Home Creators Posts Import Register Favorites Logout

Content

I'd love to get your thoughts on this video (if you don't mind spoiling the launch of the final version).

I'm planning on releasing it tomorrow. I must warn you I'm still updating some animations, visuals, color, music, etc. but I think your input could really improve it. Sometimes I get too close to a video and then I can't see what's missing.

Thanks for your input!

Derek

Files

science behind the butterfly effect

This video is about science behind the butterfly effect

Comments

D N

the b roll of the portrait of Newton looks a little unpolished

Leonardo Aguiar

The lady in the pink dress behind the pendulum animation has so much detail it distracts from the animation and explanation.

Kristian O M

The chaos animations were a bit confusing. The voice over described the attractors as points but it seems they only showed up at the very end of the last animation. So it doesn't really exists, right (as explained in the fase-base diagram)?

Patrick W. Gilmore

You state that the lines will never cross, but at 12:30 (ish), you plot a crossed line. Also, a "circle" could be considered "crossed". I would not, but it might seem that way to some.

Patrick W. Gilmore

I know the point of the circle is that at any point, you know the next step - but you might not. You can go either direction.

john kraemer

Maybe it's because I watch so much 3blue1brown, and his multipart animations are always generated from their own underlying phase space so everything has to stay consistent, but it feels very weird when at 3:20 the position on the plot changes while the velocity is still somehow zero. Also, at 6:53 when you're saying 'totally different trajectories', the trajectories on screen are still more or less the same.

Joey Luther

Does my comment change the outcome of this video? 🤔 Regardless of whether I offer advice or not - I'm still stealing precious time to read and process this message, potentially delaying or disrupting production even by a mere few seconds. Because of this message, the earth and moon could potentially collide.

Mel aka janMelantu

The final butterfly animation seems too zoomed-in, it's hard to understand until the very end

Ryan Tapping

This is a really great video! I would agree with Jon Kraemer that at 6:53, you might want to re-dub your speech to say "totally different trajectories" when the paths of the three objects are more divergent, as to not confuse the viewer -- I personally didn't really realize what animation I was watching until they diverged more (that there even were three objects to begin with, and that the demo was to show the divergent paths). Also at 9:38 you say "fog" and show an image of fog - which is great! But I think it would make your point stronger if you could show fog getting thicker over a timeline (making your point) rather than random image of San Francisco's fog. But honestly, great job. I've learned a bit about these things in my own physics classes and I think this does an excellent job of conveying the concepts without throwing in much jargon.

Mel Kimble

Once again, your production quality is excellent. The subject matter is well chosen and well presented.

Dragi Raos

I have no suggestions. It is very clear and feels way shorter than it is. Attractors probably deserve an additional video. Also, a few pointers to additional reading in the video description would come handy (Gleick's book is an obvious candidate, but you have surely encountered others, perhaps better, while researching this.)

Andreas Fink

Very good, very clear. Perhaps more words on the solar system. Is it`s instability of importance to us? We've been around far longer than the few million years you mention in which a significant effect can be expected.

Gregor Shapiro

The 'crossing' at 6:35 (ish) is on your 2D screen. My understanding is that in 3D they do not cross.

Mani Rajagopal

This is a great video!. I loved how you started the video explaining Laplace daemon that the future is predictable provided we know the initial condition accurately (emphasis on the accuracy). I am student of meteorology and since I knew the concept I enjoyed the explanation greatly. Many of my friends did not know the difference between chaos and random. It is covered in the video, but it could be emphasized more. Ending the video how the flap of a butterfly could cause a tornado could be explained bit more. If a pressure sensor is not sensitive to capture the pressure perturbation from the flap of a butterfly, the initial conditions (pressure values) given to the model has error, this would grow in time and would decide if there will be a tornado or not.

Mik Scheper

Loved it—it's amongst your best. A couple of thoughts, though: • When you said you were going to show a 3D graph, I kinda imagined it would look like the pendulum graphs you'd already shown, with axes and a white background. So it took a while for me to see the 3D graph for what it was. I never really understood whether the colour of a point represented time, or one of the values, or what. I suppose these details doesn't really matter, but maybe a few more words of explanation would've saved me a bit of head-scratching. • You used the word 'attractor' a few times, and I can only guess what that really means. Maybe you explained it while I was distracted by trying to work out the details of the 3D graph, or maybe you didn't explain it, and it's a term I've forgotten from the physics or stats I studied at uni, or they're called something else in Australia. Not sure, but I'm flagging it, in case someone else had the same experience and it's something you could make clearer. But overall, it's one of those 'wow, that's really cool!' sorts of vids. Well done.

Sese Schneider

What is a "fai-space"? I think you use this word to often

Mik Scheper

I agree that 'totally different' might not quite be the right way if putting it. What actually happens is the values get more and more significantly different, and eventually, dramatically different. I guess 'totally different', technically, would mean opposing momenta… which possibly could theoretically never happen, otherwise the system would continuously ravel and unravel and actually be a loop? 🤔

Mik Scheper

Odd, I can't even remember a lady. But you're right, there was a lot on the screen at that point, and maybe that made it take longer for me to wrap my head around what it all was. But there was an elegance to it that I liked… maybe that's not important, though. 🤷🏼‍♂️

Mik Scheper

Yeah, I kept thinking 'hold still!', but stills in vids seem to be discouraged. Maybe that's the cause of XKCD 1264… https://xkcd.com/1264/

Veritasium

for sure! and will link to them in the description

Veritasium

this is a good point. So maybe things have changed? But it's tough to tell given the timescale of our observing

Veritasium

thanks! Yeah I'll fix up the timing on that. I may add some more stuff on the fog image

Veritasium

yes, this is a little tricky... I'm working with an animator and that is the framing he chose but I agree it doesn't really show the 'butterfly'

Veritasium

haha - I filmed that (handheld - no surprise) at the Royal Society in London. I'll add stabilization

Veritasium

Phase space... yeah maybe I should add this on screen

Veritasium

thanks! Yeah the attractor is the object the trajectory seems to be "attracted" to... so in the case of a damped pendulum it's attracted to the origin. In the case of the undamped pendulum the attractor is a loop, and in the case of Lorenz, it's a butterfly-looking thing

Madison Bombardier

At 9:20 when you say that chaotic systems become unpredictable at modest time scales and then follow by saying we can't predict the future very well at all... I think this just needs a caveat, because it really depends on how many active components there are in the system (weather has a lot), how precise our initial conditions are, and the scope of applicability (as in maybe we only care about projections of ten days but our precision is good enough to reasonably project 50 days before the behaviour becomes sufficiently chaotic to the detriment of an accurate projection). And also just right at the end when you talk about how we can gain insight into the nature of a system by running an ensemble of projections even if any one projection is unpredictable... maybe this is a topic for another video, but I feel like the bounds of a system are important, because I would hate for people to watch this video and think that "totally unpredictable" means "literally any state in the universe." Even if the specific state is unpredictable, we can still be sure that the state exists within certain bounds that are very likely known.

Veritasium

right, this is the solar system as a whole though, and even though the Earth may be stable, other planets and moons may not have been a billion years ago...

Madison Bombardier

"Phase-space." He defines it at 2:30-2:50. It's a useful abstract way to talk about the state of an object in a system

Joel Menken

I only see two editorial changes. At 3:40 cut away from the pendulum graphic before it starts moving again. At 4:01 add a transition to the cut. A blur, flash, or morph translation to clean that up.

Florian Linscheid

At 6:30 I find it hard to realise that it's a 3D plot while the camera is following the point when there is no else reference point to keep track of the movements. That's why you need to watch it again to notice that at 6:34 the lines are not actually crossing. I know you mention that Lorentz used 3D space, but the animation gives no (obvious) hint to that. Otherwise I feel the main messages about chaotic systems are well presented!

Uri Rothschild

maybe explain a bit more what phase space is as it's quite an important part of the video.

Uri Rothschild

also maybe tell why the lines/path can't cross and/or what would happen (or what it would mean for the possibilty to predict the system). great video btw. loved it a lot. and it made me a bit philosophical on how small things/events/decisions etc can have huge impacts.

abcdefghij

A great video! 🙂 I would avoid room zoom out at the beginning part of the video. I also get the idea about looking into foggy future, but that bridge didn't seem to be that relevant. Thanks for the video!

Kimberly Green

I believe chaos is the mathematical manifestation of hope.

Bahman Engheta

Perhaps say a word or two about the so-called logistic map and how the choice for the single parameter of that simple recursive relation determines its evolution over time and its number of attractors, and show the famous bifurcation diagram.

Natalie Hansen

At 10:11 it would be nice to leave the "reveal" of the butterfly a bit longer, that way the viewer can start to recognize the image the shape and then you can give them the clue to see it. Overall, it's a good video, during parts I felt a bit "fatigued" trying to keep the concepts all in check, for this. I think you should lean into the "stories" aspect of things like this just a tiny bit more to create a flowing narrative to keep watchers fully engaged. I fully don't expect you to rewrite the script and the overall video is still very good. I appreciate the video's conversation feel, and the willingness to talk to the audience without talking down to them. I feel like this video should not pack more content into it. Much more and it will lose its broad appeal. As part of the science fan I like big expansive ideas, but this level of detail is appreciated especially since it stays more in its “roots” of the butterfly effect. It has some winning ideas in it: Catchy Title, and a promise of a big payoff (without overselling). Although unlike other big hits I think this will lose casual viewers a bit too quickly due to how “technical” the video feels. I don’t know how you should or if you should address that in future videos. Thanks as always for making awesome videos, and teaching me something new!

Veritasium

Really appreciate your thoughtful comments. I have tried to space some things out a little and add in some helpful text. I agree I may lose the casual viewer, not sure how to rectify that. I love those butterfly plots but I do feel like they are incredibly abstract. I'm already planning to take another crack at some of these ideas...

Veritasium

I had logistic map planned for this video but cut it due to time constraints. I am planning on making a follow-up though. Good ideas!

Bahman Engheta

Makes sense. Best to keep it manageable, especially for anyone seeing this the first time.

EenKerel

Maybe it makes sense to discuss a bit as to why this happens? systems with "memory" or internal state?

Matt Dunlap

I loved this video! ... but then again I've always had a soft spot for chaos theory. The one thought that came up was that the video didn't do much to convey the proportion of systems that display chaotic tendencies. Have we discovered one real world chaotic system or a billion? That is, outside of weather, should we think of this as normal? I know the answer, but others may not, and the fact that so much of the world is chaotic provides a strange sort of meaning to life. A defensible escape from the existential dread of determinism. Keep up the great work! Chaos is more important than most people realize!

Dragi Raos

The attractor was first mentioned when showing the phase space graph for pendulum. It seemed clear to me, but then, I am familiar with these concepts.

Jim Cruwys

I think you need a summary. You talk a lot about the butterfly effect, chaos, and the uncertainty principle but you never answer the question of "how well can we predict the future". I know it's obvious from the examples but I think a summary would be good.

LodeRock

Would the timescale mentioned at 8:49 be the Lyapunov time of the solar system? and would it be the same for all things in the solar system? Great video, really liked this concept.

Анастасия Ласкина

I like Your pronunciation! English is the third language for me, but I can understand You easily! 10⭐ for this important point for video bloggers!

Jesse Thompson

I feel like he explained that pretty thoroughly. Each point in phase space represents everything about the system that would define where it goes next. So if you ever crossed the same point, you would be doomed to flow through the precise same motion all over again. (This kills the forward evolution of the pattern. 😉)

Uri Rothschild

mostly agree. i wasn't sure if one crossing would automaticall mean that the system/path goes into a loop. or if there are scenarios with a limited number of repetitions/crossings (e.g. 1)

Marco Trevisan

Three comments: 1) You mention the word "attractor" straight away, but unfortunately this mathematical concept has a name that is very similar to the word "attraction". I remember when I was a teenager that this confused me a lot, because I used to think it was something related to magnetic fields, or gravity, until I found the definition and realised it was a mathematical thing. For this reason, I think it deserves a sentence or two saying that, «despite its name (justified by the historical development of the concept), it is not about electrodynamics or gravity, it is just a mathematical construct (describing a type of trajectories in the phase space which don't tend to change much as a whole under slight perturbations)» (or something shorter... perhaps just removing what I have in brackets). 2) As another user said, the bridge to show the foggy future and the foggy past would be ok in another context, but due the quality of the rest of the video (you even used Fragonard's "The Swing" as background for the pendulum!), one would expect something nicer to illustrate this. I personally recommend, as an example, using "London im Nebel" by Lesser Ury, which should be available because the author died in 1931. https://commons.wikimedia.org/wiki/File:Lesser_Ury_London_im_Nebel_1926.jpg 3) Are you sure you want your gmail account publicly available in a video? You show it several times in the lastpass section. Unless it is a mock address just for this (hope it is), I wouldn't personally display it here, because you would need to edit the video and post it again if you regret it later.

Marco Trevisan

By the way, a very nice video, I just point out these things because you asked for suggestions, but it is very good already!

Tomáš Kosec

As many people said: summary at the end of the video would be nice :) But as always, very good video!

Natalie Hansen

That may be enough. It could be a slight tweak to pacing can give the ideas enough time to breath I will watch the final video and add any additional feedback. I do really like this format of video you are working towards.

veritanuda

I like the graphics. They were clear and concise and added to the overall explanation. I have a certain fascination with Chaos Theory myself only in so much as it is a philosophical question of how something on a macroscopic scale can look so ordered while underneath on a microscopic scale it is so chaotic. The two diametrically opposed states seem impossible yet it also seems contrived somehow. Really it is the fuzzy edges of science that really stretch the imagination.

Robert Assing

Really nice video and good work on the graphics! I don't know if it's too late now but the is still no Shadeball here in Germany :(