Engineering a TESLA VALVE in Timberborn!
Vložit
- čas přidán 6. 07. 2024
- We're testing out the BRAND NEW water physics of Timberborn and seeing whether a tesla valve finally works in this game!
LINKS!
PATREON: / realcivilengineer
MERCH: www.realcivilengineer.com
MEMBERSHIP: / @realcivilengineergaming
REDDIT: / realcivilengineer
TWITCH: / realcivilengineer
PADDY (MY DOG): / @paddytheapprentice
STREAM ARCHIVE: / @realcivilengineerarchive
Epic Game Store Support-A-Creator Code: RCE
(In connection with Epic Games’ Support-A-Creator Program, I may receive a commission from certain in-game purchases)
#realcivilengineer #timberborn #tesla - Hry
He kept going on about how the poo water source was "more viscous" and "faster", but I think it's only because the source generates from all of its block spaces, putting it ever so slightly ahead of the other water sources he tested against by volume and position.
Went looking for this comment
Yeah if he had the regular water start at the front of the 3x3 then it would be even which he litteraly tested and got wrong lol
Doesnt viscocity reduce flow velocity cuz there is more internal friction
Yes but either the game doesn't calculate that or badwater doesn't have less viscosity because the flow rate is the same
I noticed that as well, which will you think about it?His dirty water started a block in front of the one by three in the middle water. So now all his Testing is skewed in incorrect
Doing fluid simulation is truly an absolute nightmare mathematically, I can't blame the devs either for trying to avoid doing water pressure/flow rate interaction calculations, and the turbulence intereference. What they've already added with basic head and pressure representation and better 3D water representation overall is already a huge step forward for the game! Especially given that only a small part of the playerbase like us will ever actually care to put them to use! Big ups to the devs for that!
I doubt they implemented full fluid mechanics, it'd be too resource intensive. But hey! Partial is good enough
I was about to comment this. Fluid simulation is something I still have horrible dreams about.
To an extent you’re right I mean I’m not a drainage engineer or someone who needs to calculate fluid dynamics but I will say I appreciate that you can adjust power sources manually now! That’s a really cool addition to the game that not everyone will use but that’s an awesome addition that even normies like me can appreciate 😂
I feel like they should talk with the guy who made the creeper world games. That might give them a leg up or at least a starting point or whatever.
@@walkrvc whoa, I never expected to see someone with so much culture here.
Creeper world is a banger 😅
I assume this video was recorded on June 25, so let's ad this from July 3:
"When a fluid reaches the ceiling above a covered fluid source and there’s a way for it to escape, for example to the side and upwards, it will do just that. In short, this means that you can now assemble pipe-like setups."
Also, vanilla vertical power shafts and integrated mod support! :) ;) :)
Edit at the end of the video: Also "When placing multiple Levees at once, you can now draw rectangles rather than just lines. Go crazy with the aqueducts!"
Vanilla vertical power shafts is huge.
next test, do beavers need air? make tunnels for the beavers to pass through, and flood them. make a nice long route for the beavers to pass through, so they take a longer and longer time to pass through, underwater the whole time.
This challenge feels so wrong, but it should definitely be testen!
i mean yea its a good test but i am excited for big dams with gates at the bottom or at top for accesibality
its also great for maps like the spiral as you could make an aquaduct to divert the badtide without contaminating the revivors that are closer to the source pool
"Beavers are fish so they can obviously breathe water."
t. medieval monks
Knock knock Peta open up your cruel treatment of animals have been noted.
Given that beavers could originally drown in water that was too deep, and the community responded with "what the hell is this?! they're beavers why can't they swim?!" I have to assume that beavers cannot die from prolonged water exposure (at least, no more than secondary to other problems like starvation)
It's surprisingly hard to find a game that lets you make real aqueducts. I know architecture is forbidden, but I love the look of flowing water. I love arbitrary water features that exist merely to look good. And I want a game that lets me make these things. Ideally inbedded in a city builder type game.
I can spend hours in city skylines, rerouting rivers, equalizing flow rate across a network of channels, setting up the perfect stage for a city. Really, the city is an afterthought. Landscaping is the real meat of the game IMO.
same. I can now have a sky river dropping waterfalls onto power wheels into another sky river!
I think we should give water features a pass as engineering demonstrations, like a bridge that has no practical purpuse except to demonstrate engineering principles.
Engineering isn't about purpose but function. Its engineering is the column is structural!
Update Note:
Water Wheels no longer slow the water flow. We originally introduced this mechanic to prevent players from building a perpetuum mobile, but the side effect - water wheels that would sometimes just stop working - was causing way too much confusion.
Are you one of the devs or just sharing the info
@@halflife2funjust sharing the patch notes
Kind of dumb since you still can do perpetual energy machines with poo, also in the game you don't really need that much energy anyway so it's kinda pointless.
That happened to me a few times. Had multiple wheels in one river, but the flow just went around one of them and it stopped working.
You need to use the new sluice building piece instead of the floodgate.
And he didn't seem to notice that you could control how strong each individual water source was.
He's gonna face palm so hard when he discovers them lol.
@@ranzeraRight? I think he purposely skips the dev notes for the specific reason of making his videos more entertaining. The whole ‘Incident’ could have been avoided had he read any of the details about the last update.
@@navad108but we don't speak about the "incident" lol
You can see the water current in cm/s with the water gauge
On one of the earliest episodes he tried to use the water gauge, didn't understand it, and then spent the entire rest of the series wishing he could measure the speed of the water!
Water wheels are about to get a huge buff
Water Wheels no longer slow the water flow.
"I have the Water wheel power!"
I was thinking dont water wheels confirm that the water does change speed, but then he goes on to figure out how to get the fastest water. RCE can make some normally boring videos actually interesting 😊
99% certain that what you're looking at is cell-by-cell physics i.e. each cell (size of a levee) can interact only with neighbouring cells. That gives a decent approximation of physics, but pressure etc. aren't really handled.
I don't want to say it's easy because it probably isn't but if a cell of water has its directlion to the side or head on with another cell. They could just decrease the velocity of the affected cell. Even better they could also average the direction. I imagine this would simulate turbulence and when applied in this valve shown in the video it would be effective at decreasing the energy in the system enough to have little water flow out the end.
@@Grymyrk It's not that easy, unfortunately. In the model I'm thinking of, adjacent cells try their best to make the water height level, subject to gravity. Importantly, this means that effectively water is teleported between adjacent cells. That means there isn't much to calculate velocity on, and in any case it's capped at 1 cell per logic tick (because water only interacts with adjacent cells), which isn't how velocity works.
Not knocking the approach in general though - I've used it, and it's a pretty good approximation at a macro scale.
Same way Stationeers have been doing it with atmosphere for like 10 years now, I assume ONI and a bunch of other fluid/gas management games does it the same way.
yes i think that's how i would try and implement this as well. the most basic version is like you said; that each cell only has one value for how full of water it is, and that value "trickles" along the cells to each adjacent cell on each update loop. a more advanced version might give each cell a few additional variables like the rate-of-fill and from which direction its being filled and in which direction it's being emptied... thus each cell has a velocity vector... ... but i'm not sure ... anyway its really cool!
edit: i *think* in the system i mention above there should be some emergent behaviours, some might be good such as laminar flow having a better flow-rate... but also possibly some bad side-effects like flowing around a 90 degree corner might kill the velocity too much... its hard to visualise without actually doing it and running tests
I wonder if we can apply calculus to voxel based physics models to get our every-day equations. Infinitely good approximations are no different than the real deal.
Doesn't notice the sluices. Doesn't notice that you CAN, in fact, do area building by click & hold and drawing your rectangle.
Poo sources produce vertically from 9 vertical faces, regular water is a permeable block with a source inside so it leaks from 5 faces each. So if you restrict 4 sides (bottom, left, right, back) then 6 of them can only produce vertically while 3 of them can produce in 2 directions (up and forward). This accommodates for the difference. Poo isn't more slippery, it just doesn't have the same output type.
You can also now use the waterfall for the waterwheels instead of it being an illusion.
Those sluices are so great for the testing in the second half of the video.
You have flow meter in timberborn. You should have used those. But still loved this video!
Matt glossing over the fact that you can use all the water in a damn now instead of just the top 3 layers is a big architect move.
i just tried today and i can't build "holes" on dams, do anyone knows if he's using any mod?
@@Frikiadicto Can you do a platform with levees on top?
There is pressure with the water. If you're curious, Skye Storm has some great videos on Timberborn engineering, and in many of his dam constructions goes over how to manage water pressure (super useful if you're trying to get better results from waterwheels to generate more power). The gist of it, though, is to take larger streams and force them into narrower streams
Skye Storm discovered the way Timberborne simulated Bernoulli, but there wasn't "3d (volume) pressure" in the game, it was still converting water source power/spread direction. A water source of 1 could spread in 4 directions, generating 4 0.25 cms vectors in those directions.
What Skye did that was super interesting was closer to beam forming, he artificially limited the ways water could flow with converging step drops (since water could only calculate in 2d) and thus created virtual pressure by forcing all the cms into a single channel even though it should have overflowed.
Again, NOT due to "water pressure" but "vector pressure". That said, his method of creating a turbo channel made for some amazing water wheel power generation!
So impressed that the devs changed the water, it's such a big undertaking
as I previously mentioned on your other Tesla valve experiments, another factor you failed to duplicate is reverse channel widths. if you look at your drawing again they're like half the width of the forward flowing channel.
second to properly test the concept, you need to try one with bad water at one end, and regular water at the other end and start them at the same time rather than racing them down an empty channel with no opposing pressure in the system. if you build the channels correctly, then the reversing channels should help support the flow direction of the forward channel which should prevent reverse flow from occurring
additionally, to compensate for the the lack of fine detail with which you can build, you're going to have to design it bigger (effectively increase the resolution) and you might have to tweak the orientation to the grid so that the channel intended for the flow to follow lines up efficiently to the grid where as the flow reversing channels, plus you might have to tweak the design to recreate the spirit of the idea within the limitations of this system
From what I understand, pressure build-up/difference due to varying channel width is also an important factor, one he did not test in the valve or separately.
Can you put a power wheel at the bottom of a waterfall and power it with just the falling water?
Exactly what I came for to comment.
I’ve been wondering that
czcams.com/video/kMoevpEjUaA/video.html&lc=Ugykz3tKyLwakDC8j1p4AaABAg seems to think you can, I haven't tested this yet though.
6:50 "I like poo water." -Real Civil Engineer, 2024
I was going to say that
I think any one that has watched his city skylines videos knows how obsessed he is with waste water.
Hey Matt, I’m soo into Timberborne because of you. I bought it 2 days ago and I haven’t slept in 2 days lmao. I even started streaming it on twitch. If I’m going to play might as well stream XD. I’m so glad to have been watching you for almost 2-2 1/2 years and seen how far you’ve come. Huge influence. Thank you for all the amazing videos you do and all the entertainment you’ve given me!
I believe during the pressure test, the water wheel HP difference was because the top one was only running on the water source generator, while the lower one was running on the water source generator AND the water storage above the floodgate. So the very beginning of those water wheels test, the storage would both need to be full, and the HP measured before the storage of water above the higher flood gate gets below the top of it.
Matt: With Levees you can only do straight lines, not areas
Me: Not anymore. Someone needs to read the changelog
Notlike Matt reads changelogs, but the log he showed was older than that change.
I'm sad RCE didn't test the power output of sticking a water wheel beneath a waterfall, to see if pouring it from higher generates more power.
Fun fact: poo water always runs faster when you don’t have a paddle.
Near 12:30 Matt almost turned into Mandark from Dexter's Lab... xD
Wow, this update is huge!! That solves a few of my only gripes with this game and allows for some even crazier, cooler builds! Thanks for all the experimentation, RCE!
6:52 The fastest slope is probably akin to brachistochrone: steep in the beginning, smoothly transitioning to flat.
The one Bernoulli discovered and Spider-Man uses
@@duncanchin How does Spiderman use it?
Dev: do I bloody hell need to go back to uni for fluid physics for real?
You can change the settings on the water sources to speed them up/slow them down etc....
22:25 Matt is so transcended with the strongest shape, he makes it completely on accident and so detailed as well XD
The fastest slope would be a brachistochrone curve; would love to see you engineer one!
Been a while since we've had a test video. These are always fun to watch
During the aqueduct / pipe pressure test, you could do with maybe lowering the bad water sources one level. That way half their output wouldn't immediately flow off the back. Instead that would only happen when the tank was full. Then you'd have actual pressure instead of the lower-gated tank draining itself out and not allowing for pressure, since you need a head of water for pressure to happen.
i think that to effectively simulate the path of least resistance in timberborn you would have to vary the channel width to vary resistance. since the game doesn't seem to process inertia and directional acceleration well, it just flows into available blocks rather than continuing its previous path of motion.
Mat i advice you to change the titel to something with the update like "testing the new timberborners water update"
You should make a perpetual motion machine so you can have power wheels turning during a drought!
Oh cool! I was hoping he was going to do this!
I just got into the Experimental Mode since this update came out. This update added so much fun to the game!
I love the sluices and water updates; not to belittle the other changes. Those two changes completed changed my dam designs.
Having a completely automated water system is AWESOME! It has been more satisfying than most things in the game.
You have to have a 1-deep pool for your sluices to work well. Which made having the 500HP pump worth something. So, now you can have a system of 'locks' with the sluices' that allows you to manage your entire water system in a way that you'll always have the amount of water you want where you want.
AND! a new update was snuck in, one that c hanged the power-shafts. The 'step up' power shaft has been removed and broken down into pieces with a vertical shaft introduced - not game changing but made power easier to manage.
*Whoa, I didn't realize we could put holes on the bottom of dams now! I gotta do another settlement now!
Pretty sure the game doesn't model water with fullfluid dynamics, it's just got a measure of how much water is in each place.
That's why the water flows faster down hill(all fluids spread evenly to the surrounding tiles), then slows down the instant it reaches the end(it no longer spreads preferentially forwards), as well as why the waterfall is slowest(it only has one tile with preferential spread vs the others having multiple tiles).
Very cool! Always fun to watch you geek out on the inner workings of water in games!
oh wow. I expressed my disappointment about water physics and maximum dam depth not being adressed after 2 years of development, and two weeks later, exactly that is solved and working amazingly. Hats off, devs! But it means I have to stop being a time-managing adult and give Timberborn another go!
These are awesome changes. thank you for bringing attention to this and testing it so thoroughly
hay mat, not sure if this is till a thing but, when you have a "waterfall" the flow rate over that water fall used to get capped at about 1 blocks worth of flow rate, so for your open vs pipe test, as far as i can tell, the flow rate would end up being the same for both of them roughly, at the end of your tests just add a big flat area for the water to spread out, and you should see the water flowing much faster
it's just a weird physics bug in the game to create mega dams that had a consistent output flow rate regardless of the water level
6:54
Matt sometimes just forgets that he's measuring the fastest slope and not the slope + flat part.
He should've just excluded the waterfall in the testing and dug deep enough on the 2nd slope to where the first slope ends.
Matt happens, I guess. We know him well enough for that.
I've always loved playing with water in Dwarf fortress. Making pools, waterfall and such while trying to not drown the whole fortress.
Dwarf fortress does have pressure and way to regulate it.
I had a almost Pavlovian response when seeing the title 😂 my brain immediately went "Wel-come baaack to Tim-ber-borners!"
I love these types of exploratory videos you do
Thank you for covering this! :D
FINALLY! congrats RCE for actually making a quality video!
If you don't think his other videos are of quality, why do you keep watching?
Every time I now load up Timberborn (sometimes several times a day for quick sessions) I have to sing "Welcome back to Tim-ber-Borners...".
Thanks for that.
Hey, there, Matt! Can I just say that despite having no knowledge of fluid dynamics/physics, your joy at describing them is still infectious!
Im pretty sure we were already familiar with how well "Shit flows down hill" 😂
gonna keep the vid on hand when i start playing timberborn again
thanks for the science experiments
Excited to see you play all the updates
Thank you for testing and sharing your knob.
Anyone else love hearing Matt getting excited over fluid dynamics in a game?
The amount of enthusiasm for liquid mechanics and water pressure from this man.
No wonder he's passionate about bridges.
If you really want to test pressure, you should create two tall reservoirs, on of which with a water source, and connect them with a pipe at the bottom. They should both fill up equally, but I suspect the second one will only fill up to the top of the pipe
Don't have this yet but as with so much of my game collection, I see RCE playing it and it magically ends up in my library.
Hahah, I just wachted the city skyline tesla valve video 😂😂
Did REC forget about stream gauges? Don't they tell you flow rate and such?>
Think that may have been a mod. Matt not be supported in this update
I (and i bet the timberborn devs) really appreciate that anecdote about how much fluid modelling software costs. It can be too easy to criticise when you don't understand how big a job that is!! Puts things into perspective.
I almost started an excel sheet to track food per day per beaver over my own water experimenting i was gonna do and here he goes and does the water stuff for me XD
very nice! ive been waiting for someone to experiment with the new physics! would love to see maybe a reservoir with a hole at the bottom into a pipe, and see how much verticality you can get on the other side? or so many other experiments. can you still fold water together with cleverly placed blocks to get crazy water wheel speeds?
The "slower" water is because the one that matched the bad water source was actually the 3 clean water sources at the front of the 3x3, not the middle ones. By the end of the first test, the 3 in the middle had caught up, but the 3 in front was always in step with the bad water source.
Well, all I can say is "Get ready for The Incident 3.0" 😂
Where the editor mirrored the tesla valve I kinda imagined that the editor would tell Matt to stop rambling 😂
I like imagining big engineering firms using Timberborn to simulate their liquid now.
Why not just set a bunch of water level testers and look at the height at various locations to get a crude approximation of where it’s pushing the water. Also I don’t think the visuals are synchronized to its flow rate. So anything visually in the water is affected by the physics but it being under pressure or faster won’t change the rate it flows. This is also why the steeper steps seemed ahead of the shallower steps each step down gives the visual a constant boost (probably just like 1.5*ground_speed*blocks_moved or whatever) so since they covered the same number of steps and the same number of ground blocks they arrive at the same time.
I don't think the speed with the pipes has to do with pressure but the simulated speed at which the water is falling to the floodgates. The higher one doesn't have to fall far from the source to flow out meaning it isn't gaining speed. I don't think pressure is simulated
ooh I'm definitly looking forward to future Timerborners seasons, just imagine the engineering projects that Matt will do!
Finally an Engineering video from a Real Civil Engineer!!!
Oh hell yea, this is the best. I love testing game physics almost more then playing many games lol
This is my favorite kind of video. Feels like a Special Myth buster video
A crazy suggestion: since levee can be build over the platform, try to build leveled aqueduct for the water wheel. So what i imagine is at the botom level there's some water wheel and raise the level using platform and overhang platform and put water wheel on top so after the waterr flow down you can get another power generator, or maybe build the water wheel below the waterfall.
Do water wheels work underneath a waterfall - with the water falling directly down, maybe producing more power if the aquaduct is higher?
didn't expect an episode of timberborn without any beavers can be that interesting 🦫
I am in the airport now. Can't wait to get home and turn on Timberborners.
There's a water measurment tool available in the game that give depth and water speed. No need to guess on this stuff. You can get the exact water flow speeds for each scenario.
New Timberborn content! Unfortunately my PC can't run it in double-digit FPS, so I must wait for my bank account to help me get a new PC. Fortunately I can always watch Matt play, and enjoy other games he's introduced me to as not all those games he plays are so CPU-intensive.
Thing I like about aquaducts? I can do my dwarf fortress showers.
Wow I expected you to need to make another video on this to get it right!
You and the guy who makes creeper world would get along very well.
Real engineering on this channel?
I watched another channel cover all the new features but he wasn't a *real engineer* so I was lacking this kind of advanced sciencing that shows how impressive the simulation really is!
You did steep slope and shallow slope.
Now do brachistochrone slope!
This is fascinating, i love these scientific game tests
I can not wait to see if I can get a bell syphon to work for storage!
Oooo, woot. Wish you released this not on a weekend, but I'm in
Your tests before the Tesla valve worked based on gravity/atmospheric pressure forcing the water to move faster through the tunnel. If you put a lid on them so the poo can't escape out the top/back it should force all the fluid to then move through the tunnel, adding much more pressure.
Could you do a comparison of the covered turbulent" tesla versus the covered "free flow" tesla valve? It would be a better comparison of if it works or not
26:55 they've change how you can put down levies - you can now build them in two directions at once. Meaning you can make big squares of them.
The sluice alone improved my mood for several days.
I cant believe you forgot about stream gauges!
I didn't expect RCE to make video about beta features so fast!
What a great insightful video!
I am disappointed to not have heard once Freddy Mercury saying "under pressure", but I am impatient to see the upcoming madness Matt will unleash on those poor beavers.
yoooooo extra timberborn video!
You forgot the biggest test!! Enclosing water sources in a box so the water can ONLY flow through the 1x1 hole. See if that increases the pressure?
instead of blocking the hole using a floodgate, you should be using a sluice. They're designed to be built under walls like what you did with the platform.
I wanna see Matt put a wheel at the bottom of the waterfall to see if falling water is better
All this water testing, and not a single stream gauge in sight. Sounds like an architect move, if you ask me...🤔