SamuZai
Plum Parrot
Plum Parrot

patreon


Cyber Dream readers who enjoy math, please read :)

Recently a very helpful reader pointed out that my calculations seemed wonky when it came to ship travel times. I won't spoil for those of you who haven't read that far, but let's just say I might need to do some revising! My friend who initially helped me calculate travel times seems to have made an error, and he's not available at the moment, so I took my problem to ChatGPT. Would any of you care to check the AI's math? Here's our conversation:

Me: Let's calculate some fictional ship travel times. Assume the ship can accelerate at a constant 1G. Assume it's traveling from X to Y and the distance is 1.3 Billion Kilometers. Remember the ship has to decelerate for half the journey. How long will this take?

ChatGPT: To calculate the travel time of the fictional ship, we need to consider that it will accelerate for half the distance and decelerate for the other half. Let's break the problem into two parts: the time it takes to accelerate and the time it takes to decelerate.

First, let's convert the acceleration from 1G to m/s^2. 1G is approximately equal to 9.81 m/s^2.

The total distance is 1.3 billion kilometers, which is 1.3 * 10^9 * 10^3 = 1.3 * 10^12 meters. Half of this distance will be covered during acceleration, and the other half during deceleration, so each part will cover 0.65 * 10^12 meters.

Now, we can use the following kinematic equation to solve for time:

distance = 0.5 * acceleration * time^2

Let's solve for time during acceleration:

0.65 * 10^12 m = 0.5 * 9.81 m/s^2 * time^2

Rearrange to solve for time:

time^2 = (2 * 0.65 * 10^12 m) / 9.81 m/s^2 time^2 ≈ 1.32768 * 10^11 s^2 time ≈ √(1.32768 * 10^11 s^2) ≈ 364740 s

Since the ship takes the same amount of time to decelerate, the total travel time will be twice this value:

total time = 2 * 364740 s ≈ 729480 s

Now, let's convert this to days:

729480 s / 86400 s/day ≈ 8.44 days

So, the fictional ship's journey of 1.3 billion kilometers with a constant 1G acceleration and deceleration would take approximately 8.44 days.

Me: In that example how fast was the ship going before it turned to decelerate?

ChatGPT: To find the maximum velocity of the ship just before it starts decelerating, we can use the following kinematic equation:

v_final^2 = v_initial^2 + 2 * acceleration * distance

Since the ship starts from rest, v_initial = 0. The acceleration is 1G, which is 9.81 m/s^2. The distance covered during acceleration is half of the total distance, which is 0.65 * 10^12 meters.

Now, plug in the values:

v_final^2 = 0 + 2 * 9.81 m/s^2 * 0.65 * 10^12 m v_final^2 ≈ 1.2742 * 10^13 m^2/s^2 v_final ≈ √(1.2742 * 10^13 m^2/s^2) ≈ 3569300 m/s

So, the ship's maximum velocity just before it starts decelerating is approximately 3,569,300 m/s (or about 3.57 km/s).

Me: what percentage of c is that?

ChatGPT: To find the percentage of the speed of light (c) that the ship's maximum velocity represents, we can divide its maximum velocity by the speed of light and multiply by 100.

The speed of light (c) is approximately 299,792,458 m/s.

Using the ship's maximum velocity of 3,569,300 m/s:

Percentage of c = (3,569,300 m/s / 299,792,458 m/s) * 100 Percentage of c ≈ 0.0119 * 100 Percentage of c ≈ 1.19%

So, the ship's maximum velocity is about 1.19% of the speed of light.

Comments

The other thing I was going to write, and I'm sure you have considered this, is that ship speed is probably directly proportional to how much money someone is willing to spend. Military ships are probably 5x as fast as civilian ships.

Tom S

One neat thing about orbital mechanics is that they are relatively simple math problems (for the people in the story.) In most of the sci-fi stories that I've read, the traffic control and ship personnel are able to track most ships in space very accurately. I wouldn't be surprised if Julia can hack the space traffic control computer and get an instant location of honey's ship. Of course as the author, you can write ship tracking however you like. Maybe tracking ships more than a million miles away is basically impossible with their current technology. It would be neat if you threw out some technobabble that described the tracking limitations for the world you have created. Maybe ships are required to operate transponders.

Tom S

Great question! I went back and edited the chapter with Honey's message, but it's still not obvious which ship will get there sooner. I kind of want it that way - we'll see what happens ;)

Plum Parrot

Interesting time dilation only occurs the the last 5% of c and for having a difference that would allow you to “travel into the future” (thousands of years for mere years of travel time) is the last .x and .0x range iirc

Falxie

With the planets and anything else worth visiting in orbital paths around a gravitational body, doing a straight shot between two bodies will likely mean there is also quite a bit of lateral velocity (Think t-bone interception) difference that has to be accounted for, even with a constant acceleration (infinite delta-V) engine available. If you want your relative velocity to your target to be close to 0 when you arrive, the fastest path is unlikely to be straight between the two. Rather, it will integrate gravitational slingshot paths to line up with the orbital path of the target object, then either speed up or slow down in line with its orbital path to bring relative velocity to either local orbital velocity (Target is a planet/moon/gravitationally significant body, or 0 if docking with it (ship/station/landing on the ground). As someone suggested, take a few hours to play some Kerbal Space Program to learn and experience how orbital physics works, both with planet intercepts and with meeting up and docking with other ships in an orbit. Many elements are not intuitive, such as firing the engine away from the target you want to get closer to so that you can intercept it half an orbit later. Some of this can just be wand-waived into the navigational AI's calculations, but the lateral velocities do add a non-insignificant addition to the amount of dV needed to get from point A on one local vector to point B on a different local vector.

Dnulho

I saw you set limits for commercal and fast ships. I'm more curious about when acceleration couches are used and the ratios of the eventual travel time. Given that honey's ship supposedly didn't use acceleration couches, who will arrive in Saturn first?

Kddan

on a side note, .01 c is absolutely screaming, and i highly doubt any nonmilitary org would burn fuel nonstop for the entire travel time. It would become a mass vrs speed calculation. Glyn stewarts Castle series and mercs series play around with these numbers alot.

StressedTech

Your math checks out as far as pure numbers go, you would have to ask a physicist about how gravity and relativity would factor in. I stuck to pure math and probability for my degree.

StressedTech

Appreciate the feedback, but with H-3 reactors and drives, the tech is there. (in Juliet's world) Space travel was booming before the AIs slipped their leashes, and we fell into a terrible conflict (Cybergen-Takamoto war) - now the war's over for 50 years and things are ramping up again. We still have a lot of the AI-developed tech, though there's less understanding about it than we'd like. (future plot-points!)

Plum Parrot

Somebody already mentioned relativity so I won’t mention it, but from a narrative standpoint getting to 1% of the speed of light is absolutely insane for the technology of the world. I would consider maybe 0.1% for the maximum speed just based on lack of destinations that require burning fuel that fast and structural stability of spacecraft

heh

lol

Plum Parrot

😵‍💫🤕 I saw it... I read it... I followed the chore... yet my brain still hurts!

RonGAR

Thanks :)

Plum Parrot

A lot of more informed and smarter people have checked this before me. So I will just chime in and say I love that you are actually doing the math and science roughly behind this. That website that was mentioned is the right way to do it. Another way would be to have fun and do it in a moded version of kerbal space program as they do that sort of thing with a relative degree of accuracy. Anyways thanks for the great read Parrot and hope that you find what works for you to enjoy the writing.

Faelyn Kitsune

Thank you for this. Yeah, my goal with the maths of ship travel in this series is for them to sound plausible enough that someone who is really interested would need a calculator to confirm them. Then there's enough variability mixed in with planetary positions and ship specifications that such a person might say, "close enough." I just don't want something to be so wildly off that it will break immersion.

Plum Parrot

Looks good, but ChatGPT calculated the final speed way inefficient when considering starting from a standstill. Our acceleration is 9.81m/s^2, or otherwise said: 9,81m/s speed increase per second of acceleration. So we can just do 364740s * 9.81m/s^2 and get the result straight away. Though apart from relativistic concerns, we also have gravitational impact of the source and target celestial bodies. Earth will keep the ship "down" with 1G, so for the initial start we need a little more than 1G of acceleration to lift off. Conversely, eg. the moon has a lot less gravity, so the same "braking" acceleration will be too much. Why do I think of this? Randall Munroe of XKCD did a nice explanation of the gravity effects and limits in his "What If" Blog on the article "Earth-Moon Firepole". TL;DR: when calculating with a plain 1G we are not only ignorkng relativity but also the gravitational impact of the celestial bodies travelled between. I don't have a formula ready for that, so you might just use some hand-wavium.

Shad0wlife

Nice catch, but I think it did its calculation using the meters, so it worked.

Plum Parrot

As others have said if you are sticking with the very simple model of orbital travel then this looks fine apart from one error. "3,569,300 m/s (or about 3.57 km/s)" that should say about 3570 km/s

John Growcott

Yeah, this is the answer. Travel between orbital bodies isn't really as simple as linear kinematics, because you're traveling between bodies that are in motion relative to each other, etc etc. The distance between the Earth and Pluto varies a lot depending on where they are in their respective orbits, for instance. But assuming you picked a reasonable distance, the linear math at least lets you sanity check first. It might not be precisely right for any single trip, but it's easy and a good place to start

Nim

That's really neat :)

Plum Parrot

if you set the above mentioned calculator to 'Einstein's Universe' it actually does alls the relativity stuff for you FYI.

wellwish3r

With a more or less current understanding of physics the main limiter is reaction mass. As in an engine in space needs to push on someting to move a spaceship, and the only thing around to push on is mass you carry with you (expelled as exhaust) All long range space engines that try to remain rootet in our current understanding of physics either provide constant but extremely slow acceleration or an acceleration to cruising speed. So for in-system travel keeping the speeds as low as you plan to makes sense.

wellwish3r

This is cool! Thanks :) It roughly agrees with ChatGPT, too, hehe.

Plum Parrot

Some of that is directly correct if you were not involing the mass of planets and other orbital objects but just talking a void traveling from one point to another. When you start pulling in things like gravity and slingshots and other orbital mechanics so you get to the right space its best to go and use a program to figure all that out. https://nineplanets.org/tools/space-travel-calculator/

Findell

This is helpful, and yeah, to avoid relativistic issues, we'll be capping commercial ships at around 1% of c and fast ships at less than 10%. I'm explaining this as drive limitations (they need maintenance to keep up that kind of burn, etc.), navigation difficulties, fuel costs/needs, and crew comfort/cargo risk.

Plum Parrot

Hey, so the problem here is that you are not accounting for special relativity. Once you start talking about speeds that are best expressed in fractions of ‘c’ you are going to need to incorporate that into your math if you want it to be even close to correct. In particular once you move closer and closer to the speed of light 1G of acceleration will not actually increase your relative speed by 1G/s. If you want to have your ships work like this (constant 1G) and have it be accurate you will need to work with the Lorentz factor (https://en.wikipedia.org/wiki/Lorentz_factor). This will make your calculations fairly complex for every trip you plot. My suggestion would be to have ships only accelerate to a fraction of ‘c’ (like say 0.4c) and then cruise until it is time to decelerate. This would allow you to pre calculate the acceleration/deceleration phases of all ship travel (unless the ship accelerates faster). Calculating how long the cruise stage is then becomes a simple division. This could easily be explained by the cost of reaction mass for ships. Also you then need to the decide if the cruise stage is weightless or if there is another method of generating gravity (spin or even artificial gravity plating). If you have more questions I'm happy to help though I have to say I only took 3 physics classes in University 10 years ago. I'm not an expert in the field.

wellwish3r

You're right, but in Juliet's world, the early true AIs solved H-3 fusion, and now it's kind of ubiquitous.

Plum Parrot

I know, I was grinning the whole time about that.

Plum Parrot

I did! It's really minor at that speed. ChatGPT: With a gamma value of approximately 1.000034, time dilation is minimal.

Plum Parrot

Ask it how much time the occupants of the spaceship differs from the time of an observer on earth :)

Jerry Rapp

My math also agrees with ChatGPTs math. Also very glad you take the time to check your maths. Nothing annoys me more than reading sci-fi stories where the math is so obviously wrong you don’t even need to get out the calculator to see that.

Sekander

I love that we have an ai assisting in writing a story about ai

Hampus Steinvall

so my german „highschool“ education in physics and math says this is correct. a vage memory from some science video says that a constant acceleration of 1g is impossible in the near future because we dont have any technology right now to fuel that.

t3chn0fr34q


More Creators