Perhaps the best series of videos about statistical mechanics I've ever seen, and it's not even about statistical mechanics. Waiting for the algorithm to pick this one up!
I like your comment so much because it captures so well how I feel about this whole series. As we accidentally understand the different topics we encounter, we hopefully recognise the underlying patterns of approaching such topics and develop the necessary mindsets to tackle different problems in the future that seem to be unrelated to fluids.
I love the clarity of the shown thought process. Many others would have covered the entire video in 1-2 sentences, omitting all this detail. Thank you!
Awesome! This series is probably the best I've seen when it comes to understanding fluid simulations. I'm a chemical engineering graduate and I've always wanted to program a simulation as a hobbyist but I've never quite had enough of a solid grasp on the concepts to proceed with the implementation. I could've simply followed some random tutorial but I'd rather develop an understanding of the concepts being applied to able to code my first simulation. Can't wait for the next part!
Thank you for taking the time to write such a kind comment! I'm glad you like the general approach. Personally, I focus a lot on the conceptual side (maths, problem setting, changing perspective, asking the right questions, things like that ...) and it helps to see that others share these thoughts. I hope to see you in the next part!
Just discovered this channel and spent 2 hours watching one of the more precise and lucid explanations on the internet. As a graduate on these topics, I find your perspective is incredibly unique and worth sharing with everyone. Keep up the great work! Waiting for your next upload!
Your videos has simulations as exactly how I think and they completely present my imagination which is so beautiful. Thanks a lot for this wonderful work. I liked and subscribed already. Wonderful Channel!
I'm dazzled and speechless with your amazing video. I'm a fluids engineer working with statistical thermodynamics and your video made me recover my energy and motivation towards my daily routine. Damn!
Absolutely brilliant! As a chemical engineer I am delighted with this series. This is the most intuitive derivation of how macroscopic properties are derived from microscopic states. It took some years of studying to develop this picture for myself. And here you derive it all in a visually intuitive way, in such a short period of time. Masterfully beautiful. Your videos fill me with joy. I have a newfound appreciation for having studied this field.
Outstanding video as usual, and the thumbnail this time is a significant improvement! I'm so glad to see that your videos are starting to get the attention they should. Also good to see you made a Patreon! I imagine over time you'll reach many people who will be very happy to support novel content like this :)
I'm happy to hear from you. :) Thank you again for addressing the thumbnail issue last time. I thought carefully about your comment when creating this one! :) And of course, thank you for your support!
The video is super beautiful :) I'm wondering about one thing though: When simulating fluids with those SPH-particles the SPH equations (in the zero viscosity case) are (although derived from Euler equations which are derived from the Boltzmann equation which is irreversible) time reversible and using a reversible integrator and fixed point arithmetic, we can actually see the particles forming up again if we reverse the velocities after coming from such a formed state, as done in Globally time-reversible fluid simulations with smoothed particle hydrodynamics - that we can do this, however, comes from knowing the exact particle positions and velocities - only from the 1- particle density, it is as in the colliding spheres case - just unlikely to find this behaviour. The Boltzmann entropy (based on the 1-particle phase space density) increases (statistically) while the Liouville entropy remains constant. Edit: Actually forming a dam again and clumping up are very different phenomena. For a non / weakly compressible fluid to form a dam again is quite plausible, but the system will not clump up again working against pressure forces which normal particles - not our SPH particles - could do.
The real laws of motion always include both *repulsion* and *attraction* which is why the normal curve of probability in entropy (probably) makes all possible microstates actually happen. That's why so many folks get confused by entropy, and think that it means that things only ever fall apart, and it's why these computer models are not especially accurate to real life, because entropy actually means that things fall apart and form new things, over and over. Entropy is the same as evolution, with that messy sexual reproduction via natural selection (things coming together) and random mutation (things coming apart).
Sorry but no. Entropy is not evolution. If you look at each state of a time crystal you can say it evolves into the next state but yet time crystals have 0 entropy, they are non-entropic processes. Entropy is often misunderstood because it is just that, a hard concept to understand. Look up Shannon entropy to get a better understanding. Essentially entropy is the limit of energy transfer of the microstates. Very, very few things in this universe have 1:1 energy transfer, there is usually a resistance. (Time crystals are the only example I can think of) Hence, entropy increases for most systems. Essentially entropy means things lose energy over time. Hence why you can't have perpetual motion. I think what you mean is that like evolution, entropy is a quantity which exists ONLY for the system. If you take the particle out of the box there is no entropy. Which is true!
@@88Magician88 Yeah, it's unfortunate that so many kids were taught such a weird, old fashioned, backwards idea of what entropy (the laws of reality) is. Shannon entropy is what I'm talking about, which is evolution, as in the process of generating a family tree, where individual patterns combine and then divide, over and over again. As you say, it's the number of possible microstates in a system (macrostate). Pascal's triangle models this nicely, or a Galton board, if you're curious about what it looks like. What something appears as to an observer (anything that can change, and have multiple states) will vary based on what the observer is (how it's observing). So some things seem to not be changing (matter), and others seem to change rapidly (energy). If you experience things "losing energy over time" that's because you, as an observer, are focusing on the matter-like elements of it. In reality, the universe never creates, nor loses, energy or matter. It's always the same amount. Just differently distributed. Like how a fractal keeps changing while staying the same. Reality is just one massive family tree of patterns of varying stability and change branching and recombining, to evolve all possible states of existence. Stephen Wolfram is starting to explore this if you want a different approach then my own.
@@thewiseturtle I agree! (kind of) Whenever I heard entropy being described by a prof I always had to sort of tune it out because they would say something like "this isn't really how it is but it's an analogy to help you understand", if it isn't how it is don't teach it to me, ya know? I hesitate to say I fully agree because you use evolution, which is loosely defined. Time crystals evolve over time (yes they repeat, hence the crystal but they do evolve) but are non-entropic. So I can't say I agree that entropy IS evolution but I will say they are analogous. I do love the way you think about the universe. Not many people see it like you but I think your picture is a lot closer to reality than others! Well done!
Dont gravity and electromagnetism work thru the move along spacetime=attract, move opposite to=repel? Wonder if the notion of electrons being one-handed has anything to do with this asymmetry? I forget what I'm referring to, but something about them not being reflected in a mirror
@@oUncEblUnt420 I think you are talking about electron chirality. Electrons are left-handed (spin) and positrons are right-handed (spin). The Stern-Gerlach experiment was what they used to figure this out. Spin has nothing to do with gravity or E and M. Spin has to do with angular momentum. (If the electron was a macroscopic object then it would be spinning. It's not, therefore, spin is an objectively confusing name.) The "notion of electrons being one-handed" has to do with this chirality. The claim is that there should be an even amount of left-handed electrons (electrons) and right-handed electrons (positrons). We don't see this though, we see mostly left-handed electrons. Thus you could say that chirality symmetry is broken by electrons. However, there is nothing that states the universe must have chirality symmetry. We assume it does because then we can say the universe has parity symmetry, which would be nice mathematically but there is no physics that demands this, thus things that break parity symmetry (probably) aren't paradoxical.
I think its time to revisit my patreon account now :D I would easily pay for a full fluid simulation course of this quality! Thanks for the effort to create this
All your videos on a whole another level.. It's beyond awesome.. Great animations and great in depth explanation.. Great work through and through.. Loved it..
I'd love if you could do a video solely about how you go about making these videos (research process, tools used, etc.). The pacing, colours, animations; all fantastic!
Set up a particle simulation and witness energy un-dissipate? I have tried that too ;D When I learned about Poincaré recurrence theorem as a student, I thought to myself, doesn't this destroy the 2nd law of thermodynamics? You sir, have a very satisfying answer.
Could you help me to understand what program/libraries you used to make this? I really like the visualizations you use and want to start making physics education videos
Thank you for another great masterpiece! One question: should we consider this as a part 3 of the Understanding Fluid Simulation series? I see your comments below the video and asking because: (1) The video name doesn't start with the "Understanding Fluid Simulation" prefix and (2) title picture is different. Should we expect part 3 as a separate video or this is it?
Thank you very much! This part is meant to be a short follow-up, as it addresses things that I couldn't fit into part two. So it's kind of part 2.5, if you will. I plan to make the next part (3) more in line with the first two parts.
Most everything that happens in our animal level of the universe is pretty improbable. This is what we call free will. It might be deterministic, but it's also (likely) pure randomness. So everything that possibly can happen does happen, somewhere, somewhen. Of course, not everything is possible. But nitrogen glass does happen here and there. And so does your head. So while improbable, it's not impossible. Just like falling in love, or winning the lottery.
Sorry if this is a stupid question (since I am not a physicist, just a curious man), but don’t you need to add gravity? In zero gravity, fluids tend to stay in their lower energetic form, which is a spherical shape. If your model doesn’t include gravity, will the particles eventually acquire a less energetic formation? Or maybe this doesn’t apply to single atoms, but only to molecules with more complex interaction such as hydrogen bonds? Anyway, congratulations to your series, it is truly amazing!! You should keep the channel active! Thank you for all the effort you put in it!
Yooooo this is amazing!!!!! I really appreciate how UA-cam has become a platform that shares knowledge that bewilders my understanding of the world. I've always been a keen learner but holy cow. This is beyond that, eye opening tbh. Sir I want some advice. I already am somewhat familiar with coding and programming. I don't know OOP but i do know the use of structures in MATLAB and fairly complex stuff in Wolfram Mathematica. I want to learn how to build my own simulator of fluids so I can play around see what I discover. Which language would you recommend me? What are it's advantages and what's the main framework I need to understand to make it become a reality? I'm curios about C# as Sebastian Lague seems to use it. Python I'm familiar but not proficient. I am willing to really take the dive into it. But I don't know where to start!
I like this video but when you say "reversing" it initially confuses me. Obviously, if you could magically pause a flow and reverse all the particles, that new flow cannot defy laws of thermodynamics. But that is not usually what reversal means. If you were to simply run back the clock and track the particles in a simulation (not the real world) they would go back to the initial conditions. Anyway, thank you for the very insightful video.
when you are going from a lower-level model to a higher-level model, do you rigorously derive how their coefficients are connected so that they produce the same results, or do you approximate them until they sort of match? for example, in this case did you derive the macroscopic parameters such as viscosity from the microscopic parameters such as particle freerun length and radius or did you just tweak them until they sort of match?
Personally, I like the fact that there are different approaches that can be used depending on the main objective of the analysis. Your comment addresses this aspect perfectly! Since this series is about giving structure and looking conceptually at problems, knowing that there are different ways is a good placeholder for further discussion if needed.
Randomness, viscosity ,'nothing is entering or exiting in such a pure and sealed box.', .This was left out. Tending to an unrealistic odds of system development. Seemingly breaking law of entropy.
hmmm, so destroying localized information restricts the arrow of time 🤔 that got me a philosophical epiphany on the nature of our own reality... if quantum fluctuations are truly random, they do destroy localized information, and that would explain the "only forward" nature of time. Either that or I already drank too much
or the statistical nature of quantum mechanics is already hiding an underlying deterministic concept that doesn't really have a forward time direction. Maybe "forward" or "backwards" has no meaning at all its just happens to be either case (not both) which doesn't really effect anything as particles behave the same either way. macroscopically it's just statistic so certain unlikely low entropy states like sudden perfectly straight towering atoms or every particle accumulating in the corner of a room or similar meaningful stuff just don't occur or at least extremely rare.
@@ecicce6749 great answer. i'll add my 2 cents here. evolution laws do not require a definitive arrow of time to work, since they are symmetrical. I don't know if that's relevant, but I felt like I should say it.
Hello Braintruffle . Can you please suggest what you consider as the best fluid mechanics books (videos ) for a mathematician and a hydraulic engineer , and I need also books on the subject of Numerical Analysis ... thanks in advance
Wouldn't intermolecular forces introduce dissipation and prevent time direction effects? Or are you only interested in perfect gases, based on kinetic theory?
5:15 >never undissipates, even by simulating backwards in time wait wait wait. your microscopic equations are completely time-symmetrical. are you saying that the microscopic equations don't "undissipate" or the macroscopic ones?
i guess later you clarified that it's the macroscopic laws are non-reversible because they destroy the information about microscopic state (that is reversible by itself). which makes sense. but you probably should have explicitly said that at 5:15.
I’m sure you’re familiar with the SINDy work they’re doing at UW. Is the average probability density and dissipation already built-in to SINDy by the nature of sparse optimization? I’m wondering if explicitly incorporating this probability average could be used algorithmically to reduce the cost of hunting for sparse matrix coefficients, or if it is just as costly?
I actually saw nothing wrong with the first animation aside from energy not disipating. Think the reactions were 100% efficient therefore there was a sudden drop in entropy. This can actually occur in real life but far less and needs quantum tunneling, and if you want to avoid that in the above example you should either add a drag factor or some energy loss upon collision.
This series is going to be a masterpiece
The definitive guide, yes!
It already is a masterpiece.
I'm so happy you like it! Thank you!
it's already a masterpiece. It was pure joy to watch and follow along
@@maxjost9266 A phdpiece then
Perhaps the best series of videos about statistical mechanics I've ever seen, and it's not even about statistical mechanics. Waiting for the algorithm to pick this one up!
I like your comment so much because it captures so well how I feel about this whole series. As we accidentally understand the different topics we encounter, we hopefully recognise the underlying patterns of approaching such topics and develop the necessary mindsets to tackle different problems in the future that seem to be unrelated to fluids.
@@braintruffle Peak big brain behavior. Please reproduce.
I love the clarity of the shown thought process. Many others would have covered the entire video in 1-2 sentences, omitting all this detail. Thank you!
Scripting indeed takes the most time. Hearing it being appreciated makes me happy! :)
Awesome! This series is probably the best I've seen when it comes to understanding fluid simulations. I'm a chemical engineering graduate and I've always wanted to program a simulation as a hobbyist but I've never quite had enough of a solid grasp on the concepts to proceed with the implementation. I could've simply followed some random tutorial but I'd rather develop an understanding of the concepts being applied to able to code my first simulation. Can't wait for the next part!
Thank you for taking the time to write such a kind comment! I'm glad you like the general approach. Personally, I focus a lot on the conceptual side (maths, problem setting, changing perspective, asking the right questions, things like that ...) and it helps to see that others share these thoughts. I hope to see you in the next part!
Really respect the thought you put into now just the how but also the why to your methodology. Super excited for the rest of this series!
One of the most underrated channels on yt
This is shaping to be one of the most exquisite series on youtube. Outstanding work so far, truly.
Thank you soo much! I'm glad you like it. :)
This has to be one of the best educational series of videos I have ever seen! Right up there with 3Blue1Brown and others
Thank you for the appreciation! I feel honored. :)
I hope this channel gains more popularity. The videos are visually stimulating. And you learn a lot in the process
Hell yeah, more fluid dynamics
Just discovered this channel and spent 2 hours watching one of the more precise and lucid explanations on the internet.
As a graduate on these topics, I find your perspective is incredibly unique and worth sharing with everyone.
Keep up the great work! Waiting for your next upload!
This is making me have deep thoughts about the nature of time and determinism in the real universe.
The clarity of your explanations and the beauty of your animations are admirable.
Thank you so much for this!
As a side note, do you use manim for the animations?
Your videos has simulations as exactly how I think and they completely present my imagination which is so beautiful. Thanks a lot for this wonderful work. I liked and subscribed already. Wonderful Channel!
I'm very happy that it fits to your imagination! I hope you like the next parts too :)
I'm dazzled and speechless with your amazing video. I'm a fluids engineer working with statistical thermodynamics and your video made me recover my energy and motivation towards my daily routine. Damn!
Amazing job. Never stop the channel. Congratulations from Brazil.
Absolutely brilliant! As a chemical engineer I am delighted with this series. This is the most intuitive derivation of how macroscopic properties are derived from microscopic states.
It took some years of studying to develop this picture for myself. And here you derive it all in a visually intuitive way, in such a short period of time. Masterfully beautiful.
Your videos fill me with joy. I have a newfound appreciation for having studied this field.
10:17 when your microscopic simulation tells you your full legal name and address
You have only 13k subs and this video only has 22k wiews? im really expecting you to blow up very very soon, you have big potential
This is the top tier of UA-cam. Congratulations!
Such an interesting way of thinking, wanting less information to be able to simulate faster
This man DELETING TIME for better sims. This series is awesome!
Fantastic video again, I'm in love with this series. Thank you so much!
ive never heard of this cfd method and its absolutely genius
This is the best youtube series on youtube, making 3b1b look like a fool
incredible quality of teaching and animation
Outstanding video as usual, and the thumbnail this time is a significant improvement! I'm so glad to see that your videos are starting to get the attention they should. Also good to see you made a Patreon! I imagine over time you'll reach many people who will be very happy to support novel content like this :)
I'm happy to hear from you. :) Thank you again for addressing the thumbnail issue last time. I thought carefully about your comment when creating this one! :) And of course, thank you for your support!
Phenomenal video, exited for the next one already!
So exited for that series! Have fun creating it!
Extremely interesting! Very much looking forward to the next episode in those series!
One of the best simulation animations I've ever seen but the voice is good for sleeping. Sleep is a good thing tho
Absolute master class. This is exquisite content.
Geez your visuals are stunning! Loving this series! Thanks :D
Amazing videos! Can’t wait for the next ones
This is criminally underrated
Thank you for giving us hope for the next parts!
Waiting for the next video eagerly. Very very eagerly.
The video is super beautiful :)
I'm wondering about one thing though: When simulating fluids with those SPH-particles the SPH equations (in the zero viscosity case) are (although derived from Euler equations which are derived from the Boltzmann equation which is irreversible) time reversible and using a reversible integrator and fixed point arithmetic, we can actually see the particles forming up again if we reverse the velocities after coming from such a formed state, as done in Globally time-reversible fluid simulations with smoothed particle hydrodynamics - that we can do this, however, comes from knowing the exact particle positions and velocities - only from the 1- particle density, it is as in the colliding spheres case - just unlikely to find this behaviour. The Boltzmann entropy (based on the 1-particle phase space density) increases (statistically) while the Liouville entropy remains constant.
Edit: Actually forming a dam again and clumping up are very different phenomena. For a non / weakly compressible fluid to form a dam again is quite plausible, but the system will not clump up again working against pressure forces which normal particles - not our SPH particles - could do.
This channel is incredible.
The real laws of motion always include both *repulsion* and *attraction* which is why the normal curve of probability in entropy (probably) makes all possible microstates actually happen. That's why so many folks get confused by entropy, and think that it means that things only ever fall apart, and it's why these computer models are not especially accurate to real life, because entropy actually means that things fall apart and form new things, over and over. Entropy is the same as evolution, with that messy sexual reproduction via natural selection (things coming together) and random mutation (things coming apart).
Sorry but no. Entropy is not evolution. If you look at each state of a time crystal you can say it evolves into the next state but yet time crystals have 0 entropy, they are non-entropic processes. Entropy is often misunderstood because it is just that, a hard concept to understand. Look up Shannon entropy to get a better understanding. Essentially entropy is the limit of energy transfer of the microstates. Very, very few things in this universe have 1:1 energy transfer, there is usually a resistance. (Time crystals are the only example I can think of) Hence, entropy increases for most systems. Essentially entropy means things lose energy over time. Hence why you can't have perpetual motion.
I think what you mean is that like evolution, entropy is a quantity which exists ONLY for the system. If you take the particle out of the box there is no entropy. Which is true!
@@88Magician88 Yeah, it's unfortunate that so many kids were taught such a weird, old fashioned, backwards idea of what entropy (the laws of reality) is. Shannon entropy is what I'm talking about, which is evolution, as in the process of generating a family tree, where individual patterns combine and then divide, over and over again. As you say, it's the number of possible microstates in a system (macrostate).
Pascal's triangle models this nicely, or a Galton board, if you're curious about what it looks like.
What something appears as to an observer (anything that can change, and have multiple states) will vary based on what the observer is (how it's observing). So some things seem to not be changing (matter), and others seem to change rapidly (energy).
If you experience things "losing energy over time" that's because you, as an observer, are focusing on the matter-like elements of it. In reality, the universe never creates, nor loses, energy or matter. It's always the same amount. Just differently distributed. Like how a fractal keeps changing while staying the same. Reality is just one massive family tree of patterns of varying stability and change branching and recombining, to evolve all possible states of existence. Stephen Wolfram is starting to explore this if you want a different approach then my own.
@@thewiseturtle I agree! (kind of) Whenever I heard entropy being described by a prof I always had to sort of tune it out because they would say something like "this isn't really how it is but it's an analogy to help you understand", if it isn't how it is don't teach it to me, ya know?
I hesitate to say I fully agree because you use evolution, which is loosely defined. Time crystals evolve over time (yes they repeat, hence the crystal but they do evolve) but are non-entropic. So I can't say I agree that entropy IS evolution but I will say they are analogous. I do love the way you think about the universe. Not many people see it like you but I think your picture is a lot closer to reality than others! Well done!
Dont gravity and electromagnetism work thru the move along spacetime=attract, move opposite to=repel? Wonder if the notion of electrons being one-handed has anything to do with this asymmetry? I forget what I'm referring to, but something about them not being reflected in a mirror
@@oUncEblUnt420 I think you are talking about electron chirality. Electrons are left-handed (spin) and positrons are right-handed (spin). The Stern-Gerlach experiment was what they used to figure this out. Spin has nothing to do with gravity or E and M. Spin has to do with angular momentum. (If the electron was a macroscopic object then it would be spinning. It's not, therefore, spin is an objectively confusing name.)
The "notion of electrons being one-handed" has to do with this chirality. The claim is that there should be an even amount of left-handed electrons (electrons) and right-handed electrons (positrons). We don't see this though, we see mostly left-handed electrons. Thus you could say that chirality symmetry is broken by electrons. However, there is nothing that states the universe must have chirality symmetry. We assume it does because then we can say the universe has parity symmetry, which would be nice mathematically but there is no physics that demands this, thus things that break parity symmetry (probably) aren't paradoxical.
This is amazing! I've watched all the videos in the series so far and I'm excited for the future videos. I would love to build my own CFD.
Amazing! Looking forward to the following episodes!
I think its time to revisit my patreon account now :D
I would easily pay for a full fluid simulation course of this quality! Thanks for the effort to create this
Thank you very much for your support! It helps a lot! :)
I want more of that thank you for this video i love your voice is so relaxing
Your videos are amazing! Keep up the good work!
All your videos on a whole another level.. It's beyond awesome.. Great animations and great in depth explanation.. Great work through and through.. Loved it..
Great videos, great channnel. 🙌🏻 Thank you for your good work. 🙏🏻
Absolutely beautiful
Amazing video please keep making moar!!
This is the type of in-depth thinking I've been missing
I'd love if you could do a video solely about how you go about making these videos (research process, tools used, etc.). The pacing, colours, animations; all fantastic!
OH BOY! HERE WE GO!
Great channel! Subscribed!
Can't wait to see more! I absolutely adore this kind of stuff
The legend has returned!
simply beautiful, great work!
please make sure to discuss both space and time discretization methods in detail. We would love such a video
Loving this series!
Can not wait for the next episode!
Set up a particle simulation and witness energy un-dissipate? I have tried that too ;D
When I learned about Poincaré recurrence theorem as a student, I thought to myself, doesn't this destroy the 2nd law of thermodynamics? You sir, have a very satisfying answer.
awesome, mind blowing , marvellous kindly keep it up
Could you help me to understand what program/libraries you used to make this? I really like the visualizations you use and want to start making physics education videos
Thank you for another great masterpiece! One question: should we consider this as a part 3 of the Understanding Fluid Simulation series? I see your comments below the video and asking because: (1) The video name doesn't start with the "Understanding Fluid Simulation" prefix and (2) title picture is different.
Should we expect part 3 as a separate video or this is it?
Thank you very much! This part is meant to be a short follow-up, as it addresses things that I couldn't fit into part two. So it's kind of part 2.5, if you will. I plan to make the next part (3) more in line with the first two parts.
While watching this I got hit by the brick of solid nitrogen that improbably materialized above my head.
Most everything that happens in our animal level of the universe is pretty improbable. This is what we call free will. It might be deterministic, but it's also (likely) pure randomness. So everything that possibly can happen does happen, somewhere, somewhen. Of course, not everything is possible. But nitrogen glass does happen here and there. And so does your head. So while improbable, it's not impossible. Just like falling in love, or winning the lottery.
@@thewiseturtle That's not how probabilities work.
@@TheQxY What's the probability that that's not how probabilities work?
Also, which isn't how probabilities work. I said several different things.
Are there delicate fluid phenomena other than time-reversibility that are lost in this approach?
fantastic animations :)
Sorry if this is a stupid question (since I am not a physicist, just a curious man), but don’t you need to add gravity? In zero gravity, fluids tend to stay in their lower energetic form, which is a spherical shape. If your model doesn’t include gravity, will the particles eventually acquire a less energetic formation? Or maybe this doesn’t apply to single atoms, but only to molecules with more complex interaction such as hydrogen bonds? Anyway, congratulations to your series, it is truly amazing!! You should keep the channel active! Thank you for all the effort you put in it!
Yooooo this is amazing!!!!! I really appreciate how UA-cam has become a platform that shares knowledge that bewilders my understanding of the world. I've always been a keen learner but holy cow. This is beyond that, eye opening tbh.
Sir I want some advice.
I already am somewhat familiar with coding and programming. I don't know OOP but i do know the use of structures in MATLAB and fairly complex stuff in Wolfram Mathematica. I want to learn how to build my own simulator of fluids so I can play around see what I discover. Which language would you recommend me? What are it's advantages and what's the main framework I need to understand to make it become a reality? I'm curios about C# as Sebastian Lague seems to use it. Python I'm familiar but not proficient. I am willing to really take the dive into it. But I don't know where to start!
Amazing script!
Can't wait
Waiting for upcoming videos :)
Love your videos
Great animation quality.
Where is the next part?! I miss this series so much, even if it only has 3 videos
So good!
Great Explaination❤❤❤❤
LETS GOOOOO!
This was epic!!!
I like this video but when you say "reversing" it initially confuses me. Obviously, if you could magically pause a flow and reverse all the particles, that new flow cannot defy laws of thermodynamics. But that is not usually what reversal means. If you were to simply run back the clock and track the particles in a simulation (not the real world) they would go back to the initial conditions. Anyway, thank you for the very insightful video.
Agree with all the other comments here! What tools do you house for your simulations and animations?
when you are going from a lower-level model to a higher-level model, do you rigorously derive how their coefficients are connected so that they produce the same results, or do you approximate them until they sort of match? for example, in this case did you derive the macroscopic parameters such as viscosity from the microscopic parameters such as particle freerun length and radius or did you just tweak them until they sort of match?
Personally, I like the fact that there are different approaches that can be used depending on the main objective of the analysis. Your comment addresses this aspect perfectly! Since this series is about giving structure and looking conceptually at problems, knowing that there are different ways is a good placeholder for further discussion if needed.
Nice simulation
Reminds me of early 2-minutes paper video.
Great
Beautiful and instructive video but any encoder’s worst nightmare ^^
Me who understands none of the technical jargon:
OOOOOOHHHH Pretty glowiee thinggiiess and Soothing music + voice
Randomness, viscosity ,'nothing is entering or exiting in such a pure and sealed box.', .This was left out. Tending to an unrealistic odds of system development. Seemingly breaking law of entropy.
Disappointed by this TIME this video is published because i wish it was earlier that i can now find the future parts in the channel
hmmm, so destroying localized information restricts the arrow of time 🤔 that got me a philosophical epiphany on the nature of our own reality... if quantum fluctuations are truly random, they do destroy localized information, and that would explain the "only forward" nature of time. Either that or I already drank too much
or the statistical nature of quantum mechanics is already hiding an underlying deterministic concept that doesn't really have a forward time direction. Maybe "forward" or "backwards" has no meaning at all its just happens to be either case (not both) which doesn't really effect anything as particles behave the same either way. macroscopically it's just statistic so certain unlikely low entropy states like sudden perfectly straight towering atoms or every particle accumulating in the corner of a room or similar meaningful stuff just don't occur or at least extremely rare.
@@ecicce6749 great answer. i'll add my 2 cents here. evolution laws do not require a definitive arrow of time to work, since they are symmetrical.
I don't know if that's relevant, but I felt like I should say it.
Hello Braintruffle .
Can you please suggest what you consider as the best fluid mechanics books (videos ) for a mathematician and a hydraulic engineer , and I need also books on the subject of Numerical Analysis ... thanks in advance
Awsome. What rendering technique do you use? Blender + python?
I also want to know the same thing.
Which software/library do you use to generate these amazing demonstrations?
These blue light dots make my brain white hot
Wouldn't intermolecular forces introduce dissipation and prevent time direction effects? Or are you only interested in perfect gases, based on kinetic theory?
5:15 >never undissipates, even by simulating backwards in time
wait wait wait. your microscopic equations are completely time-symmetrical. are you saying that the microscopic equations don't "undissipate" or the macroscopic ones?
i guess later you clarified that it's the macroscopic laws are non-reversible because they destroy the information about microscopic state (that is reversible by itself). which makes sense. but you probably should have explicitly said that at 5:15.
I’m sure you’re familiar with the SINDy work they’re doing at UW. Is the average probability density and dissipation already built-in to SINDy by the nature of sparse optimization? I’m wondering if explicitly incorporating this probability average could be used algorithmically to reduce the cost of hunting for sparse matrix coefficients, or if it is just as costly?
Amazing tutorial. What do you use to animate so fluidly?
From which software you are making these simulations?
Yes, the averaging operation is indeed irreversible, as is obvious in retrospect.
The language you use is very heavy, dumbing it down just a tad bit would make it alot more digestible for more people, great visuals though!!
Maybe the dumb people should step it up instead of lowering the bar 😂
I actually saw nothing wrong with the first animation aside from energy not disipating. Think the reactions were 100% efficient therefore there was a sudden drop in entropy. This can actually occur in real life but far less and needs quantum tunneling, and if you want to avoid that in the above example you should either add a drag factor or some energy loss upon collision.