Latest / In the Loop AI / SpaceX Starship Isn’t About Mars - It’s About Revolutionizing The Internet
Transcript
- Josh: [0:03] So just a few days ago spacex successfully launched their 10th test flight of starship which is the mars rocket that is supposedly going to get humanity to mars that's a very interesting and noble goal but there's a really interesting goal in the meantime in the intermediary before we get people on mars and that's the internet story so i kind of want to use this episode as an opportunity to share what spacex is working on and what i think a lot of people are overlooking, Josh: [0:27] which is the Starlink satellites that are coming off of these flights. So did you first, first things first, did you get an opportunity to watch this earlier this week? Because it was pretty incredible. Ejaaz: [0:36] My two interactions with the SpaceX rocket stuff, Josh, is this video that you're seeing on my screen right now, where this big metal canister that I think is about 30 stories high is getting caught by these claw-like machine things. And then And a video that was going viral this week where apparently one of these ships just blew up. Can you please explain what the hell is going on? Josh: [0:59] Yeah. Okay. So we're actually up to ship number 37 for Starship. And I believe almost every single one of them has blown up, which is a good thing. Because this is how SpaceX works. Oh, it's a good thing. It is. Yes. You quickly iterate. You destroy things. You learn. You gather data. Then you rebuild. Then you try again. So what you're seeing is a lot of rebuilds and retrying again. It turns out getting a skyscraper into low Earth orbit is a fairly challenging thing. What we're seeing here on your screen that you shared is actually the most exciting part of the Starship program up to date. So the way Starship works is there's actually two pieces. When we say Starship, normally people refer to both pieces stacked on top of each other, but it's actually two separate things. There's the booster, which sits on the bottom, and then there's a Starship that's on top that has nine engines or something like that. And the reason that we have two separate parts is because Earth's atmosphere, super dense, really hard to get out of. Josh: [1:46] We need the entire bottom half of the rocket just to get out of the atmosphere. Then the bottom part breaks back and it returns starship goes out into deep space so what we're seeing here is actually the sole reason why this program is able to exist which is these things called chopsticks or mechazilla or whatever you want to call it the proper technical term was actually able to catch the booster and this gets into a topic that i wanted to i've been wanting to talk about for a while and we never had an opportunity to which is the idiot index ejes have you ever heard of the idiot index before no. Ejaaz: [2:16] But i feel like i should be on it yeah Josh: [2:18] Okay i'm i'm excited to fill you in on the idiot index. So the idiot index essentially is the delta between the cost of the goods to create an item and the amount of money that you're able to sell the item for, or the amount of money that it costs to take the item to market. So for example, with a rocket ship, the old programs, say the Saturn V took maybe $50 million of materials to build the rocket, but the actual cost of the launch was billions. So that means the idiot index is quite high. It. Josh: [2:49] With Starship, the goal is to get that idiot index down to as low as possible to get to as close as the cost of the rocket as possible. So how do you do this? Well, the problem with these old rockets was that when you launch them into space, that was the one way mission, they're gone. They never really come back to Earth. And if they do, they're with parachutes, they land in the ocean, and everything gets destroyed. So your sunk cost is the billion dollars that it costs to build the rocket ship and launch it. With Starship and the SpaceX launch in general, the goal is for reusability of these rockets. So if you're able to reuse a rocket just twice that takes your idiot index say from 20 times to 10 times and then if you do it three times it brings it down even more and even more and the most recent falcon 9 rocket that they've been using actually just finished its 30th launch so what they've been able to do is 30 from one booster that means what used to cost a billion dollars for one is now a fraction of the cost and can now be used at least 30 times over it's still going so it's really impressive what they've been doing but that's on the smaller rocket the falcon So what we're talking about this episode is the Starship rocket, which is way more expensive, way bigger, way more complex. But that demo that you saw with the Mechazilla arms was actually able to catch it. And that proves the point like, hey, we can actually make a rocket that is rapidly reusable, which lowers that cost significantly. So we get closer to one on the Josh: [4:05] idiot index, meaning we can make these for cheaper. Ejaaz: [4:08] And yeah, that's that's kind of how that works. So I have a question, and probably one that deserves to be on the idiot index, which is, why do we need such a massive rocket? Like, what's Elon's purpose behind building something that's way bigger than the Falcon 9? Josh: [4:24] So the dream, the goal he's going to sell everyone on, the goal that he has been selling everyone on is that, hey, we need to get to Mars. And in order to get to Mars, I mean, he calls it the great filter, passing the great filter. One of the great filters that needs to be passed is making humanity multi-planetary. So in the case something happens on Earth, well, we have redundancy on Mars. But the problem with that is in order to have redundancy on a separate planet, you need to be fully self-sufficient in the case that the ship stopped coming from the home planet. And in order to be self-sufficient that means you need every single element that life requires so if you miss say you miss calcium for example and you don't have one element that's critical it ruins everything so that means you need about 100 million tons of payload on the planet to create this redundancy which means really you just want the biggest rocket ship possible to get the most amount of payload per launch out into space but this has interesting second order effects in the intermediary which we're going to get into a little bit later in the episode which is you could also just launch bigger satellites. And that's what's going to happen in the intermediary is now that we have a bigger ship, well, now you could launch much bigger satellites that actually make the services that we use on our home planet much, much better. Ejaaz: [5:32] Okay, so you just mentioned that the payload would need to be maybe like millions of pounds, theoretically. Can you like help me then understand what this prototype that Elon's talking about here? Like maybe I can pull up this infographic, Josh, and you can tell me a bit more about it. Josh: [5:47] Yeah, so what we're seeing here, there's V2, V3, V4. V4 is the ideal state that we're going to have while we go to Mars. V2 and V3 are kind of the intermediary steps to get there. The problem with V2 and V3 is they're still blowing up a lot. So we would love to have the V2 or V3 version working flawlessly, but they're being used to learn, basically. These are the intermediary steps to getting to the final form. Every single rocket launch, you'll notice it's a bit different, the outcome. So some launches, they're testing different things. Some launches, they're going into different levels of orbit. They're re-entering Earth's atmosphere at a different angle to make it heat up more and test the heat shield. So they're kind of doing all these different tests for each one. V4 that you see is going to be close to the final form that we're going to Mars with. And really what it is is it's just slightly bigger slightly more engines slightly more payload so the plan is to probably launch one or two more of these version twos move up to version three and then eventually get to version four okay and then in that sense we'll hopefully be around next year and we will have a clear shot at actually sending one of these to mars saying we as if it's if i'm involved but i feel like i feel like i'm invested shareholder yeah there you. Ejaaz: [6:54] Go well hang on a second. When you started this episode, you said that this Ejaaz: [6:59] was all to actually do with Starlink. Isn't that an internet satellite thing? What's this going to do with that? Josh: [7:06] Okay, this is the part that's cool because this is the part that actually affects us. If you don't care about going to Mars, if that's not your thing, that's fine. This is something that should be your thing. This is something you should care about. It's the Starlink network. Basically what the Starlink network is, it's a series of these satellites that exist in low Earth orbit. They orbit around the Earth, and they provide internet to you anywhere, anytime, anyplace. And what's happening here, we have the tweet pulled up right here from Sawyer Merritt. Josh: [7:31] You're starting to see the exponential growth of this network, right? Like you've kind of, you see a very slow trickle and then the line starts to go steeper. And as you watch this line grow, it's important to understand, oh yeah, here's the, okay, this is cool. So drag around the planet here. You could actually press play, I believe. And then, yeah, so you could see the way the satellites move around earth. This is the current starling constellation, which is incredible. Josh: [7:55] Like if you, if you had infrared vision and you looked up at the sky right now and you had binoculars and you could see everything flying around this is what it would look like there's just this unbelievable array of satellites rotating around earth at super high speed all talking to each other insane this is the currency of the network starlink satellite this is all starlink what you're currently seeing is all starlink satellites oh my god this is funny anomaly where you kind of see the top and the bottom are really missing a lot and it's because they're actually rotating around earth's orbit and they're trying to optimize for where the most people are and there isn't that many people there so for them to go vertically around earth as host a horizontally like latitudinally it's just better so they're going to work on getting the arctic and the the polar sides of the earth going but currently for now this is what the earth looks like so these v3 satellites that are launching now enabled through starship are really impressive because each one of these launches now for the starship is equivalent to 20 launches of the previous rocket which is the falcon 9 meaning each one of these is equivalent to 60 terabits per second of bandwidth, which is roughly equivalent to 2.4 million simultaneous 4K streams or 12 million 1080p streams, which is a tremendous amount of bandwidth per launch. Josh: [9:07] And previously, at the cadence that Falcon 9 was launching, it would take about three months to launch 20 ships to be the equivalent of SpaceX or Starship. So if you launch Starship at the same cadence as Falcon you're getting essentially a 400 times multiple on the throughput that is enabled via these satellites and what's even more interesting is that the latency of these satellites because they're much more powerful and they're in low earth orbit it actually drops to around 20 milliseconds now for people who aren't familiar with what latency is it's kind of the response time between your computer and the server you're interacting with and it's about how long the signal takes to go from one to another well 27 milliseconds is about the average this is below the average so if you are a gamer if you are a streamer if you're recording live like this the latency between the one connection to a computer and the other will actually be lower than a standard residential internet connection which is interesting because okay now that we're putting out terabits of bandwidth into the low earth orbit and it reaches we saw the chart all around the world anytime any place it's faster than residential internet it has lower latency and it's really you see a very clear path to Starlink actually that chart continuing up that exponential curve and going even higher because man as soon as this is available at that level for me I will totally want to be a user right, Same. Ejaaz: [10:29] I mean, I was just thinking about that simulation that we just showed, or rather not a simulation, but a live like kind of graphic feed of where all the satellites are. There are like millions of these things. So all I can think of is when I'm on a flight and I don't need to connect to shitty airline Wi-Fi anymore, and I can just connect to one of these satellites. Or if I'm walking around and I don't need to use like 4G or 5G, I can just connect to one of these satellites. That's pretty insane. Am I thinking about that correctly? Like I basically can bring out a PC gaming laptop in a park and run like, I don't know, World of Warcraft and there'll be like no latency. Is that a fair example? Josh: [11:07] Yeah, in a park in the middle of nowhere with no service at all. And because these new connections have direct to sell, which is it's called direct to sell. Basically, you can actually receive connections directly on your cell phone. So currently it's limited to t-mobile users if you are a t-mobile user and you are in the middle of nowhere you can actually still receive service from these starlink terminals and we looked at that map earlier and it was incredibly impressive but a really interesting fact is that one launch of a starship with these new v3 starlink terminals is equivalent to 13 of that entire network overnight in one single launch so if you do about eight of these you have essentially doubled the network. And I mean, these are launching basically every single week. Josh: [11:48] So the throughput is going to go through the roof as soon as this happens to work. Ejaaz: [11:53] Okay, so I have an obvious question, which is, how does any competitor survive this? You've got a madman launching satellites that are increasing the network by 13% with every launch, and he's reduced the cost of every launch, if what I'm understanding you're saying is correct. And he owns a space company that's doing that. Surely he wins right Josh: [12:15] It's really hard to compete it is super hard to compete because nobody really has the ability to do this and what you're seeing is i mean if we go back to that that post you had up about the flight 10 which is like what happened on flight 10 showing the live stream we could just kind of walk through what they did that is so impressive like essentially what they did is they. Josh: [12:33] They took this starship and they sent one part of it into low Earth orbit. They sent the other part and landed back in the ocean. Both landed right where they needed to be. They tested, we call it the PEZ dispenser, where the gate actually opens up. It shoots these Starlink terminals out of space. These are all really complicated problems that took 37 iterations and like almost 20 years of technology evolution to get there. And there's no company that's really close. I'd love to find the chart. There's a chart out there somewhere that shows the amount of satellites sent to low earth orbit from every country in the world up until SpaceX started. And then SpaceX is equivalent to every country in the world times like 10. And it's a really challenging thing because, I mean, believe it or not, rocket science is pretty freaking hard. And there really only is one company Josh: [13:17] so far that's figured it out at scale. And they still haven't, granted, we're talking as if, or I'm talking as if they figured it out. They have not. They haven't successfully deployed any of these satellites. Every one of these starships is still blown up. They haven't actually landed any of them but there's a very clear path of getting there which is what i think i'm personally super excited about is assuming they continue on this rate of progress there's only a few more test flights left until they have a functioning like ship that's actually capable of going into orbit and deploying these satellites i. Ejaaz: [13:46] Mean i mean that is just insane so so let's just play that scenario out and assume that they figure out how not to let their rockets crash and they consistently put out these satellites, increasing their network by 13% each time or whatever that statistic was. How can people get access to this, Josh, once this is live? So you mentioned like normal people like you and I, if we have T-Mobile network phone, we can get access to it. Can anyone else kind of like spin up or get access to a Starlink satellite if they were a non-T-Mobile user? Josh: [14:16] Yeah, you have to, you'd have to buy, and this is totally not a sponsor. I don't even own one of them, but if they'd like to, hey, please come along. But yeah, you, you basically, what you need to do is you need to buy a satellite dish. Like if you remember the old TV cable plans where you needed a satellite dish sitting on your roof, it's just like that. It looks a little more elegant, but you basically need a little satellite dish. They sell them in different sizes. And what the satellite dish does is it's just a receiver for the laser beams that these satellites are shooting down. And it will connect with a few of them. It'll kind of triangulate itself. And then that's how you get the connection. Ejaaz: [14:44] So you just kind of like attach this to your house or your apartment or windows? Josh: [14:48] Yes, exactly. Anything with a clear view of the sky. And the reason I don't have one is because it currently, like, it kind of sucks if you don't need it. It's really meant for rural areas that are not high density because the bandwidth of the network isn't very high right now. So I have a few friends that live upstate. They live on a farm. They all have Starlink connections because it's the best internet by far. And it's great. You get about 200 megabytes down, about 20 megabytes up. and and you're able to live a totally normal life with that but in densely populated areas there's just not enough bandwidth currently to support that so it doesn't make sense to have it so right now the current value of the network lies in these remote areas these these rural areas if you're going for a hike in the Josh: [15:28] middle of nowhere you want to send a text message in an emergency okay. Ejaaz: [15:31] So josh can you explain why this flight 10 was such a big deal like i understand that maybe it didn't explode at a specific time and it exploded maybe at the right time but can you maybe like give me a layer deeper like what exactly happened what was so Josh: [15:44] Amazing there's important terminology that you should i need you to know right now it's called a rud which stands for rapid unscheduled disassembly which is basically the professional term for the rocket exploded so a lot of these these launches result in a rud a rapid unscheduled disassembly this flight did not so in the past few, And it started losing. So as it was flying in low Earth orbit, it started going off kilter and then it kind of started flipping out of control and then it burned up in the atmosphere and it was terrible and it went all through the Atlantic Ocean and it was kind of a mess. The next time on the test platform, the rocket actually exploded. And not only did the rocket explode, but it blew up the entire launch site. So that set them back a month or two. Josh: [16:25] So there's been a lot of these problems of rockets blowing up when they're not supposed to. These rapid unscheduled disassemblies. This time, everything blew up on time, which was actually really good. So the expectation was everything was going to explode, but everything exploded just at the right time. And one of the most important parts that they've been trying to test for many test flights but haven't is this term called the PEZ dispenser. And that's what we're seeing on screen right now, which basically is this gigantic door, the bay door on the side of the ship. it opens up and it dispenses these gigantic version three robots or version three satellites. Josh: [16:56] And what we're seeing here is the first deployment of the satellites. And what happened last time is there was a pressure problem and they couldn't get the door open. So this is a big deal because this is the actual process required to get these new version three satellites that enable this global network of internet connectivity to work. And they demonstrated for the first time that it actually works. So the things that happened here is PES dispenser opened, successfully deployed the satellites, successfully brought the starship back to Earth, successfully brought the booster back to Earth. These were all good things because when you're building rockets, one of the hardest things, and one of the reasons why we haven't really had rapidly reusable rockets ever, is because our atmosphere is really freaking hot. And it's like super, super hot when you come in really, really fast. And it melts basically all the metal in the world. Like there's no metal that you can build, that we know of at least, that will survive the heat and friction of flying back into the Earth's atmosphere at hypersponic speeds. So what they did is they created this thing called a heat shield. And what they use is they actually, it's so funny. It's like, I feel like. Josh: [17:55] One of our moms could work this job where they're in a bakery and they're just baking these ceramic plates. And that's all it is, is they bake these series of ceramic plates that coats the bottom of the rocket. And it turns out ceramic is really good at being heat resistant. So what they've been doing is trying to test that heat resistance, test the heat shield by bringing the rocket back over and over again. And this time they actually held up for the first time as well. So these heat shields, these little ceramic tiles, they all coated the bottom of the rocket and they prevented the rocket from overheating and exploding too early so those are the things that sounds. Ejaaz: [18:26] Like the ship's ready to go what what is there left to figure Josh: [18:30] Out it's pretty good okay so in terms of what's coming next this is speculation but elon kind of confirmed some of it and we have a post about this it's from this guy named everyday astronaut he's kind of the number one spacex fanboy with the perks of getting invited to go to a lot of launches and talks to elon so he's speculating hey version 10 actually went really good but version 11 like we spoke about earlier is still the v2 rocket it's not that that new rocket that they really want to test so they kind of have this intermediary launch that they're going to try to test certain things which means that we're probably not going to get an actual starship catch until version 13 which is two launches from now and the starship catch is going to be what you saw with the mechazilla arms except for the upper stage not the lower stage so the goal two launches from now is they will land, 400 feet of rocket within about a 15-minute window, which is going to be incredible. First, the booster comes in, they take it, they grab it out of the sky, move it out of the way, place it on the ground, and then they catch the top part and they place it right back on top. And the idea is that you'll be able to eventually relaunch that back into space Josh: [19:32] and be on your way with your rapidly reusable Mars rocket. So it's going to be really exciting. Ejaaz: [19:39] Josh, when I think about everything that you've just told me, the image or rather the scene that comes into mind was Josh: [19:47] Many of the Star Trek movies. Ejaaz: [19:48] Particularly the recent one, where it starts off with the protagonist basically getting abducted into a spaceship, but it's set in the future where you have these massive structures which launch and catch rockets. I think of Iron Man, Tony Stark, where you have similar structures as well with his spaceships. And this is being built right in front of our eyes. And it almost doesn't seem real. Like everything that you've told me doesn't seem real until I watched like the live stream and these clips of these ships being caught by these massive mechazilla arms. And so I'm thinking about ships that you've mentioned, these big ships, I'm thinking about sending humanity to Mars. I'm thinking about satellites being dispersed at breakneck speed at extremely high temperatures, which seem impossible. Ejaaz: [20:33] Where do you think this like ends up? There's so many use cases for this beyond just the internet. And I bet you have like some crazy examples. Oh, ideas. Like, do you? Josh: [20:42] Yeah, it's really, first of all, it's really inspiring to see this stuff work. Like, I will never forget the feeling of not only seeing the most recent catch, because that happened back in October, and that's when it went super viral. But there's been so many of these milestones that seemed impossible, that when they do them, it's like, I can't believe that just happened. I mean, the first one was just landing that first rocket back on Earth. Then they landed two of them at once. Then they landed 30 in a row. then they landed a skyscraper and it's like every time you see them progress you're like there's surely there's no way they can do it and then they go and do it and it's it's really unbelievable and there's a lot of these these second order effects that you're hinting at so one is the internet i mean sure we have gigabit internet laser beaming down from earth onto us but there's also this really cool opportunity to to think about what it unlocks when you have these rapidly reusable rockets so i think the other day you sent a message about spacex in regards to their low earth orbit mission so kind of like we recently had on blake shull who does supersonic airplanes this is like supersonic on steroids where if you can actually create these rockets that are really safe that can go. Josh: [21:49] Into low earth orbit, then come back down. You could essentially traverse any part of the world in about half an hour. So to go from like New York to Tokyo, about 34 minutes to go from New York to London, about 15 minutes. And it creates this really bizarre world where, I mean, you can, things just, it feels futuristic. It feels like it's something out of Star Trek, right? Like you go to a launch pad in New York. Ejaaz: [22:13] It's that meme where it says the world if, and then they describe a problem and whatever. And then it's like this futuristic kind of world with like cars flying everywhere. Like what you just said, like transporting me from LA to Tokyo in 30 minutes is just insane. And I think like when stuff like that starts to launch, people will really start to believe that this space thing can be real, I guess is what you're saying. Josh: [22:36] Yeah, and there's this great post you're sharing from Nikita who is just like talking about, he spent the day at Starbase and he never really understood the scope of what's happening here. And I think this is a common conception amongst everyone is like, oh yeah, this guy's building rockets. But the idea of going to Mars seems so far off that it's almost, it's impossible to get excited about because it doesn't feel real until you go to Starbase and literally an entire city and you go see it for yourself. I mean, he's describing it right here. Yeah. The Manhattan Project. It's an entire city built for a singular mission of going to space. It's incredible. And these things take a long time it's been decades since they started it will be decades until they get to where they want to go but the fact that we're able to watch the progress in real time through these launch live streams it's it's amazing to see and it's one of those things that it's a bucket list item maybe we should do that you guys we'll put on our bucket list to go to starbase for a launch one day. Ejaaz: [23:29] I'm down to see one of Josh: [23:31] These yeah because it really it's inspiring to watch like over my computer i can't imagine what it must look like in person to see what's going to happen when these things really start to get going, the effect they're going to have. It's going to be amazing. Ejaaz: [23:43] It's, it's, I mean... The main takeaway from this last bit of this episode, Josh, is that it's just so inspiring and real. I just, I remember watching SpaceX's first live stream for the Falcon 9 catch. That was like a, whatever, an hour and a half long stream from start to finish. And they put together this like really cool clip of like everything that happened. And I think I watched that clip like coming on like 200 times. It was just so inspiring watching it for the first time and now it's just happening every day casually and they're taking it up a notch by you know whatever catching 7 000 tons at 300 stories high or 30 stories high feet of shuttle and then dispersing satellites it's just so cool and i guess i can't wait to see more of it as late as that sounds it's so cool to watch yeah it's Josh: [24:37] Amazing how how quick it's funny you mentioned that because my dad and i used to watch every single SpaceX launch, period. Every one that went up, we would watch the live stream. And then it got to a point where they were just launching so many of these rockets and every one of them was landing and it was just expected. I'm like, I can't keep up with all of these. And they just keep pushing the envelope forward. And now the new one is Starship. And I mean, again, they're doing the damn thing. They landed the booster. They caught it. Hopefully in two launches, they'll catch the top actual ship and they're just going to keep chugging along. And it seems like this impossible mission, but every single one of these test flights proves that it's not as they get closer and closer to realizing it. And it's exciting. It's like, it's definitely, it's something to wake up and be inspired about that. We are making progress in this really difficult thing we've never been able Josh: [25:22] to figure out in this really narrow window of time in which it's actually possible to do. So yeah, that has me feeling feeling really good and on top of just getting really fast internet and then eventually getting to go from new york to la in like 15 minutes it's also gonna be pretty cool to to have the plausible reality of of having life be multi-planetary and that or at least going sending something off into space to go explore and see what what's out there so yeah that's that's an exciting thing for for me to to end on too yeah and i think we'll probably wrap it there this episode had nothing to do with ai in fact this entire mission is probably one of the last great things humanity will accomplish without AI. Josh: [26:00] So shout out to the SpaceX engineers. They did all of this the hard way. There were no large language models in existence when this code started being built. We were also recording this episode. It is currently 11.20 p.m. Eastern Standard Time. I'm exhausted. So if we were stumbling over our words at all, it is because it is very late. But we really wanted to get you an episode for this week. And there wasn't a ton of super exciting AI news. I really love the SpaceX stuff. Ejaz seems like you're pretty stoked about the SpaceX stuff. It's really like why don't we get on and just try our best to record something about something we're interested in and hopefully it resonates with you hopefully it was informative if it if wasn't i'm sorry we're trying our best it's really but we wanted to get you an episode out there so if you enjoyed it please don't forget to to like subscribe share it with your friends who might really appreciate this it really goes a long way we are 47 now on the spotify leaderboards we are close to slipping out of top 50 so we got to put our foot on the gas we got to keep on going keep Keep sharing with your friends. But yeah, that's pretty much it. Thank you all so much for listening, for sticking with us till the very end of the episode. We really appreciate it, and we will see you. Music: [27:03] Music