/sci/ - Science & Math » Thread #11937988

74KiB, 458x333, main-qimg-fa181c77585491c6ba2490c4213c38cb-c.jpg

View Same Google iqdb SauceNAO

Anonymous Sat 25 Jul 21:40:39 2020 No.11937988 View Reply Original Report

Quoted By: >>11938012 >>11938032 >>11938099 >>11938214 >>11938234 >>11940014

Can someone redpill me on entropy?
It is generally said to be a loss of information of order in a system but given that the amount of blue and red in pic related does not change and that you need the same amount of information to explain its position in said system does not imply loss of information.
I don't fucking understand this shit.
help pls

Anonymous

View Same Google iqdb SauceNAO Howard Bloom.jpg, 78KiB, 1280x720

Anonymous Sat 25 Jul 2020 21:49:25 No.11938009 Report

Quoted By:

https://www.youtube.com/watch?v=PIa9hjsIQJ4

Anonymous

Anonymous Sat 25 Jul 2020 21:52:31 No.11938012 Report

Quoted By:

>>11937988
>eu garbage
into the trash it goes

Anonymous

Anonymous Sat 25 Jul 2020 21:53:31 No.11938013 Report

Quoted By: >>11938025

OP, this may clear things up: https://physics.stackexchange.com/questions/75146/entropy-and-information

Quoting that:

>I believe conflicting concepts of information and entropy under the same names are at work here.

Basically there's a terminology issue.

As for this particular example, I'm confused about your question. To describe the position and color of all of the balls using the least amount of information, you need to use more information to describe the configuration on the right than the one on the left. So the one on the right has more entropy.

As a dumb example, if the number of balls is known to the recipient ahead of time, you could say "half blue half red" to describe the left one, and you'd have to say "red blue red blue blue red red red ..." for the right. The right will always require more information.

Anonymous

Anonymous Sat 25 Jul 2020 21:59:06 No.11938025 Report

Quoted By: >>11938046

>>11938013
You could say half are blue and half are red still. Care to extrapolate?

Anonymous

Anonymous Sat 25 Jul 2020 22:01:23 No.11938032 Report

Quoted By: >>11938050 >>11938051

>>11937988
In the picture to the left you need only one bit of information to describe the state, namely whether the blue balls are to the left of the wall or to the right. The other picture however allows for a lot more of configurations and you need more bits to specify one of those configurations.

Anonymous

Anonymous Sat 25 Jul 2020 22:07:52 No.11938046 Report

Quoted By: >>11938050

>>11938025
Not the same anon, know nearly nothing on the subject formally but:
Left: blue x < 0, red x > 0
Right requires more information to specify

Anonymous

Anonymous Sat 25 Jul 2020 22:09:21 No.11938050 Report

Quoted By: >>11938057 >>11938203

>>11938046
>>11938032
>>11938032
Suppose red and blue are +1 and -1 respectively. The whole system is 0 which is purple. One configuration or another within the box doesn't change this description.

Anonymous

Anonymous Sat 25 Jul 2020 22:09:32 No.11938051 Report

Quoted By: >>11938085

>>11938032
Okay you got me here but it still feels odd.
Like if you go through the position which in itself is information you need one bit of information to say red or blue no matter the configuration given each ball has some sort of identity.

Anonymous

Anonymous Sat 25 Jul 2020 22:11:44 No.11938057 Report

Quoted By:

>>11938050
i started thinking about this more, I think it is an analogy which demonstrates a principle once it is known, but fails to explain it on its own merit. Disregard my questions.

Anonymous

Anonymous Sat 25 Jul 2020 22:17:13 No.11938071 Report

Quoted By: >>11938085

Can someone explain the relationship between chaos and entropy if there is any?

Anonymous

Anonymous Sat 25 Jul 2020 22:22:24 No.11938085 Report

Quoted By: >>11938094 >>11938161

>>11938051
>given each ball has some sort of identity
This part is what you get wrong but its not explained in the picture so not your fault exactly.
For this to work you do have to make the important assumption that each ball is indistinguishable from his same-color peers. In more practical terms this would be gas molecules in a box where its physically impossible to tell two different particles apart and only the configuration matters.

>>11938071
Entropy is defined as
$S = - k_B ln(Z)$
where Z is the amount of states that a system can assume and $k_B$ is the Boltzman constant.

Chaos is when shit gets lit on fire yo.

Anonymous

Anonymous Sat 25 Jul 2020 22:29:37 No.11938094 Report

Quoted By:

>>11938085
One one hand it feels like I have learned something but on the other hand it feels like you choose the one perspective that makes you able to explain the problem away.
Anyways I am going to reflect on the information this thread has givin me to try to reach some understanding.
Thank you.

Anonymous

Anonymous Sat 25 Jul 2020 22:32:17 No.11938099 Report

Quoted By:

>>11937988
In terms of thermodynamics, I've heard it being defined as the diminishing in energy available to produce work. For example, the sun is a giant ball of plasma, but it has a huge mass, and, therefore, huge gravitational potential energy, which createsw pressure on its core, smashing atoms into each other, fusing them, creating new elements, creating photons, which escape to the solar system, hit earth, feeds the climate and organic processes, which dissipates that heat until there is no more energy that can create that amount of work, beacuse it's all been dissipated.

or so the mainstream physics go, more or less

Anonymous

Anonymous Sat 25 Jul 2020 23:02:50 No.11938161 Report

Quoted By:

It looks like so far, this guy>>11938085
is the most correct. That equation, where he points out Z is the amount of states that system can assume is a very good explanation. And then where he points out "chaos" is when shit gets lit on fire is a way of saying ignore the word "chaos".

Put it this way: The system on the left has 2 sections, one red, one blue. How many ways can you move that around? probably some number less than 20. The system on the right is all jumbled up, so this can be rearranged some very, very large number of times, thus entropy is much more here. This is the *only* reason why people consider the word "chaos".
Now, when people say "well, the one on the right is unchanging and normalized, so how is that entropy (chaos)?" they are missing the part where the number of ways the system can be rearranged is the important part. This is why the final state of the universe is max entropy even though literally nothing will be going on...you can rearrange 10^100 individual atoms many, many, many more times than you can rearrange 10^30 stars/planets.

Anonymous

Anonymous Sat 25 Jul 2020 23:06:34 No.11938174 Report

Quoted By:

The information theoretic definition of entropy came well after the statistical mechanical/thermodynamical definition and is largely meaningless in a physical context. It is more significant for CS and information theory, signal processing than anything else. You gain literally no insight from understanding it as the amount of information in a system.

Anonymous

Anonymous Sat 25 Jul 2020 23:23:07 No.11938203 Report

Quoted By:

>>11938050
That's not the description, that's the sum of some arbitrary value assignment.

If you want to describe the left imagine you can say
"The balls are arranged in a 12x11 grid, where the left half is blue and the red half is red"
If you want to describe the right side you have to specify the individual balls since they're not clustered.

Anonymous

Anonymous Sat 25 Jul 2020 23:28:08 No.11938214 Report

Quoted By:

>>11937988
Entropy and thermodynamics aren't actually real.

Anonymous

Anonymous Sat 25 Jul 2020 23:36:53 No.11938234 Report

Quoted By:

>>11937988
It's true that it takes the same amount of information to describe the microstate of both sides, i.e. exactly which molecules are where. But thermodynamic entropy is also related to how the system looks on the large scale: the macrostate. The amount of entropy, more or less, is the amount of information needed to encode the microstate GIVEN that you already know what the macrostate is. Tons of the possible microstates look homogenous on the large scale, like on the right, so encoding a specific microstate based on that requires a large number of additional bits (high entropy). On the other hand, comparatively few microstates separate the molecules, so knowing that they're separate narrows the number of microstates down significantly, requiring less information to encode one of those few microstates (less entropy).

Anonymous

Anonymous Sun 26 Jul 2020 15:49:54 No.11940014 Report

Quoted By:

>>11937988
High Entropy actually translates to high information content. Entropy is really talking about the randomness of the data distribution. On the left, the red and blue are sorted in a simple way. You could describe the position of each ball, but you could also just say "blue on the left, red on the right" and express the distribution just as accurately. No such simple description exists for the randomly arranged balls on the right. The only way to describe the situation on the right is to specific the position of every ball - there is no overarching pattern. In CS, low entropy means more compressible (that is the file size will be smaller after a compression algorithm). This is because the data is highly patterned and you can simply express those overarching patterns rather than specifying every bit. High entropy means little or no compressibility because there is no overarching pattern - the only way to express the data is by stating the value of every bit exactly.

Capcode	All Only User Posts Only Moderator Posts Only Admin Posts Only Developer Posts
Show Posts	All Only With Images Only Without Images
Deleted Posts	All Only Deleted Posts Only Non-Deleted Posts
Ghost Posts	All Only Ghost Posts Only Non-Ghost Posts
Post Type	All Only Sticky Threads Only Opening Posts Only Reply Posts
Results	All Grouped By Threads
Order	Latest Posts First Oldest Posts First

Your latest searches