This video explains why I had a blue splodge on my face in the last video! The sponsor is 80,000 hours. Head to 80000hours.org/steve to start planning a career that is meaningful, fulfilling, and helps solve one of the world’s most pressing problems.
If you put small holes in the clear acrylic at the "top" of dead end pockets that are sealed it might cause the water to go more places. presumably more water would go down then out of the holes because gravity should still be stronger.
You know what I think can solve the puzzle flowing through the main path right away?? Smoke! Pushing smoke into the puzzle, it will flow through the path of least resistance from the start
Yes, that is almost the same as at 6:36 where the pre-filled maze is fed with different colored water. If the air pockets were also pre-filled with water, then it would be exactly the same.
Legitimately almost didn't click because I didn't want to sit through 8 minutes of "hype" for an underwhelming reveal at the end of the video. This was not that, and I was pleasantly surprised
Regarding surface tension, try solving the smaller maze with 'wetter' water. Reduce the surface tension of water by dissolving some alcohol in it. :) Might be interesting with different fluids as well.
@@ericmoss6110 omg, superfluid helium would be an amazing test as it also can crawl *up* - it'd presumably solve the maze pretty much exactly by the keep-hand-on-wall method. That said, if water was already a difficult thing to seal against, helium manages to be nigh impossible lol
AT LEAST two (separate) parts. Not two parts. Many mazes have many separate disconnected walls. (There's no requirement for only having one possible solution in a maze.)
There are couple of important things this experiment shows, especially when dye is added. It shows what happens when there are unused lines in water delivery piping (like for example building hydrants sharing same line with tap water). It can become breeding ground for bacterias, eg. Legionella, despite water moving, and that's why it's important to use backflow prevention valves.
'water age' is indeed a critical factor in delivery quality water. This is why municipalities and other agencies controlling water supply model and test for water age as part of the design of new expansions to an existing system. Areas of known longer ages, are regularly flushed. Additionally, modifications to the distribution system can help eliminate these regions. one tool in this fight is preventing a dendritic (tree-root) branching of the system. By carefully utilizing loops and and multiple paths, it is possible to significantly reduce water age.
This is also why I always run the tap until the water is genuinely cold, to ensure it’s not been sitting around for who knows how long in the building or the street. Sometimes it happens quickly when there must be high turnover but sometimes it takes a few minutes.
@@kaitlyn__L Temperature isn't a bad guide, although the 'new' water from the street (which is buried, so likely cooler) will have to cool your warmer piping down, so once you feel the temp start to drop, you've likely got 'fresher' street water at the tap
That's why I don't trust the campaigns that say that water treatment is so good that you can drink straight from the tap. I do believe that the water from the treatment plant is great, but there's no way I trust those pipes between them and my flat, not to mention those in the building itself.
@@karol30660 I'm sorry that you distrust your water distribution system. I would suggest you sample your own water from your own tap and have it tested (at your own cost) against your national drinking water standards at a private, independent lab for your own peace of mind. If, in the unlikely event that your water falls outside of the national standards, your local distribution maintaining agency will be VERY interested in your results.
I had never thought of a maze being in two halves - and yet it's so obvious when you think about it! Steve is like the Johnny Ball of the digital age 🙂
@@Khaim.m Yea! I got that once I got that it was in two halves - just never thought of either of those things until I watched this! I love the way his videos make you look at everyday things in a new way 🙂
For the smaller maze, you can just fill a flat metal/ceramic tray with just barely enough solvent to cover the bottom. Then put the maze into it with the side you want to solvent weld. Leave it in there for like 10-30 seconds. Then take it out, and set it onto the panel you want to weld it to. Because the maze is so finicky, I'd still do the first side with the syringe. This process works for all kinds of parts and actually tends to give a cleaner, more even, weld. However, since you're using laser-cut pieces you'll have plenty of micro-cracks in them after getting them into contact with solvent (unless you tempered them afterwards), they are probably just hard to see in the black acrylic. For large pieces this works too but of course you'd need a pretty large tray (or one custom-made for the job) and a lot of solvent. You also don't need to actually submerge the pieces, just lowering them until the solvent pulls itself onto the welding surface is actually better (because it only coats the surface and because the solvent doesn't get squeezed out) but a bit finicky, you might be able to put some uncoated steel wires into it as an offset. Source: Sometimes do this at work, but for most pieces the syringe method works just fine.
Regarding using the "hand on one wall" method to go through a multipath maze, as long as you pick a wall right at the entrance then you should still always find your way out. Any extra paths would create a third piece to the maze, but by necessity it would have to be in between the two outer parts which must be the two parts bisected by the entrance and exit.
I wrote a longer comment about this but also if there are 2 entrances and 2 exits, you must find a path from an entrance to an exit, and the paths between each entrance and the exits connect, there is a chance the wall you follow will lead you on a path between the two entrances instead of to an exit, but that's the only thing I can think of that would defeat the algorithm, barring dropping someone in the middle of the maze instead of starting them at an entrance... or having no paths from the entrance to an exit, obv.
@@JuniperHatesTwitterlikeHandles You can theoretically have traps and one way doors in a real maze, so that you can go through a particular section one way, but not leave through the same area - meaning that you'll get trapped in the section beyond the one way door. Or you place the exit in the "middle" part by having a tunnel up or down from one of the rooms that's only accessible through the third part of the maze.
This only works if the starting point is along the outer wall of the maze (which is _generally_ the case, but not necessarily). If you are starting from a random location within the middle of the maze, then you have no way of knowing whether the wall you touch would be connected to the exit or not.
It is also possible for the exit to the maze to be a staircase or ladder or something which is placed within the maze. That would mean that you could follow the outer wall in a complete loop and never find the exit.
@@icravedeath.1200 I'm pretty sure UE does offer water simulation. Blender has had physics simulations for a long while now. Don't know about accuracy in a scientific level, but they are more than amazing for what ever I will ever make :D
The "keep your right hand on a wall" always works, IF you start doing so from the entrance! A maze may have two or more paths to the exit and therefore has more than two wall structures, but it will always have just two outside ones.
Actually it doesn’t work at all if there are any floating walls in the maze, as in walls that don’t connect to the outside. Left or right wall doesn’t matter. Sometimes the floating wall is obvious and sometimes it’s not. Oh, and it only works if you can touch the outside wall directly from the entrance.
@@thehunters3402 At the entrance there is always an outside wall! It may be part of a 'floating' wall, yes, but then there´s more than one entrance. In that case, switch the same hand to the other wall of the newfound entrance.
With your second complicated, big maze, I don't think it is surface tension but an air lock, the path repeatedly goes up then down so if there is a bubble in each up bit, it requires a minimum head of water to overcome the sum of the uphill parts.
5:40 True, if you only start using the 'hand on wall'-tactic as a backup plan, after you've already been in the maze for a while a multi-solution one can trap you. Doing so from the start will work (for any 2 dimensional maze, where 'out the other side' is the goal. Get to the centerpiece artwork (or staircase to the exit-tunnel), are made out of 1 piece and may loop you back to the entrance )
If you're able to mark the walls in any way at all, if you loop back to the same point on the wall, you are following a center wall that doesn't connect to the entrance or exit. Switch to the wall opposite and follow that one. If necessary, just keep mapping these interior walls until you get to a wall that leads to the entrance or exit. I follow walls all the time in games to fully explore areas and find places I've missed. If you're in a maze in real life and you fear you don't have all the time in the world to walk around--definitely find a way to mark your exploration or break yourself out of there.
If you wanted to simulate the simulation you could put a small vent hole at the top of each isolated section of the geometry and then just cover/plug the hole once water starts leaking out of the venting hole. That should be easily done in the larger versions and you just need something like flex tape to cover the vents. it won't look as pretty but the physics should work.
Also, by treating every maze surface by a hydrophobic repellent, you could reduce the surface tension friction. Would be fun to see the results and how they compare to the vacuum simulation 🙂
I love by the way how you included the animation clip at the beginning but still motivated us to watch the original which supports its creator. Just love it
Also that he wasn't 'shaming' the animation creator (for lack of a better word) for not having realistic graphics but instead said things along the lines of 'he animated something I can't recreate'.
5:28 Even in a multiply-connected maze (one with multiple successful paths), you can be sure that the walls of the ENTRYWAY will be the walls of the exit, as long as the entry and exit are on the outside border of the maze. I didn't realize that until you showed your "coloring the walls" at 2:50 then mentioned that two paths would create a middle part. That middle part won't be the left or right wall of the entrance, but would necessarily be an "island" in the middle. The interesting thing is that you can't just wander around and decide later to use this algorithm: you need to commit to it at the very entrance to the maze. And, it only works if the exit is also on an external wall, not say a stairs in the middle of the level, or say a goal or treasure of some kind not adjacent to an exterior wall.
Though generally not a problem, if you're starting from the middle of a maze (e.g. you're blindfolded and carried to the starting point, or you forgot to use the strategy until later), then there'd be no guarantee.
It could also be the case that the exit/goal is somewhere in the middle of the maze, and there's only one entrance/exit, then it could be made of a single piece and thus following a wall will take you around the entire maze (including both entrance and exit)
This shows the difference between physical and digital simulations. In digital there is always something overlooked like the ire pressure in this situation which is why its always important to test the digital one on a small scale first.
Yup. It's interesting that lots of times when I'm reading good info on using a simulation or simulator of/for something it's recommended to do the real thing along with a simulator. I have peripherals for simulator games but should probably get rid of them because I don't access to the real world counterpart to get the most out of the simulation.
The point of a making a simulation in science is to put in all the variables that you think effect the outcome and their strengths and interactions and so on and then you compare it with real life observations. And then you expect to find that your simulation is inaccurate at some point and you try to find out what caused that inaccuracy and by finding out you learn something new... Simulations are always inaccurate, they are supposed to be, as soon as a simulation is perfectly accurate your field of study is solved, but, so far, there have always been more questions left to answer in any field of study and thus more inaccuracies left to iron out... At in the natural sciences, biology, chemistry, physics, etc. I am not sure, since I am not a mathematician, but I think there are parts of mathematics that are solved...
@@LutraLovegood I mean, you never run a simulation once and if it gives the result you want call it done. You always run it hundreds of times so you can do statistical analyses on the results... When I did my research the standard was 1000 iterations of a single set of input parameters and then analyse the results, which took about 3 hours on my fairly good gaming computer, but this was 10 years ago, my current computer would do it A LOT faster... The speed depends entirely on the nature of the simulation, some are really fast, others require hours and hours on university owned super computers.
The blue water is basically doing the same thing that the air did in the smaller maze. It's blocking the "wrong" paths, and instead of surface tension, there's just the boundary between two liquids of the same density - they're not going to mix very much at all in such a small time frame, so it's basically just a question of whether the blue water has anywhere to go when the red water pushes against it.
This video actually made me realize that if you just try to visualize how the maze is in two pieces it becomes easier to solve visually. Pretty cool! I also think it works how a siphon works.
It’s also why if you keep going right, or keep going left, you will always solve the maze. Unless, of course, the maze is broken up into more than 2 parts.
Which, by the way, knowing such information is good for you in emergency situations. If you are ever in a building under evacuation and it is pitch black inside, put your hand up against the wall and follow it. You will eventually reach a door and/or window.
Love this! Mazes were "my thing" at school. We had exercise books with squared paper in them for maths, and I used to draw mazes all the time, so I became really familiar with a lot of stuff about them. The "stick to one wall" rule is a good window into some of them, as it makes you think about where the entrance and exit might be (could be in the middle), whether it still works with multiple entrances, whether there's a way to use it to escape a maze if you start in a random spot (there are algorithms that can eliminate dead ends as you go, but they require some sort of memory or external modification)... A third dimension also stirs things up. The maze at Longleat in England (which was the largest in Europe in the 1980s, not sure whether it still is) has six bridges that go over the hedge paths. This allows you to see some of the connectivity from above, which people imagine might help - but the presence of the bridges also makes the topology more sneaky and the maze harder. I LOVE MAZES! Cool to see your construction skills applied to them like this. Excellent video.
@@nikelsad I made mazes with server plugins a couple years back. There are also "picture to schematic converters". So combining that with a normal image maze generator should also be possible. Particular mods or names i don't have, though. I only generate mazes for CNC laser action.
Dude, that sounds awesome!! I think one day I'll make this a birthday trip for my ex-girlfriend. She also loves everything logic-related. This christmas I gave her the hardest version of the Perplexus 3D labyrinth, she fucking loved it and tried cracking it for a over a whole day 😊 Have you heard of those? They are more about the motor skill to bring the ball through the maze. The hardest one is actually pretty hard! Thanks for the unintended present idea 😉
I came here from the short to ask what would happen if you put dye in after it was solved, very glad to see that it worked exactly as I expected! Thank you for doing more than just answering the face question and thinking of more interesting things that can happen in the situation!
Hey Steve, I do quite a bit of solved welding acrylic. For the second half you can do the dip soak method, where you 'soak' the acrylic in a thin puddle of solvent (such as in a custom HDPE or PTFE tray) and then clamp it to the 2nd clear acrylic part. Great video!
@@SteveMould Yes, you would want to soak the acrylic in the solvent to soften it, making it tacky so that when you make contact and clamp with the other piece of acrylic the two will bond in the same way that applying the solvent via capillary action does. Soak time will depend on a couple of different factors such as if the acrylic is cast vs extruded, if it has any colorants or co-polymers in it, and which solvent weld you are using (I am assuming you are probably using weld-on 3 or 4 or some other dichloromethane base solvent), and what clamping pressure you are using. You'll need to play around and find what works for you, but I would start with a 15 second soak and go from there. Since you are using lasercut acrylic, you'll need to be extra mindful of the acrylic crazing when you do this. You may need to anneal your parts before soaking, but I hope you can get away with not annealing since it can deform and change the dimensions of the part. The stresses in the crystalline lattice near the cut edge (the cause of crazing) are built up differently between laser cutters depending on the ambient temperatures, speed of the air assist, etc. If you are able to stop the cut edge from cooling as rapidly that will help with the crazing. The easiest way to do this is to lower the airflow of the air assist.
I actually used this once in a dnd campaign. Our DM put us in a huge mirror maze room. Using a combination of a few magic items, I made a torrential downpour of water and we waited a minute before following the flow
How did you keep it from back-flowing out of the maze? I mean this method only really works because of the verticality of the maze making it so the water cannot simply flow back out of the entrance. You would likely need way more than a minute in order for the water to rise to the point where you could detect a flow is a big room. My solution to DM's throwing mazes at casters is always just fucking blow through it or use divination magic to scout the correct path, ain't nobody got time for that.
This is amazing. You could hypothetically have a huge containment of water with multiple maze plugs of different tensions to simulate different distributions instead of larger or smaller ducts
I absolutely LOVE that this had unexpected turns, like just completely ignoring whole sections because air pressure kept them out Also interesting that near the end for a while, the air was bubbling up from extraneous sections and pushing some water towards the exit (again because of air pressure!)
This is a great analogy for another phenomenon. I remember when people were very surprised when slime mold could solve mazes. Were they really clever enough to find that out? No, they simply did the same thing as the water. They went everywhere, but when they reached a dead end, the slime mold that followed that path would simply die, as it was not part of the shortest path.
I really think you should look into getting the big versions of the mazes into a gallery. They are really beautiful art pieces against that white background. The blue liquid is really radiant
The water in the tank came to a halt like a car going uphill with its engine’s torque just short of power. The water in the maze needed more weight pushing down so it could push up all those nooks and crannies and fight against gravity and surface tension alike. Truly fascinating. :)
@SteveMould What a great video! It was cool to see how fluids actually travels through a maze with real world physics. Also What a great sponsor. 80,000 hours was something I didn't know about till now and I'm going to look at that organization ASAP.
Regarding the 2 paths leasing to 3 walls you can 'hug'. It still works holding the hand on a wall, if you do so when you enter the maze. The 3rd middle part that loops wont be one of the walls at the entrance. So unless walls are shifting its a pretty safe approach
Definitely one of these concepts that I never would've thought about on my own but are really interesting if someone brings them up. If you wouldn't mind the mess, you could maybe drill small holes near the top of the "air chambers", so the water could push the air out of the maze and get closer to bergman's simulation?
You've scaled up one of the dimensions of the system without scaling any of the other relevant variables (viscocity, density, simulation time). Neat to see how the system changes from that one input. I'd be interested to see this run by scaling all of the relevant variables to keep the system's relevant dimensionless numbers the same (reynolds number, etc). Would make for a great fluid dynamics example. Smarter every day or veritasium did a video on a boat research that talks about this a little EDIT: It's veritasium's video at a naval wave generator
Science usually makes me want my head to implode but you make is sound so fun! Your tone of voice and accent and way of explaining is just perfect, love this video, you got a new subscriber!
I had a feeling it would be surface tension being too strong for the water to overcome it. You present your paradoxes and caviats in such a good manner its more like an assist than me just knowing it
dude this was pretty dang good. you dont know how hard it is trying to explain how water works to people. and this video is gonna help them understand with a visual.
6:00 not only just surface tension but as you said earlier for the small mazes, air pressure. The first lip looks impossible if no air prevents the water from moving. All the little pockets of air that you see have formed along the correct path (unlike the incorrect path in the small model, although there is still some of that) to help hold up that massive head of water.
You're right, I would even say that it's pretty much only about pressure, surface tension has very little impact here. Pressure difference is nicely visible in the little siphon in the top left corner - free water surface is much lower on the right side than on the left which means that air pressure on the right is higher. And you can see why - because the air on the right is sealed off while that on the left can still escape (that happens in 5:30). You can also nicely see how the air pressure grows in the whole sealed right part of the maze. In about 5:15 the right part is sealed off and then water goes further up to top left corner of the maze (reached at about 5:27) on the left side of that seal, and you can see how the free water surface on the right side also moves up but much less, just enough to reduce volume and increase pressure of the air trapped there so that it equalizes growing hydrostatic pressure of water column on the left. Also, as the water stops flowing, the difference in height between tank above maze and surface on the other side of first siphon and that of to two sides of that column of water dividing right and left side are visibly similar (not exactly the same, because there are other siphons with uneven pressure on both sides).
correction, at 6:10 it isn't surface tension stopping the water, it is the air trapped in the system separating the water and thus the column of water is shorter on the way down than up, so the pressure isnt enough to go all the way up again
It's not the air trapped in the system that's stopping it. When there is enough pressure at the top, the air trapped in the system just gets pushed out.
@@unfa00 Adding soap to the water would be an easy way to test the hypothesis. If the hypothesis is correct, the water with soap will stop flowing at a lower level than the water without soap.
@@FirstLast-gw5mg yes if the pressure is higher the water flows, but it doesnt push out the air. Otherwise it wouldnt stop flowing once the pressure drops enough
Breadth-first search versus depth-first search versus ant-colony-optimisation for maze solving it'd be interesting to explore which algorithm solves what types of maze first and come up with metrics to describe certain maze classes like length of passages versus number of junctions etc :D
I know there is no way he sees this comment since it’s an old video but I think if you were to lay it flat record from top down and have the water enter through a sealed port. The results would be much different as the air would have more room to move.
@@kebien6020 air pressure is also gravity related though. Like what makes water move and air move? Maybe I'm being too pedantic, don't take it too serious.
@@mcmarkmarkson7115 only atmospheric air pressure comes from gravity. The one involved into not letting water in on some sections of the maze despite gravity pushing the water in that direction is not atmospheric air pressure.
Can I suggest that you hold the maze horizontally, and make the maze with the base and the walls (not the top). Fill the maze completely with clear water, and allow continuous flow of clear water into the entrance and out the exit. Once it's settled down, use some coloured dye to reveal the path of the flowing water.
Lay the maze on its side and take the top sheet of acrylic off. You'll have to be careful with the flowrate or increase the depth of the walls to stop it flowing over the top but you wouldn't have the issue of airlocks on the dead ends. Would become a pressure driven flow instead of a gravity driven flow. Id think it would still look different to the simulation but would imagine it would fill alot more of the odd routes up.
I was thinking put small holes over one side of the acrylic (back side so it doesn't spoil how it looks too much) and then make sure the amount of water going in is greater than the amount that can leak out of the holes. Again don't think it would quite match the simulation but think it would be close.
@@insertnamehere4775 Yes/No, having a feed at one end and it being empty at the other it will eventually flow through. with a very low amount of head. However, it might not look like the simulation exactly, but interested to see how it compares with what is shown in the video.
As a physicist, I must say that it looks oddly satisfying (and accurate). This is the first time I see this channel btw. Immediate subscription activated 😌
Your casts are uniquely appealing... information presented in a manner we can participate in... solve/understand what we are experiencing 😁 Much gratitude for your energy ❣️
You could try substituting the water with some type sort of oil used in vacuum pumps to test the maze under a vacuum. The oil shouldn't be too hard to get your hands on and they all have very low vapour pressures
4:55 - you dont't need vacuum, allowing the air to escape is sufficient. Thus making small holes at horiontal lines of the maze would sufficient. That of course creates a problem that water would leak through as well, but water has much larger buoyancy than air, so having a small plug at the hole, that would seal it once water reaches there might do the trick (though I am aware it is not that simple to actually build it that way, but you know, in principle it should/would work :D )
How about adding a fine mesh in front of vent holes. At the top there should be relatively low pressure, so a sufficiently fine mesh should prevent water from leaking out due to surface tension. However air could still pass through that mesh.
This is a visualization of Parallel Breadth-First Search, which is quite interesting. With multiple molecules in the water (a vast number), it can search for the shortest path simultaneously. The BFS search also has the advantage of finding the optimal solution. As mentioned in the video, the air pressure prevents the water from going the wrong way, making it very different from the maze-solving video. However, that simulation can also be like the real world by adding some functions. Typically, the maze-solving agents we see in videos can only sense if their nearest cell is open or closed. But in this situation, thanks to the air pressure, it can sense much further. Have no idea how to implement the sensing feature, it can be done somehow. For anyone wondering about implementing this algorithm...
I never thought about being able to "cheat" my way through a digital maze by coloring it 😂 Very logical if you think about it but it never came to my mind, that every (solveable) maze has to be made of at least two pieces!
2D mazes. A line can run through a 3D maze without separating it. Ex: a torus, ex: a donut, or a sphere with a kind through it. You would need at least a 2D object to separate it
Your tests are great to be used by cave divers, people going deep underground exploring cave systems. So that in case of storms outside and flooding in the caves, they ca use as a model to predict safe places for setting up camps. I've seen a story about such a situation of a deep dive and a storm that lasted several days outside leading to huge flooding in the cave.
Based on this, I think it would be an interesting although potentially simple challenge to purposefully design mazes to defeat this. Firstly, design a solve path that requires a large amount of distance to be covered vertically against gravity. Second, design dead ends that cover a large amount of distance vertically with gravity and have a large amount of volume. Lastly, make the channels small. Water will fill the dead ends first and you will need a large amount of water to do so. You will then need a large amount of head pressure to even get water through the solve path. If done properly, you'd have a maze that basically completely fills before being solved and requires either a source of high pressure water or an exceedingly large reservoir. Taken to the extreme, you could have a maze where the total resistance to flow is so high, water can't make it through.
I love how you straight-up admitted "the reason the small, complex maze doesn't behave properly is just because I fricked something up". No complicated terminology to dance around the problem; you laid it right out that you couldn't get it watertight, and THAT'S why it misbehaves.
0:54 is the most cool at all, solves immediately without going in the wrong way. I would be nice to see a complicated maze. You managed to fix the leaks with that nice cutting out technique so. Also if you use pressured air, then the solveing would happen emidiateally.
@@u1zha Even light does not goes immediately. There is no thing in the universe which you cannot possibly say "Immediately is relative", so what is the point of this correction?
Hi Steve, your water maze solver seems to work the same way as Metadynamics does in molecular-dynamics (MD) simulations. Here we'd like to accelerate the dynamics to get the molecular system out of a 'rut' and enable it to explore other parts of configuration space. Sometimes the ruts are too deep and you end up waiting forever for the computer simulation to jump out by chance, but metadynamics accelerates this process by filling in the ruts (potential wells, really) so that, like your water maze solver, the rut is no longer accessible, and then the simulation can 'spill out' into new configurations.
You should totally try with the maze on its side and an open (to air) top. No need for a vacuum silly 😊 Acrylic bottom, light table below, camera above, and multiple stages of colored dye would be my recommendation. The water will try to "fill" all parts of the maze and reach equilibrium. Depending on the setup with input/output controls, you can show a wide variety of phenomena. Electricity's tendency to find "the path of least resistance" would be one interesting one to see (probably best done with a color dye change). The water analogy for electricity is something I've always found helpful: water (mass) = charge water height = voltage (flow of) water = current
wow this is a cool concept. how did you come up with the maze pattern ? is it a pre-determined maze pattern or did you make it yourself? great video as always , simple and easy to understand concepts. having a visual to go along with a concept is always so much more understandable.
I actually searched stock images sites! I needed the file in vector format so I could send it to the laser cutters (after adding the tank). Stock image sites are good for vector files.
Neat! I would love to see a b-side of this where once the maze has been solved by the water, seal the exit and then see how much water pressure it would take in the tank to force all of the air bubbles out of their pockets
Every now and then I find videos like these that are on more of the complicated side of things and they really intrigue me. It just scratches the right spot on my brain to keep watching
Hi, great video, I love the mazes! I have a different hypothesis for why the water doesn't fully drain from the reservoir though (5:50), I believe it is likely that the maze is essentially creating numerous "P" traps that causes air bubbles between the different sections. After all the P traps are full, new water would have to compress the air in the bubbles to make room for the new incoming water, causing the resistance. Though if you added enough water pressure to push the air bubbles all the way through I bet you would get some proper syphon action!
Yes, it can't be surface tension because, although he points at lots of places where surface tension happens, the water flow is stopped by the first one of those places, so the rest of the surface tension sites are irrelevant.
What earned my sub? Was it the in depth and interesting explanation? Was it the amazing content and engineering? Sure these helped, but I can find them on most channels. No, the thing that has earned you my sub was the innovation, ideas, and problem solving you had on the side: Like mazes being two pieces or laser cutting holes in the 3rd layer. You make great stuff, so keep it up. Bravo Steve!
Possibly beyond the abilities of your laser cutter, but if you could make a membrane with holes which were small enough to let through air but not water you might be able to recreate the BergmanJoe maze that way. Lots of "air escape hatches"
Or one could "lay down" the maze and leave the top open. All you'd need to make water flow in with some urgency is an upward slope (a funnel or an angled tube for example) at the entrance. One could also make a small ceiling (or height gate) at the entrance limting the height the water will reach, in order to avoid splashover. This may be less difficult to execute than laser cutting a membrane
These are all one specific kind of maze: ones with one path, no loops, and an exit on the edge. Would be interesting to see this on a maze with a central exit and loops
You should be able to simulate the “airless” maze with bubble values that allow air out but not the water. Even just a tiny hole in the upper corner of any air dam.
Being that my last completely random recommendation from RUclips was about men in prison being "forced" into doing certain "activities" this was a HUGE step up. Thank you, this was done really well too!!
@@emrahyalcin yeah but if you lay it sideways, then the water won’t go down towards the exit it’ll just exist and in both the simulation and the real thing it actually has gravity pulling it down in the situation you’re talking about gravity is the thing spreading it not moving it
Did you consider trying this on a smaller scale? I'd like to see the same demonstration where capillary action becomes the dominant force instead of gravity.
Steve, have you considered repeating the water maze experiments using vacuum pressure? This is commonly done when filling products with oil or dielectric fluid that will be used underwater, where the fluid will be kept at external pressure and separated from the water using a flexible membrane or bladder. It is important to remove enough air so that bladder will not break when external pressure is high. A vacuum is applied during the filling process to remove air, but the amount it can remove is a function of multiple things, such as geometry, density, head, vacuum level, external air pressure, viscosity, and surface tension. It is more complicated to do than it looks.
So what happens if you lay the maze flat and fill it? I imagine you'd have to create a small tank at the inlet end so the water level can start out higher than the height of the maze.
Summary: spill a glass of water on a table. To some degree it would just fill the whole maze evenly. Assuming low enough flow rate at the inlet, there would always be a layer of air above the water but still within the maze. Therefore the behavior would probably be almost the same if there was no cover on top. In that case you could recreate the problem, by putting some objects on a table and then spill a glass of water (or more scientifically - pour water on the table at an approximately constant rate). Keep in mind that momentum of the fluid can impact the path significantly.
@@DarthBiomech the maze would still fill up based on gravity. You can't really pile up water because it just flows outwards, because the adhesion and surface tension forces are usually not strong enough to overcome the pressure caused by gravity pulling higher-up water down. So as all fluids, it spreads out relatively evenly.
@@sebastianjost Yeah, but in vertical example water tries to reach the lowest point, so it fills all lowest compartments first until it finds a hole, but on a flat surface it should be spreading out evenly equally in all directions on all intersections. In theory.
It depends on whether the maze is large enough for the water and air to pass each other. If it's a small maze, and the water and air can't pass, it'd solve the maze just as it does when the maze is vertical, because the trapped pockets of air have nowhere to go and can't get back out the top. But if it's a large enough maze that the water and air can pass, then it would fill the whole maze more or less evenly, and as the maze filled with water, the air it's replacing would come out both the entrance and the exit. Of course water would also flow out the exit, assuming that it was free to do so (if the exit was also an open tank, then the flow would stop when the maze was full and the two tanks had equal water heights).
This video explains why I had a blue splodge on my face in the last video!
The sponsor is 80,000 hours. Head to 80000hours.org/steve to start planning a career that is meaningful, fulfilling, and helps solve one of the world’s most pressing problems.
Awesome Steve :)
Nah. I’m good. Thanks tho
If you put small holes in the clear acrylic at the "top" of dead end pockets that are sealed it might cause the water to go more places. presumably more water would go down then out of the holes because gravity should still be stronger.
egg
Well you know Mercury has a pretty low pressure before boiling. Or you could try making a super fluid. That's always fun.
You know what I think can solve the puzzle flowing through the main path right away?? Smoke! Pushing smoke into the puzzle, it will flow through the path of least resistance from the start
Yeah! And as a bonus it can solve mazes where the exit is above the entrance
My guess would be though, that smoke would fill much more of the maze over time, because it doesn't have the surface tension of water.
Yes, that is almost the same as at 6:36 where the pre-filled maze is fed with different colored water. If the air pockets were also pre-filled with water, then it would be exactly the same.
@@OMNI-Infinity electricity kinda solves a maze in real life all the time - that's how the lightnings get their shapes
If you blow enough capacitors at one end it will create the smoke you need to solve the maze!
Directly to the experiments in less than 1 minute, now this is the channel that deserves to be popular.
Legitimately almost didn't click because I didn't want to sit through 8 minutes of "hype" for an underwhelming reveal at the end of the video.
This was not that, and I was pleasantly surprised
i mean it kinda is, not everyone has 2m subs
its just straight to the damn point
Yeah just forward the video 9 waste of time minutes
I love experiments.
Regarding surface tension, try solving the smaller maze with 'wetter' water. Reduce the surface tension of water by dissolving some alcohol in it. :) Might be interesting with different fluids as well.
Smart
Or pure alcohol, or add soap. Great idea!
@@SteveMould Or superfluid helium-4! 😂
I saw soap bubbles in your red/blue section so assume you used a detergent or other surfactant.
@@ericmoss6110 omg, superfluid helium would be an amazing test as it also can crawl *up* - it'd presumably solve the maze pretty much exactly by the keep-hand-on-wall method.
That said, if water was already a difficult thing to seal against, helium manages to be nigh impossible lol
Nice, now the alligator can finally have his shower
Nice where’s my water reference
I forgot that existed until just now
Almost didn’t know what you meant, w game tho!!
This was super nostalgic NGL 🙌🏼
the fact that a maze is two separate pieces blew my mind more than the water maze. It makes perfect sense but I have never thought of that!
100% lol, same here! I never thought of that and I STILL can't wrap my head around *WHY* that is, lol
+1
I'm confused why it has to be two pieces?
Oh, right, because one of them never touches the other.
It can be made of 3 pieces or 4 pieces or 5 pieces
Realizing every maze has two halves was a surprising revelation I didn’t expect.
Certain types of mazes*
Notice there are no loops in these?
It's actually pretty obvious when you consider a solution must exist and there are no islands
AT LEAST two (separate) parts. Not two parts. Many mazes have many separate disconnected walls. (There's no requirement for only having one possible solution in a maze.)
@@elliott614 That would make it 3 parts.
There are couple of important things this experiment shows, especially when dye is added. It shows what happens when there are unused lines in water delivery piping (like for example building hydrants sharing same line with tap water). It can become breeding ground for bacterias, eg. Legionella, despite water moving, and that's why it's important to use backflow prevention valves.
'water age' is indeed a critical factor in delivery quality water. This is why municipalities and other agencies controlling water supply model and test for water age as part of the design of new expansions to an existing system. Areas of known longer ages, are regularly flushed. Additionally, modifications to the distribution system can help eliminate these regions. one tool in this fight is preventing a dendritic (tree-root) branching of the system. By carefully utilizing loops and and multiple paths, it is possible to significantly reduce water age.
This is also why I always run the tap until the water is genuinely cold, to ensure it’s not been sitting around for who knows how long in the building or the street. Sometimes it happens quickly when there must be high turnover but sometimes it takes a few minutes.
@@kaitlyn__L Temperature isn't a bad guide, although the 'new' water from the street (which is buried, so likely cooler) will have to cool your warmer piping down, so once you feel the temp start to drop, you've likely got 'fresher' street water at the tap
That's why I don't trust the campaigns that say that water treatment is so good that you can drink straight from the tap. I do believe that the water from the treatment plant is great, but there's no way I trust those pipes between them and my flat, not to mention those in the building itself.
@@karol30660 I'm sorry that you distrust your water distribution system. I would suggest you sample your own water from your own tap and have it tested (at your own cost) against your national drinking water standards at a private, independent lab for your own peace of mind. If, in the unlikely event that your water falls outside of the national standards, your local distribution maintaining agency will be VERY interested in your results.
Maze-solving algorithm for red water:
1. Find someone who solved the maze with blue water
2. Copy their homework
I had never thought of a maze being in two halves - and yet it's so obvious when you think about it! Steve is like the Johnny Ball of the digital age 🙂
Not only that, but the solution is exactly the path between the two halves. (I think. Proving this is left as an exercise for the reader.)
@@Khaim.m Yea! I got that once I got that it was in two halves - just never thought of either of those things until I watched this! I love the way his videos make you look at everyday things in a new way 🙂
It's for the same reason why there can't be any two-dimensional beings. Their digestive system would just cut them in half.
@@lonestarr1490 What about the top of the head? 🤔
@@lonestarr1490octopuses eat and poop through the same hole, so it could work
I loved the solution of cutting slots in the upper acrylic plate to allow the solvent to reach the surfaces you wanted to solvent weld.
For the smaller maze, you can just fill a flat metal/ceramic tray with just barely enough solvent to cover the bottom. Then put the maze into it with the side you want to solvent weld. Leave it in there for like 10-30 seconds. Then take it out, and set it onto the panel you want to weld it to. Because the maze is so finicky, I'd still do the first side with the syringe.
This process works for all kinds of parts and actually tends to give a cleaner, more even, weld. However, since you're using laser-cut pieces you'll have plenty of micro-cracks in them after getting them into contact with solvent (unless you tempered them afterwards), they are probably just hard to see in the black acrylic.
For large pieces this works too but of course you'd need a pretty large tray (or one custom-made for the job) and a lot of solvent. You also don't need to actually submerge the pieces, just lowering them until the solvent pulls itself onto the welding surface is actually better (because it only coats the surface and because the solvent doesn't get squeezed out) but a bit finicky, you might be able to put some uncoated steel wires into it as an offset.
Source: Sometimes do this at work, but for most pieces the syringe method works just fine.
7oq
Regarding using the "hand on one wall" method to go through a multipath maze, as long as you pick a wall right at the entrance then you should still always find your way out. Any extra paths would create a third piece to the maze, but by necessity it would have to be in between the two outer parts which must be the two parts bisected by the entrance and exit.
I wrote a longer comment about this but also if there are 2 entrances and 2 exits, you must find a path from an entrance to an exit, and the paths between each entrance and the exits connect, there is a chance the wall you follow will lead you on a path between the two entrances instead of to an exit, but that's the only thing I can think of that would defeat the algorithm, barring dropping someone in the middle of the maze instead of starting them at an entrance... or having no paths from the entrance to an exit, obv.
@@JuniperHatesTwitterlikeHandles You can theoretically have traps and one way doors in a real maze, so that you can go through a particular section one way, but not leave through the same area - meaning that you'll get trapped in the section beyond the one way door.
Or you place the exit in the "middle" part by having a tunnel up or down from one of the rooms that's only accessible through the third part of the maze.
Another solution I've seen in real world mazes is to have checkpoints in the maze that have to be reached for the maze to be considered complete.
This only works if the starting point is along the outer wall of the maze (which is _generally_ the case, but not necessarily). If you are starting from a random location within the middle of the maze, then you have no way of knowing whether the wall you touch would be connected to the exit or not.
It is also possible for the exit to the maze to be a staircase or ladder or something which is placed within the maze. That would mean that you could follow the outer wall in a complete loop and never find the exit.
Someone should make a free demo and share or code for a 3d simulation, this could be in unity or unreal or something like that.
Could be a neat water physics simulation to use for games.
@@icravedeath.1200 I'm pretty sure UE does offer water simulation. Blender has had physics simulations for a long while now. Don't know about accuracy in a scientific level, but they are more than amazing for what ever I will ever make :D
If you want a realistic level of simulation you'll probably have to implement the whole rendering engine and physic system by yourself
Unity is no longer an option.
The "keep your right hand on a wall" always works, IF you start doing so from the entrance!
A maze may have two or more paths to the exit and therefore has more than two wall structures, but it will always have just two outside ones.
I thought the same thing!
Actually it doesn’t work at all if there are any floating walls in the maze, as in walls that don’t connect to the outside. Left or right wall doesn’t matter. Sometimes the floating wall is obvious and sometimes it’s not. Oh, and it only works if you can touch the outside wall directly from the entrance.
@@thehunters3402 At the entrance there is always an outside wall! It may be part of a 'floating' wall, yes, but then there´s more than one entrance. In that case, switch the same hand to the other wall of the newfound entrance.
@@sthenzel I know. Just expanding on the conversation because that’s what I like to do.
I've always wanted to map the Paris Catacombs by spray painting a white line along the right wall.
With your second complicated, big maze, I don't think it is surface tension but an air lock, the path repeatedly goes up then down so if there is a bubble in each up bit, it requires a minimum head of water to overcome the sum of the uphill parts.
Agree
Agree too
My kitchen sink agrees
@@LucasHartmann My car cooling system agrees.
Physics agrees
man, you really nailed the lighting on the larger mazes, the contrast of the "walls" is perfect
If you get stuck in the Backrooms- just flood it
gonna have to find a lottt of water for that
@@inimmzyou would need a bottomless water bucket more precisely
Flood fill just got real
Indeed
How do I do flood fill with voxels as fast as possible
The fuck
Tf is flood fill
5:40
True,
if you only start using the 'hand on wall'-tactic as a backup plan, after you've already been in the maze for a while a multi-solution one can trap you.
Doing so from the start will work (for any 2 dimensional maze, where 'out the other side' is the goal. Get to the centerpiece artwork (or staircase to the exit-tunnel), are made out of 1 piece and may loop you back to the entrance )
If you're able to mark the walls in any way at all, if you loop back to the same point on the wall, you are following a center wall that doesn't connect to the entrance or exit. Switch to the wall opposite and follow that one. If necessary, just keep mapping these interior walls until you get to a wall that leads to the entrance or exit. I follow walls all the time in games to fully explore areas and find places I've missed. If you're in a maze in real life and you fear you don't have all the time in the world to walk around--definitely find a way to mark your exploration or break yourself out of there.
E
@@EEEEEEEE real
If you wanted to simulate the simulation you could put a small vent hole at the top of each isolated section of the geometry and then just cover/plug the hole once water starts leaking out of the venting hole. That should be easily done in the larger versions and you just need something like flex tape to cover the vents. it won't look as pretty but the physics should work.
Obligatory flex tape meme here
Putting in a air release valve would also work so he could avoid manually covering the holes
Also, by treating every maze surface by a hydrophobic repellent, you could reduce the surface tension friction. Would be fun to see the results and how they compare to the vacuum simulation 🙂
Wouldn't that be like creating a maze with multiple exits or paths? Not really a maze?
@@YoungGandalf2325 See them as emergency exits and hang little "no exit, staff only" signs on them. The air can be considered staff.
This is the first video ever where I saw the short, thought it was cool and actually watched the real video. That’s how good your content is
I love by the way how you included the animation clip at the beginning but still motivated us to watch the original which supports its creator. Just love it
E
Also that he wasn't 'shaming' the animation creator (for lack of a better word) for not having realistic graphics but instead said things along the lines of 'he animated something I can't recreate'.
5:28 Even in a multiply-connected maze (one with multiple successful paths), you can be sure that the walls of the ENTRYWAY will be the walls of the exit, as long as the entry and exit are on the outside border of the maze. I didn't realize that until you showed your "coloring the walls" at 2:50 then mentioned that two paths would create a middle part. That middle part won't be the left or right wall of the entrance, but would necessarily be an "island" in the middle. The interesting thing is that you can't just wander around and decide later to use this algorithm: you need to commit to it at the very entrance to the maze. And, it only works if the exit is also on an external wall, not say a stairs in the middle of the level, or say a goal or treasure of some kind not adjacent to an exterior wall.
This doesn't work in 3d mazes (multi-level)
Though generally not a problem, if you're starting from the middle of a maze (e.g. you're blindfolded and carried to the starting point, or you forgot to use the strategy until later), then there'd be no guarantee.
@@tcxd1164 I explained that in detail.
@@lqr824 yeah, I reread that more clearly after I already posted the reply, but oh well.
It could also be the case that the exit/goal is somewhere in the middle of the maze, and there's only one entrance/exit, then it could be made of a single piece and thus following a wall will take you around the entire maze (including both entrance and exit)
This shows the difference between physical and digital simulations. In digital there is always something overlooked like the ire pressure in this situation which is why its always important to test the digital one on a small scale first.
Yup.
It's interesting that lots of times when I'm reading good info on using a simulation or simulator of/for something it's recommended to do the real thing along with a simulator.
I have peripherals for simulator games but should probably get rid of them because I don't access to the real world counterpart to get the most out of the simulation.
I'd correct that to say "there's usually something overlooked". It's theoretically possible to have a perfect simulation, but can be very hard to do.
The point of a making a simulation in science is to put in all the variables that you think effect the outcome and their strengths and interactions and so on and then you compare it with real life observations. And then you expect to find that your simulation is inaccurate at some point and you try to find out what caused that inaccuracy and by finding out you learn something new...
Simulations are always inaccurate, they are supposed to be, as soon as a simulation is perfectly accurate your field of study is solved, but, so far, there have always been more questions left to answer in any field of study and thus more inaccuracies left to iron out... At in the natural sciences, biology, chemistry, physics, etc. I am not sure, since I am not a mathematician, but I think there are parts of mathematics that are solved...
@@Kholdaimon Solved is one thing, but how fast is the simulation? What if you need to run hundreds of simulations for engineering and design?
@@LutraLovegood I mean, you never run a simulation once and if it gives the result you want call it done. You always run it hundreds of times so you can do statistical analyses on the results... When I did my research the standard was 1000 iterations of a single set of input parameters and then analyse the results, which took about 3 hours on my fairly good gaming computer, but this was 10 years ago, my current computer would do it A LOT faster...
The speed depends entirely on the nature of the simulation, some are really fast, others require hours and hours on university owned super computers.
The water looks more like water than real water.
Thats....not even possible
That last part is so interesting. Blue water has figured out the maze and now the red water can go directly to the end with virtually no loss
I wonder if there's some kind of force(?) flowing upstream from the end to the start??
@@janbam1778 A negative pressure difference, basically. Like pulling water through a drinking straw.
If it was a human sized maze maybe we could follow the current and also get out
The blue water is basically doing the same thing that the air did in the smaller maze. It's blocking the "wrong" paths, and instead of surface tension, there's just the boundary between two liquids of the same density - they're not going to mix very much at all in such a small time frame, so it's basically just a question of whether the blue water has anywhere to go when the red water pushes against it.
That's because the 1st pass water has blocked off all the dead ends, making it so any pressure can only be relieved directly towards the exit.
This video actually made me realize that if you just try to visualize how the maze is in two pieces it becomes easier to solve visually. Pretty cool! I also think it works how a siphon works.
It’s also why if you keep going right, or keep going left, you will always solve the maze. Unless, of course, the maze is broken up into more than 2 parts.
Which, by the way, knowing such information is good for you in emergency situations. If you are ever in a building under evacuation and it is pitch black inside, put your hand up against the wall and follow it. You will eventually reach a door and/or window.
@@howardbaxter2514 Yeah, but it might only be an internal door...
Love this! Mazes were "my thing" at school. We had exercise books with squared paper in them for maths, and I used to draw mazes all the time, so I became really familiar with a lot of stuff about them. The "stick to one wall" rule is a good window into some of them, as it makes you think about where the entrance and exit might be (could be in the middle), whether it still works with multiple entrances, whether there's a way to use it to escape a maze if you start in a random spot (there are algorithms that can eliminate dead ends as you go, but they require some sort of memory or external modification)...
A third dimension also stirs things up. The maze at Longleat in England (which was the largest in Europe in the 1980s, not sure whether it still is) has six bridges that go over the hedge paths. This allows you to see some of the connectivity from above, which people imagine might help - but the presence of the bridges also makes the topology more sneaky and the maze harder.
I LOVE MAZES! Cool to see your construction skills applied to them like this. Excellent video.
And now you could generate big mazes in minecraft and explore them in virtual reality... yes, the world is crazy like that today.
@@chrisakaschulbus4903 do you mean a particular mod for mazes?
@@nikelsad I made mazes with server plugins a couple years back. There are also "picture to schematic converters". So combining that with a normal image maze generator should also be possible. Particular mods or names i don't have, though.
I only generate mazes for CNC laser action.
Dude, that sounds awesome!! I think one day I'll make this a birthday trip for my ex-girlfriend. She also loves everything logic-related. This christmas I gave her the hardest version of the Perplexus 3D labyrinth, she fucking loved it and tried cracking it for a over a whole day 😊 Have you heard of those? They are more about the motor skill to bring the ball through the maze. The hardest one is actually pretty hard!
Thanks for the unintended present idea 😉
@@muenchhausenmusic You're welcome! Yes, I think I know the puzzle you're referring to. I don't have one but I think I've tried it once.
I came here from the short to ask what would happen if you put dye in after it was solved, very glad to see that it worked exactly as I expected! Thank you for doing more than just answering the face question and thinking of more interesting things that can happen in the situation!
Why are the water videos always the best?
Watching this is so soothing. I feel like someone could make an entire channel of just water mazes.
Hey Steve, I do quite a bit of solved welding acrylic. For the second half you can do the dip soak method, where you 'soak' the acrylic in a thin puddle of solvent (such as in a custom HDPE or PTFE tray) and then clamp it to the 2nd clear acrylic part. Great video!
That’s basically what I did for the smaller ones! Question for you: is it a good idea to soak for a long time? I just dipped.
@@SteveMould Yes, you would want to soak the acrylic in the solvent to soften it, making it tacky so that when you make contact and clamp with the other piece of acrylic the two will bond in the same way that applying the solvent via capillary action does. Soak time will depend on a couple of different factors such as if the acrylic is cast vs extruded, if it has any colorants or co-polymers in it, and which solvent weld you are using (I am assuming you are probably using weld-on 3 or 4 or some other dichloromethane base solvent), and what clamping pressure you are using. You'll need to play around and find what works for you, but I would start with a 15 second soak and go from there.
Since you are using lasercut acrylic, you'll need to be extra mindful of the acrylic crazing when you do this. You may need to anneal your parts before soaking, but I hope you can get away with not annealing since it can deform and change the dimensions of the part. The stresses in the crystalline lattice near the cut edge (the cause of crazing) are built up differently between laser cutters depending on the ambient temperatures, speed of the air assist, etc. If you are able to stop the cut edge from cooling as rapidly that will help with the crazing. The easiest way to do this is to lower the airflow of the air assist.
I actually used this once in a dnd campaign. Our DM put us in a huge mirror maze room. Using a combination of a few magic items, I made a torrential downpour of water and we waited a minute before following the flow
That’s clever, love it!
How did you keep it from back-flowing out of the maze? I mean this method only really works because of the verticality of the maze making it so the water cannot simply flow back out of the entrance. You would likely need way more than a minute in order for the water to rise to the point where you could detect a flow is a big room. My solution to DM's throwing mazes at casters is always just fucking blow through it or use divination magic to scout the correct path, ain't nobody got time for that.
@@ShiningDarknes I play as a ratfolk, so I have an unfair advantage
@@sethb3090 how so?
@@ShiningDarknes Why wouldn't you exit through the entrance, as long as you get out right?
This is amazing. You could hypothetically have a huge containment of water with multiple maze plugs of different tensions to simulate different distributions instead of larger or smaller ducts
I absolutely LOVE that this had unexpected turns, like just completely ignoring whole sections because air pressure kept them out
Also interesting that near the end for a while, the air was bubbling up from extraneous sections and pushing some water towards the exit (again because of air pressure!)
This is a great analogy for another phenomenon. I remember when people were very surprised when slime mold could solve mazes. Were they really clever enough to find that out? No, they simply did the same thing as the water. They went everywhere, but when they reached a dead end, the slime mold that followed that path would simply die, as it was not part of the shortest path.
I really think you should look into getting the big versions of the mazes into a gallery. They are really beautiful art pieces against that white background. The blue liquid is really radiant
The water in the tank came to a halt like a car going uphill with its engine’s torque just short of power. The water in the maze needed more weight pushing down so it could push up all those nooks and crannies and fight against gravity and surface tension alike. Truly fascinating. :)
@SteveMould What a great video! It was cool to see how fluids actually travels through a maze with real world physics. Also What a great sponsor. 80,000 hours was something I didn't know about till now and I'm going to look at that organization ASAP.
l put my hamster in a sock and slammed it against the furniture.
Doing satisfying experiments and explaining the science at the same time is great. Keep up the good work!
E
Regarding the 2 paths leasing to 3 walls you can 'hug'. It still works holding the hand on a wall, if you do so when you enter the maze. The 3rd middle part that loops wont be one of the walls at the entrance. So unless walls are shifting its a pretty safe approach
The title of the RUclips video (Maze Vol.01 Simulation) is a maze simulation using OpenFOAM.
Channel subscriber: Take Den [OpenFOAM & Blender]
Definitely one of these concepts that I never would've thought about on my own but are really interesting if someone brings them up.
If you wouldn't mind the mess, you could maybe drill small holes near the top of the "air chambers", so the water could push the air out of the maze and get closer to bergman's simulation?
and connect them to tubes pointing upwards so the pressure would equilibrate and you don't have blue liquid everywhere :)
I think, that mercury could solve this problem
@@nico4331 this is the solution. Just need the tubes to reach higher than the top tank and you'll have no problem with water escaping that way.
You've scaled up one of the dimensions of the system without scaling any of the other relevant variables (viscocity, density, simulation time). Neat to see how the system changes from that one input.
I'd be interested to see this run by scaling all of the relevant variables to keep the system's relevant dimensionless numbers the same (reynolds number, etc). Would make for a great fluid dynamics example. Smarter every day or veritasium did a video on a boat research that talks about this a little
EDIT: It's veritasium's video at a naval wave generator
Science usually makes me want my head to implode but you make is sound so fun! Your tone of voice and accent and way of explaining is just perfect, love this video, you got a new subscriber!
I had a feeling it would be surface tension being too strong for the water to overcome it. You present your paradoxes and caviats in such a good manner its more like an assist than me just knowing it
I love how straightforward the video was. Thank you for getting right to the point and not wandering around.
Unlike *_”cough”_* Vsause-
dude this was pretty dang good. you dont know how hard it is trying to explain how water works to people. and this video is gonna help them understand with a visual.
6:00 not only just surface tension but as you said earlier for the small mazes, air pressure. The first lip looks impossible if no air prevents the water from moving. All the little pockets of air that you see have formed along the correct path (unlike the incorrect path in the small model, although there is still some of that) to help hold up that massive head of water.
You're right, I would even say that it's pretty much only about pressure, surface tension has very little impact here.
Pressure difference is nicely visible in the little siphon in the top left corner - free water surface is much lower on the right side than on the left which means that air pressure on the right is higher. And you can see why - because the air on the right is sealed off while that on the left can still escape (that happens in 5:30).
You can also nicely see how the air pressure grows in the whole sealed right part of the maze. In about 5:15 the right part is sealed off and then water goes further up to top left corner of the maze (reached at about 5:27) on the left side of that seal, and you can see how the free water surface on the right side also moves up but much less, just enough to reduce volume and increase pressure of the air trapped there so that it equalizes growing hydrostatic pressure of water column on the left.
Also, as the water stops flowing, the difference in height between tank above maze and surface on the other side of first siphon and that of to two sides of that column of water dividing right and left side are visibly similar (not exactly the same, because there are other siphons with uneven pressure on both sides).
There is good explanation on practical engineering channel about this phenomenon.
ruclips.net/video/zdkp9N3qfkI/видео.html
this is the perfect video to watch in 4K
correction, at 6:10 it isn't surface tension stopping the water, it is the air trapped in the system separating the water and thus the column of water is shorter on the way down than up, so the pressure isnt enough to go all the way up again
It's not the air trapped in the system that's stopping it. When there is enough pressure at the top, the air trapped in the system just gets pushed out.
Though possibly if the surface tension was lower, air could have escaped?
@@unfa00 Adding soap to the water would be an easy way to test the hypothesis. If the hypothesis is correct, the water with soap will stop flowing at a lower level than the water without soap.
@@unfa00 the air still wouldnt travel downwards
@@FirstLast-gw5mg yes if the pressure is higher the water flows, but it doesnt push out the air. Otherwise it wouldnt stop flowing once the pressure drops enough
Breadth-first search versus depth-first search versus ant-colony-optimisation for maze solving it'd be interesting to explore which algorithm solves what types of maze first and come up with metrics to describe certain maze classes like length of passages versus number of junctions etc :D
This was very satisfying to watch. I like that you mention the little details you discover/know along the way
I know there is no way he sees this comment since it’s an old video but I think if you were to lay it flat record from top down and have the water enter through a sealed port. The results would be much different as the air would have more room to move.
This is actually quite cool. Reminds me of how water in minecraft also does pathfinding.
More like gravity doing its work
@@mcmarkmarkson7115 and air pressure. And surface tension for the smaller ones. The only one where gravity does the work alone is in the simulation
@@kebien6020 air pressure is also gravity related though. Like what makes water move and air move?
Maybe I'm being too pedantic, don't take it too serious.
i myself was thinking of terraria, like with the air pockets, how you can dig sideways and then dig upward to creste an air pocket to avoid drowning
@@mcmarkmarkson7115 only atmospheric air pressure comes from gravity. The one involved into not letting water in on some sections of the maze despite gravity pushing the water in that direction is not atmospheric air pressure.
Can I suggest that you hold the maze horizontally, and make the maze with the base and the walls (not the top). Fill the maze completely with clear water, and allow continuous flow of clear water into the entrance and out the exit. Once it's settled down, use some coloured dye to reveal the path of the flowing water.
Lay the maze on its side and take the top sheet of acrylic off.
You'll have to be careful with the flowrate or increase the depth of the walls to stop it flowing over the top but you wouldn't have the issue of airlocks on the dead ends.
Would become a pressure driven flow instead of a gravity driven flow.
Id think it would still look different to the simulation but would imagine it would fill alot more of the odd routes up.
I was thinking put small holes over one side of the acrylic (back side so it doesn't spoil how it looks too much) and then make sure the amount of water going in is greater than the amount that can leak out of the holes. Again don't think it would quite match the simulation but think it would be close.
Then you wouldn’t have gravity
@@insertnamehere4775 Yes/No, having a feed at one end and it being empty at the other it will eventually flow through. with a very low amount of head. However, it might not look like the simulation exactly, but interested to see how it compares with what is shown in the video.
@@MyMmmd _"...put small holes..."_
That just hurts my brain :p It's like adding more cold :p "Please close the door, you're letting all the cold in!"
As a physicist, I must say that it looks oddly satisfying (and accurate). This is the first time I see this channel btw. Immediate subscription activated 😌
Your casts are uniquely appealing... information presented in a manner we can participate in... solve/understand what we are experiencing 😁 Much gratitude for your energy ❣️
You could try substituting the water with some type sort of oil used in vacuum pumps to test the maze under a vacuum. The oil shouldn't be too hard to get your hands on and they all have very low vapour pressures
4:55 - you dont't need vacuum, allowing the air to escape is sufficient. Thus making small holes at horiontal lines of the maze would sufficient. That of course creates a problem that water would leak through as well, but water has much larger buoyancy than air, so having a small plug at the hole, that would seal it once water reaches there might do the trick (though I am aware it is not that simple to actually build it that way, but you know, in principle it should/would work :D )
How about adding a fine mesh in front of vent holes. At the top there should be relatively low pressure, so a sufficiently fine mesh should prevent water from leaking out due to surface tension.
However air could still pass through that mesh.
Surface tension should keep the water from flowing through tiny air holes. But it's hardly worth doing, since we already know what happens.
This is a visualization of Parallel Breadth-First Search, which is quite interesting. With multiple molecules in the water (a vast number), it can search for the shortest path simultaneously. The BFS search also has the advantage of finding the optimal solution.
As mentioned in the video, the air pressure prevents the water from going the wrong way, making it very different from the maze-solving video. However, that simulation can also be like the real world by adding some functions. Typically, the maze-solving agents we see in videos can only sense if their nearest cell is open or closed. But in this situation, thanks to the air pressure, it can sense much further. Have no idea how to implement the sensing feature, it can be done somehow.
For anyone wondering about implementing this algorithm...
I never thought about being able to "cheat" my way through a digital maze by coloring it 😂 Very logical if you think about it but it never came to my mind, that every (solveable) maze has to be made of at least two pieces!
2D mazes. A line can run through a 3D maze without separating it. Ex: a torus, ex: a donut, or a sphere with a kind through it. You would need at least a 2D object to separate it
Your tests are great to be used by cave divers, people going deep underground exploring cave systems.
So that in case of storms outside and flooding in the caves, they ca use as a model to predict safe places for setting up camps.
I've seen a story about such a situation of a deep dive and a storm that lasted several days outside leading to huge flooding in the cave.
Based on this, I think it would be an interesting although potentially simple challenge to purposefully design mazes to defeat this. Firstly, design a solve path that requires a large amount of distance to be covered vertically against gravity. Second, design dead ends that cover a large amount of distance vertically with gravity and have a large amount of volume. Lastly, make the channels small.
Water will fill the dead ends first and you will need a large amount of water to do so. You will then need a large amount of head pressure to even get water through the solve path. If done properly, you'd have a maze that basically completely fills before being solved and requires either a source of high pressure water or an exceedingly large reservoir.
Taken to the extreme, you could have a maze where the total resistance to flow is so high, water can't make it through.
That was very nice !!! Coming from the shirt and absolutely loved that you thought about phone users and used the tool to link to this bigger video ❤
I love how you straight-up admitted "the reason the small, complex maze doesn't behave properly is just because I fricked something up". No complicated terminology to dance around the problem; you laid it right out that you couldn't get it watertight, and THAT'S why it misbehaves.
It might be only a 9 min video but i am pretty sure that the entire project took not less than 9 days of hardwork and dedication
Not only water but also gravity
0:54 is the most cool at all, solves immediately without going in the wrong way. I would be nice to see a complicated maze. You managed to fix the leaks with that nice cutting out technique so. Also if you use pressured air, then the solveing would happen emidiateally.
"Immediately" is relative. Of course there would be some movement imparted to air in the dead ends too.
@@u1zha Even light does not goes immediately. There is no thing in the universe which you cannot possibly say "Immediately is relative", so what is the point of this correction?
"maybe i can try with a liquid that doesnt boil in vacuum, but that is hard". I think you found a new problem to solve😂. Lovely video, thanks😊
That reminds me of the trope you see in fiction all the time, where people navigate out of a mazelike cave system by following a breeze.
The air doesn't smell so foul?
When in doubt, Pippin, always follow your nose? 🙂
E
This is exactly what's happening here. You could follow the breeze.
Watching water solve a maze is like witnessing the world's most patient plumber
Hi Steve, your water maze solver seems to work the same way as Metadynamics does in molecular-dynamics (MD) simulations. Here we'd like to accelerate the dynamics to get the molecular system out of a 'rut' and enable it to explore other parts of configuration space. Sometimes the ruts are too deep and you end up waiting forever for the computer simulation to jump out by chance, but metadynamics accelerates this process by filling in the ruts (potential wells, really) so that, like your water maze solver, the rut is no longer accessible, and then the simulation can 'spill out' into new configurations.
You should totally try with the maze on its side and an open (to air) top. No need for a vacuum silly 😊
Acrylic bottom, light table below, camera above, and multiple stages of colored dye would be my recommendation.
The water will try to "fill" all parts of the maze and reach equilibrium. Depending on the setup with input/output controls, you can show a wide variety of phenomena. Electricity's tendency to find "the path of least resistance" would be one interesting one to see (probably best done with a color dye change).
The water analogy for electricity is something I've always found helpful:
water (mass) = charge
water height = voltage
(flow of) water = current
If you put the maze on its side, there will still be enclosed spaces; in practice the highest roof will just be lower and jagged.
I agree, or just lay down the mazes that have already been built and fill from one end.
wow this is a cool concept. how did you come up with the maze pattern ? is it a pre-determined maze pattern or did you make it yourself?
great video as always , simple and easy to understand concepts. having a visual to go along with a concept is always so much more understandable.
I actually searched stock images sites! I needed the file in vector format so I could send it to the laser cutters (after adding the tank). Stock image sites are good for vector files.
Neat! I would love to see a b-side of this where once the maze has been solved by the water, seal the exit and then see how much water pressure it would take in the tank to force all of the air bubbles out of their pockets
Every now and then I find videos like these that are on more of the complicated side of things and they really intrigue me. It just scratches the right spot on my brain to keep watching
I think you're the content creator I've followed the longest, maybe 10+ years now. Can't believe you're still producing really interesting stuff
Hi, great video, I love the mazes! I have a different hypothesis for why the water doesn't fully drain from the reservoir though (5:50), I believe it is likely that the maze is essentially creating numerous "P" traps that causes air bubbles between the different sections. After all the P traps are full, new water would have to compress the air in the bubbles to make room for the new incoming water, causing the resistance. Though if you added enough water pressure to push the air bubbles all the way through I bet you would get some proper syphon action!
yep, heaps of airlocks
you get them all the time in caravan grey water pipes
ruclips.net/video/oHzAuLcE7DU/видео.html&ab_channel=DamianCandetti
just like this video
Yes, it can't be surface tension because, although he points at lots of places where surface tension happens, the water flow is stopped by the first one of those places, so the rest of the surface tension sites are irrelevant.
What earned my sub? Was it the in depth and interesting explanation? Was it the amazing content and engineering? Sure these helped, but I can find them on most channels. No, the thing that has earned you my sub was the innovation, ideas, and problem solving you had on the side: Like mazes being two pieces or laser cutting holes in the 3rd layer.
You make great stuff, so keep it up. Bravo Steve!
Possibly beyond the abilities of your laser cutter, but if you could make a membrane with holes which were small enough to let through air but not water you might be able to recreate the BergmanJoe maze that way. Lots of "air escape hatches"
Or one could "lay down" the maze and leave the top open. All you'd need to make water flow in with some urgency is an upward slope (a funnel or an angled tube for example) at the entrance. One could also make a small ceiling (or height gate) at the entrance limting the height the water will reach, in order to avoid splashover. This may be less difficult to execute than laser cutting a membrane
I especially love these water videos because I’m a visual person. Seeing the transparencies helps explain a ton. Keep them coming!
These are all one specific kind of maze: ones with one path, no loops, and an exit on the edge. Would be interesting to see this on a maze with a central exit and loops
Indeed
system: you are a maze master
Put some dye in to see if the water in the trapped parts actually makes it way out. Interesting!
Its always a joy to see your 2d liquid models!
Agreed
You should be able to simulate the “airless” maze with bubble values that allow air out but not the water. Even just a tiny hole in the upper corner of any air dam.
Being that my last completely random recommendation from RUclips was about men in prison being "forced" into doing certain "activities" this was a HUGE step up. Thank you, this was done really well too!!
I cant believe how many things I learned from this. I was not expecting this much from a "pouring water in maze" video! 😃
0:49 Just going to put this here to immediately get to the most satisfying part
at 4:50 You could just make lie down the maze horizontaly to the ground instead of vertically. so air woulndt be issue.
No, because in the simulation, there’s gravity
@@Skyblue92u gravity still applies to the situation because as the water volume increases , gravity will make it spreaded
@@emrahyalcin yeah but if you lay it sideways, then the water won’t go down towards the exit it’ll just exist and in both the simulation and the real thing it actually has gravity pulling it down in the situation you’re talking about gravity is the thing spreading it not moving it
Did you consider trying this on a smaller scale? I'd like to see the same demonstration where capillary action becomes the dominant force instead of gravity.
Steve, have you considered repeating the water maze experiments using vacuum pressure? This is commonly done when filling products with oil or dielectric fluid that will be used underwater, where the fluid will be kept at external pressure and separated from the water using a flexible membrane or bladder. It is important to remove enough air so that bladder will not break when external pressure is high. A vacuum is applied during the filling process to remove air, but the amount it can remove is a function of multiple things, such as geometry, density, head, vacuum level, external air pressure, viscosity, and surface tension. It is more complicated to do than it looks.
This is fascinating, really nice work on constructing the mazes too.
wtf?
This is actually very similar to how valve bodies work in automatic transmissions. Great stuff!
So what happens if you lay the maze flat and fill it? I imagine you'd have to create a small tank at the inlet end so the water level can start out higher than the height of the maze.
Huh, yeah, a completely level maze wouldn't be able to fill up via gravity, so it's interesting how it'd behaved then.
Summary: spill a glass of water on a table.
To some degree it would just fill the whole maze evenly.
Assuming low enough flow rate at the inlet, there would always be a layer of air above the water but still within the maze.
Therefore the behavior would probably be almost the same if there was no cover on top.
In that case you could recreate the problem, by putting some objects on a table and then spill a glass of water (or more scientifically - pour water on the table at an approximately constant rate).
Keep in mind that momentum of the fluid can impact the path significantly.
@@DarthBiomech the maze would still fill up based on gravity. You can't really pile up water because it just flows outwards, because the adhesion and surface tension forces are usually not strong enough to overcome the pressure caused by gravity pulling higher-up water down. So as all fluids, it spreads out relatively evenly.
@@sebastianjost Yeah, but in vertical example water tries to reach the lowest point, so it fills all lowest compartments first until it finds a hole, but on a flat surface it should be spreading out evenly equally in all directions on all intersections. In theory.
It depends on whether the maze is large enough for the water and air to pass each other.
If it's a small maze, and the water and air can't pass, it'd solve the maze just as it does when the maze is vertical, because the trapped pockets of air have nowhere to go and can't get back out the top.
But if it's a large enough maze that the water and air can pass, then it would fill the whole maze more or less evenly, and as the maze filled with water, the air it's replacing would come out both the entrance and the exit. Of course water would also flow out the exit, assuming that it was free to do so (if the exit was also an open tank, then the flow would stop when the maze was full and the two tanks had equal water heights).
Water really is the universal solvent
It’s always a joy to see your 2D liquid models!
This is honestly a really neat video with some very clever bits of trivia! Thanks!
Now this is cool!
Love your content, always so interesting and educational.
Same!
So much built up tension near the end. I was expecting it to cut off abruptly😭