particles/energy in the system. The laws of physics can tell you what 'came before' in the negative time direction and what 'comes after' in the positive time direction. If there is practically no difference between the two directi ...[text shortened]... ce being that there are far fewer possible pasts than futures, so the past seems more solid.
22 Nov 12
Originally posted by humyIt is not just human memory that suffers from the arrow of time bias. If you analyze a rock you can determine much about its history millions or billions of years and possibly even know quite a lot about where its constituents came from prior to it being formed. All this based on the laws of physics.
I will have to spend some time thinking about that.
Yet those same laws of physics tell us practically nothing about what the rocks future holds.
Clearly there is something about the laws of physics that allows a rock to retain detailed records of its past but practically nothing of its future. This is tied into causation in that we see the past causing the future and not the future causing the past.
Entropy is an integral player in all this.
- Joined
- 10 Dec 06
- Moves
- 8528
22 Nov 12
1 edit
Originally posted by twhiteheadWell, we can predict a rocks future ( albeit not indefinitely, but neither can we predict its past indefinitely with any certainty) by using its past as a a model for its future ( I agree that we have to model its past first, so that gives its past some level of priority). Specific events that happen to the rock which might catastrophically shape it, may not be predictable using the model, but the bulk effects of all the natural forces on the rock over a larger time scale can lead us to a conclusion about the rocks existence that's probably not to far off (ie. that the rock will probably return to its constituents)
It is not just human memory that suffers from the arrow of time bias. If you analyze a rock you can determine much about its history millions or billions of years and possibly even know quite a lot about where its constituents came from prior to it being formed. All this based on the laws of physics.
Yet those same laws of physics tell us practically not ...[text shortened]... sing the future and not the future causing the past.
Entropy is an integral player in all this.
- Joined
- 10 Dec 06
- Moves
- 8528
22 Nov 12
I just had a thought, what if our perceptions are travelling at the extreme of some "time wave". We can see the past of the wave but only to somewhere near the zero point of the wave at this present time, but at some point in later time we may be able to begin to see the crest of another wave forming, and thus be able to predict our distant future by assuming that the wave properties remain the same. Thus the further in time we move, the more of the future of the wave can be predicted.