So This is How Scientific Accuracy Dies; With 869,000 Subscribers
Like many people, I sometimes watch Youtube videos when I'm bored. Usually science and engineering videos, since they're interesting and allow me to learn useful things even while "wasting time". I do sometimes err into less useful topics; gaming, movies, etc. So it was unsurprising when Youtube recommended me a video by Generation Tech about the engineering of spaceships in Star Wars.
An interesting subject! So I clicked on the video, only to discover that it was... terrible. Not the aesthetics or production quality
If you like science you might find it fun to watch the video and see how many errors you can spot. Here are the ones I noticed:
0:59: "Cyclone Ciara pushed the jet stream to close to 200 mph. With that kind of tailwind the 787 Dreamliner I was on eclipsed speeds of 800 miles per hour, which is about 200 mph faster than the Dreamliner's usual top speed of 587 mph. They assured us that it was completely safe despite the fact that we were traveling faster than the speed of sound in a composite-body aircraft."
First, it's unclear what the relevance of the composite body is. The implication seems to be something like "composites are unsafe at high speeds", but that's not true; composites are relatively new to all airplanes (the Dreamliner being one of the first airliners to use them extensively), but they're also seeing increased use in fighter jets as well. The slow adoption in both areas is due to the general slowness of learning how to manufacture and maintain new materials, not anything about carbon fiber specifically being unsafe or otherwise unsuitable for high speed.
But more importantly, saying that the Dreamliner was going "faster than the speed of sound" in the jet stream is like putting a snail on a bullet train and being impressed by how fast the snail is going. Sure, it's traveling quickly relative to the ground, but all of the engineering challenges
He does address this by saying "apparently what we were doing was kind of like getting on one of those moving walkways at the airport", but it seems like he didn't actually understand what this means, since the safety implications are no different from normal flight; as far as the plane and the air around it are concerned, it is normal flight. He goes on to use his "supersonic" plane flight as a comparison to show that Star Wars fighters are so slow, but the reasons he gives for their slowness are all aerodynamics reasons, and thus a Star Wars fighter would be just as capable of getting a boost from a jet stream.
4:50: "The speed of sound differs depending on your altitude and the temperature of the air. At sea level it's around 760 mph, and then when you go up higher in altitude to like 43,000 feet it slows down to 660 mph. That's because cold air is denser and so things move slower though it."
Allen is correct that there is a relation between cold things and the speed of sound, but the rest of the explanation does not make sense.
The air at 43,000 feet is less dense than the air at ground level, not more so. Gasses expand to fill the volume they're in, and the lower pressure at higher altitudes mean there's less force pushing them together, so they expand. This effect is much larger than the increased density that comes from the colder temperatures; air at 43,000 feet is only about 20% the density of air at ground level.
Denser fluids have more and heavier particles to bump into, making the friction higher, so "things" do, in some sense, move slower through them. But sound is not a physical object pushing through the air; sound is the air. Accordingly, the speed of sound does not decrease with density.
Now it is true that, if you replace the molecules with heavier molecules, this increases the density, and it also increases the speed of sound. Heavier molecules (at the same temperature) move slower, and since sound is just the molecules bumping into each other, the speed of the individual molecules has a large impact on the speed of sound. This is why your voice is higher-pitched if you breathe helium; helium atoms are lighter than air molecules, so it has a faster speed of sound.
But if you increase the density without changing the molecule size, by packing more molecules into the same amount of space (i.e. increasing the pressure), this has no effect on the speed of sound at all!
The density changes caused by temperature are the second type, not the first. Making air colder cannot somehow make the molecules bigger; it just makes them pack together more closely, since there's less bouncing around pushing them away from each other. So why does colder air have a slower speed of sound?
Because that's an inherent property of temperature, it has nothing to do with density at all. Recall that making the molecules bigger lowers the speed of sound because those molecules are moving slower. Making the molecules colder also makes them slower, so for the same reason, they transmit sound more slowly.
This one is, admittedly, a tough one. Googling the question will turn up myriad incorrect explanations how the speed of sound varies with the properties of the gas; many of which, funnily enough, contradict each other. Some sources claim that higher density increases the speed of sound, and other sources claim it decreases the speed of sound, when in fact neither of these is true.
And the reason why people think it decreases with density is because there are three simple ways to increase a gas's density: 1. replace it with a denser molecule, 2. make it colder, 3. lower the pressure. Two out of those three ways do lower the speed of sound, just for different reasons. So people notice the correlation and fail to realize that it doesn't hold true in general.
5:30: "When you approach the sound barrier, the air in front of your aircraft begins to compress at an increasing rate, and that increases drag on your entire airframe and makes acceleration harder and harder. There's actually a formula out there that states that the force required to push through atmosphere increases by the cube of velocity."
Here is the formula in question:
You may notice that the speed is squared, not cubed.
I think he got force confused with power; he goes on to state that a car doubling its speed needs 8 times the horsepower, which is true, since the power needed to push through air does scale with the cube of the speed.
But also, this formula is inapplicable to the sound barrier! Drag only scales as the square of the speed while at relatively low speeds, less than Mach 0.8 or so.
9:26: "The drag coefficient is basically a number that determines how much drag your airframe creates; the higher the number, the worse it is. Just for context, the average schoolbus has a drag coefficient of around 0.65, a Mazda Miata has a drag coefficient of about 0.38. F1 cars have a very high drag coefficient of anywhere from 0.7 to 1.4, because they need a lot of downforce for those sharp corners they have to take. The Boeing 747 has a drag coefficient of 0.03, and for modern fighters it's probably even lower, closer to 0.02. So guess how your favorite starfighters did in EC Henry's simulation. The T-65 B X-wing got a drag coefficient of 0.45. That's a worse drag coefficient than a Toyota Hilux."
No. The drag coefficient is a number that determines how much drag your airframe creates relative to some reference area.
An easy way to see why Allen's explanation must be wrong is that drag coefficients don't depend on the size of the object; a sphere has a drag coefficient of about 0.47 regardless of the size of the sphere; but obviously a larger sphere would require more force to push through the air, since it has to move more air out of the way.
For cars and simple shapes, drag coefficients are generally calculated as the ratio of drag to projected frontal area; i.e., its silhouette if you look at it directly from the front or back. Lowering a car's height does reduce its drag, and thus is a good idea if the goal is to save on gas or allow it to go faster with the same engine, but this will not necessarily decrease its drag coefficient, since its frontal area has also been lowered. Drag coefficients are thus a measure of how streamlined a car is given a constant height and width.
But airplanes have different priorities; if you want more space, you can always just make it longer. The focus is on lifting as much weight as possible, and lift depends on wing area. So drag coefficients for planes are calculated relative to their wing area, not their frontal area. This makes them completely incomparable to those for cars. If you were to calculate a drag coefficient for a Boeing 747 using the same method as is used for cars, it would be much higher than 0.03.
I don't know how EC Henry calculated drag coefficients for the Star Wars fighters, but at the beginning of his video he shows some simple shapes for comparison and gives their frontal drag coefficients. They seem to be in the same software he goes on to use for the star wars fighters, so I would guess that he's using the frontal method there too.
(And this is the more useful comparison, because Allen goes on to explain that Star Wars fighters don't rely on aerodynamic lift to stay up, they use "repulser lifts" (antigravity) instead. So their design goal for in-atmosphere flight can ignore lift and just focus on general drag reduction.
11:15: "Extra control surfaces for stability reduce drag."
No they don't. They're for stability. They increase drag.
Now, do these errors invalidate the whole video? Certainly not. The general gist is "going supersonic is hard, starfighters are designed primarily for space, so it wasn't worth compromising on space performance in order to make atmospheric performance better". This conclusion is still supported even after correcting all the physics mistakes.
But if the details don't matter, why include them? Because many people are interested in them! The simple explanation would be a 30-second video at best; people presumably are watching these videos because they want a deeper, more detailed, more technical explanation. So a defense along the lines of "this is pedantic, the details don't need to be fully accurate" is self-defeating; Generation Tech certainly thought they mattered enough to be worth making a video about.
Nor is this an isolated case of sloppiness. Generation Tech released a second video two days later, focusing on starfighter performance in space:
This one, somehow is even worse.
1:51: "The International Space Station is in low earth orbit. While there is some atmosphere creating drag at these altitudes, the space station still travels at around 17,400 miles per hour. And this is with very limited propulsion; it requires around just 25-50 pounds of thrust for this thing to stay in orbit. Which is amazing when you take a look at the specs on the SR-71 Blackbird. It could fly at 2193 mph, but to reach this speed it needs a thrust of 50,000 pounds. So the SR-71 could fly at 1/8 the speed of the ISS and requires 1000 times more thrust to do this."
The average drag force on the ISS is about 0.1 pounds, so that's how much continuous thrust it would require to stay in orbit.
In practice they don't run a thruster constantly, but rather let the ISS coast and decelerate for a while, then reboost it every few weeks. This often uses engines in the 25-50 pounds range, which is presumably where Allen got that number, but this is by no means required. They've also used the Cygnus rocket engine at 110 pounds of thrust, and even the two Zvezda engines at almost 1400 pounds of thrust, simply firing them for a shorter duration to get the same resulting increase in speed.
Allen's statement here is like observing that a car is driving at 25 mph with a 150 horsepower engine, and concluding from this that "you require 150 horsepower to maintain 25 mph". No. You can use 150 horsepower to maintain 25 mph, but the minimum power actually needed for this is far less.
Accordingly, his final number is off by nearly two orders of magnitude. The SR-71 actually needs 50,000 times as much thrust as the ISS to stay up, not 1000.
3:21: "So why can the ISS fly with so little effort at such high speeds in low orbit? Well this has a lot to do with orbital mechanics."
The limiting factor for speed on Earth is running into stuff, mostly air. The faster you want to go, the more heat and stress this puts on your vehicle, and the more thrust you need to overcome the friction. The ISS is in (almost) a vacuum, so encounters (almost) no friction. The maximum speed it could get to just depends on how much fuel it has to accelerate with; once going at a certain speed, it takes practically no fuel to continue moving at that speed, no matter how fast it is.
This has nothing to do with orbital mechanics; it's just friction, or the lack of it. Neuton's first law: an object in motion stays in motion unless acted upon by some force. No external force, no reduction in speed. This is equally true on Earth without any need to consider orbital mechanics, such as in vacuum trains that can go much faster than regular trains without needing anywhere near the same power requirements as a supersonic airplane.
Now it is true that the ISS being in orbit allows it to avoid lift-induced drag. Any airplane has to push air downwards to stay up, which comes with a drag penalty. But at high speeds this is overwhelmed by other forms of drag, so only around 10% of the SR-71's thrust went towards keeping it aloft; the rest was from air friction.
3:21: "If you travel fast enough, you can eventually reach what's known as escape velocity and escape the influence of gravity of that specific planet. Notice I say velocity and not distance, and that is because you need that velocity to generate force to counteract the gravitational pull force."
Well first off, the concept is escape speed, not escape velocity. It has no directional component. I can't be too mad about this one because even the Wikipedia article is titled "Escape Velocity" and then corrects itself down in the body of the article that it's actually a speed. For some reason "escape velocity" became the slightly more common term (though "escape speed" is also quite common), and even scientists who know perfectly well that it's not a velocity are too conformist to switch to the correct description.
But ok, moving on to more important issues: Velocity does not create a force. It's actually the exact opposite: if you're moving at a constant velocity, this means there is no force acting on you.
If you start out slowly coasting away from a planet, gravity will slow you down over time and eventually pull you back. But gravity gets weaker the further away you go, so there's some speed at which you'll never get pulled back. This is escape speed. It has nothing to do with "counteracting gravity", and in fact it's clearly not doing so; if you're traveling at escape speed away from a planet, you'll be constantly slowing down the whole time, because gravity is pulling back on you and there is not any force counteracting this pulling you forwards. But the strength of that gravitational pull will keep getting weaker and weaker as you get further and further away, so as long as you started out fast enough, you'll never reach 0 speed, and thus never stop moving away from the planet.
This also means that you do not "escape the influence of gravity". Gravity is always acting on you, no matter what speed you're moving at.
4:55: "Escape speed, 25,000 mph, is only necessary if you want to fly in a direct line away from Earth. In you want to get into low earth orbit, a much lower speed like 17,000 mph is ok. That's the speed that the International Space Station is traveling at. This allows you to distance yourself away from the planet significantly without actually escaping. The ISS is about 250 miles above the surface of the planet and at that range gravity's pull isn't nearly as strong. Because you aren't traveling fast enough to just escape, and you're traveling fast enough to get outside of gravity's influence where it's not too strong, you basically start orbiting."
The strength of gravity at 250 miles up is about 89% of its strength on the surface. Noticeable, but not really a significant difference.
Allen seems to keep confusing speed with distance, apparently thinking that getting to a certain speed makes gravity weaker or something, but as explained above, this is not true. It you're 250 miles up, it doesn't matter if you're stationary, moving at 17,000 mph, or moving at 1 million mph, Earth's gravity will act on you with the exact same strength in all three cases. It is only distance that determines the strength of gravity, not speed.
Regardless, you do not need weaker gravity in order to orbit. A spacecraft could even orbit at a height where gravity is much stronger, such as in a low orbit around Jupiter. The reason orbits around Earth are only possible above 100km or so is not because gravity is slightly weaker up there, it's because any lower and the air friction is too strong. Your craft will burn up or slow down too much and hit the ground before it even makes it around once. If the atmosphere (and all the mountains) were to disappear, the ISS could orbit just a few feet above the trees with no issue whatsoever.
Similarly, it is not accurate when he says that having a certain speed is what allows you to get to a certain altitude. Anyone who has used an old elevator is familiar with the fact that there's no lower limit on the speed with which your altitude can increase. Indeed, modern rockets that are trying to put stuff in orbit will generally get to around the needed altitude while going much slower than 17,000 mph, and then accelerate sideways to get the needed speed after already being at that altitude.
The only relevance of speed is that you need a certain speed if you want to stay at that altitude without continuing to expend fuel. That is, you need a certain speed to stay in orbit.
This works just like one of those coin funnel things that trick you into thinking you're playing a cool physics game but actually they're taking your money. Gravity is always pulling down on the coin, but since it's also moving to the side, it just ends up going around the hole rather than falling in. (The coin eventually falls in because friction slows it down, just like the atmosphere does if you're orbiting too low. With no friction it would just keep going around forever.)
For any given distance from the center, the coin travels at a specific speed. If you were to briefly touch a coin and slow it down, it'll fall down to a lower point to continue circling. And if you throw a coin in too fast, it'll just keep going in almost a straight line and jump out of the funnel. (If the funnel were infinitely large, it would just end up in a higher orbit.)
As the Hitchhiker's Guide to the Galaxy puts it, "There is an art to flying, or rather a knack. The knack lies in learning how to throw yourself at the ground and miss." The way to do this is to just pick a sideways speed that means you've moved off to the side by a distance of the orbit's radius by the time gravity has pulled you to the center of the planet.
5:57: "Now because gravitational pull exists around all planetary bodies, we can even use other planets to slingshot you around, increasing your velocity at a very high rate. It's like when you're in the playground as a kid, you run at one of the poles and swing your arm out and hold on to the pole as you run by, it will swing you around and it will increase your trajectory quite a lot. It's the same principle. So there is a sweet spot when you're approaching a planet that will allow you to essentially use the force of the planet pulling you down to slingshot. [Shows with his arms a spaceship moving towards a globe slowly, then going around it and moving away much faster.]"
I mean, you can just try the playground thing and you'll immediately notice that your speed does not magically increase. Grabbing a pole with your arm just means you start going in a circle instead of a straight line.
Orbit is no different. If a spaceship approaches a planet at some speed, it will leave the planet at the exact same speed, no matter how close it got.
The real effect that's called the "slingshot effect" is totally different. The details aren't important here, but basically, if I'm in space and Mars is about to pass by me at 13,000 mph, I can just move myself a little closer to where Mars is going to pass, and Mars's gravity will pull me along with it, so now I'm traveling at 13,000 mph too. This means I don't have to waste energy accelerating myself if I wanted to go in that direction anyway.
7:48: "In our world, escape velocity is the only way to escape the gravity well of a planet and enter vacuum."
kAlready mostly covered above. A) gravity is not the same thing as an atmosphere; getting into the vacuum of space does not mean you're no longer in a planet's gravity well. This is in fact the entire reason why orbits are possible; in order to orbit you have to still be inside a planet's gravity so that there's some force pulling you in a circle around it, but you have to be outside the atmosphere so there isn't any friction.
And B) escape speed has nothing to do with ability to gain altitude. A rocket with enough fuel could climb at just 1 mile an hour and enter vacuum a few dozen hours later.
9:36: "This also means that starfighters don't actually have to use orbital mechanics to fly around when battling above a planet; they can indeed dogfight, a.k.a. change directions rapidly. When you really can't do when you're subjected to orbital mechanics, because you're flying at such fast speeds; it just takes a long time to decelerate and accelerate."
This is one of those statements that seems so fundamentally confused about the subject at hand, it's not even really "wrong", it's just nonsensical, and accordingly is kind of hard to respond to.
The purpose of using fast speeds in a dogfight is to avoid getting hit, and to hit the enemy. A faster and more nimble plane is more able to dodge bullets and outpace attackers, along with being better able to catch up to a fleeing enemy, or maneuver around them to a better attacking position. All of this requires you to be moving quickly relative to that enemy. Your speed relative to a planet a few hundred km away is completely irrelevant.
Many battles in Star Wars focus on destroying a particular large ship, like a Star Destroyer or the Death Star. In Star Wars lore these ships aren't orbiting, they're just stationary relative to the planet and using their antigravity to stay up. But in Allen's hypothetical where repulsorlifts didn't exist, these large ships would presumably just be orbiting the planet. So if the rebels want to destroy one and need to approach close by to do so, they'd want to enter the same orbit, and be mostly stationary relative to the target ship.
Then the battle could play out very similarly to how it does in the movies. The defender starts shooting, the attacker dodges the shots but can't go too fast because then they'd be leaving the vicinity of their target, etc. All of the rapid acceleration is happening on a small scale relative to the other ships in that area, so the fact that they're all currently in the same orbit around a planet has nothing to do with their movements. Calling this unrealistic because they'd be going too fast is like seeing two people swordfighting on a train that's traveling 200mph and going "this swordfight is so unrealistic, people can't swing swords at 200mph".
This is a really odd misunderstanding, since previously at 4:28 Allen (correctly!) explains that speed is relative and there's no such thing as "stationary" in general, only stationary relative to some other object. But he seems to have forgotten this by the end of the video.
What's saddening about all of these mistakes is that they would have been so easily avoided. These aren't complicated technical points that require a life's study to understand; these are, by and large, very basic aspects of aerodynamics and orbital mechanics than the average beginner to the field would have to learn before they could get anywhere at all.
In many cases a skim of the relevant Wikipedia page would have told Allen all he needed to know. For example Wikipedia's page on drag coefficients has a whole section that addresses the different definitions and reference areas, and clearly cautions readers (in bold text!) that the numbers for airplanes and cars are not measuring the same thing and not directly comparable.
Frankly, I cannot see how a well-meaning amateur could make such a large quantity and severity of mistakes. To be clear: there's absolutely a place for citizen educators; the internet provides a vast trove of information at our fingertips, making it easier than ever for a smart and dedicated individual to learn the basics of a field of study in just a few days or weeks. But they have to actually, you know, try.
It's ironic that Allen mentions Kerbal Space Program as a good educational resource for spaceflight, because he's absolutely right. The Youtuber Scott Manley got started out doing Kerbal Space Program videos, before transitioning into talking about orbital mechanics and rocket design more generally. Now he's a legitimate aviation and rocketry expert making deep technical explainers, and rarely does anything Kerbal-related at all anymore. Even actual NASA engineers have positive things to say about KSP. Having played myself, I think it would be difficult for any serious player to have as poor an understanding of spaceflight as Allen has demonstrated while still being able to complete the game's missions.
So unfortunately, while I of course cannot read his mind and don't know what goes into producing Generation Tech behind the scenes, the available evidence seems to point towards Allen just not caring very much about accuracy in his videos. This is further supported by the fact that if you look at the comments on these videos, people have already pointed out many of the errors I described here. Yet despite getting dozens or hundreds of likes and being near the top of the comment section, Generation Tech never replied, nor issued any correction.
I held out hope that maybe he would at least learn from this and do better in the future. But a month later he's right back at it, confusing gravity with pressure and still not understanding relative motion.
Just to be sure I wasn't unfairly maligning the channel, I emailed them with my concerns, and asked if I was missing anything in my analysis here. They never replied.
So, what does this mean for their other videos? Any Generation Tech video that touches on science cannot be trusted, but most of their videos are pure in-universe discussions. Presumably they'd actually know what they're talking about there?
Well I'm not a Star Wars nerd myself, so I can't comment on their level of knowledge about the lore. I can however notice that many of the common-sense explanations they give for in-universe decisions... don't actually make sense.
Like here, where Allen says that making your ship vertical allows you to "stack weapons on top of each other without blocking each other", and "gives you overlapping fields of very deadly fire". Huh? The sight lines of your weapons are determined by their position on the outside of the ship, and the contours of that ship. But whether a ship is "vertical" or "horizontal" just depends on what direction you make your artificial gravity point. If the pictured external geometry is the ideal one for the pirate's desired weapon configuration, it's still entirely their choice how to arrange the floors on the inside.
I don't know how to interpret statements like this other than as attempts to fill the air with something that sounds smart and hope that no one is actually going to think about what he just said.
Maybe the lore details are fine. But I find it helpful to remember the existence of the Gell-Mann Amnesia effect. If someone is saying complete nonsense in a field where I have the knowledge to recognize it, this is evidence that they're likely to be saying nonsense in other fields too.
