Increase effect of mass on platform speed

[skaudus]

TL;DR

Platform top speed should have a higher dependency on mass to discourage "cigar" ships.

What?

Currently, mass has a disproportionately small effect on platform speed when compared to width. This encourages the use of "cigar" ships, which are extremely long and thin but potentially very heavy.

Weight 251 t, 18 tiles wide, 4 thrusters, 150 fueling rate

12-16-2024, 10-41-13.png (628.69 KiB) Viewed 6067 times

https://www.desmos.com/calculator/eykhbatbn6

Using this calculator, the speed of the attached ship is ~203 km/s.

- Increasing the width to 30 gives a speed of ~151 km/s. Increasing the number of thrusters to 7 (maximum per width) gives a speed of ~158 km/s.

- Comparatively, quadrupling mass to 1000 (the maximum possible in the calculator) gives a speed of ~195 km/s.


The advantage of extra width is better asteroid collection. However, the gains are minuscule compared to the penalty width gives. For the above ship, doubling width and fuel rate gives a speed of ~183 km/s.

- Making the ship 4 tiles narrower and removing one of the engines would give a speed of ~225 km/s, and a further reduction of 4 tiles and 1 engine would give a speed of 250 km/s.

Why?

As per the above, any ship can be *significantly* improved by taking all its production and stacking it vertically, turning it into a cigar shape -- the thinner, the better. The gains are considerable -- 50-80% speed increases are not uncommon. Any increases in mass due to routing spaghetti are insignificant compared to chopping a couple of tiles of width. This significantly bottlenecks platform design.

Since mass is the parameter shown in-game, I suggest flipping the two in the formula (with some adjustments), so that mass gives the biggest penalty to top speed whereas width would only impact acceleration.

In real life, spacecraft delta-v is inversely proportional to mass, while shape has almost no influence as long as mass distribution is somewhat symmetrical.

[mmmPI]

I think it's a positive that "cigar-shape" are "more efficient". That create an incentive to not make "bricks" it sort of mimick "aerodynamism" even if doesnt make much sense in space, it incentivize players to make "realistic" spaceship shape regarding the norms set by science fiction ( yeah not real science i concede ). Or at least be rewarded when trying to make them slick.

I don't think it would be a problem if mass had more effect, but i wouldn't want to see it swap with platform width as the biggest offender here.

I think it could lead to a situation where the fastest ship is just an horizontal lane of thruster, and the wider the faster. And it wouldn't feel like "optimizing" anymore if "more is always better".

[skaudus]

The issue is that currently, the optimal shape is a vertical line.

By making mass more important, it allows for other shapes, such as triangles, or irregular organic shapes.

12-16-2024, 12-04-58.png (1.17 MiB) Viewed 6055 times

This design has a very low mass but a wide scoop up front to get asteroids. Despite it being a skeleton, it's much slower than an equivalent cigar shape.

I know it's a lot slower because there was a ship race a month or so ago, and every top scoring ship was a cigar.

[CyberCider]

I completely agree with you, but I think “cigar ships” are actually the smallest problem this causes. To me, the most damaging aspect of the way speed works now is the fact that there’s no reward to building space-efficiently.

As long as the backside of the ship is covered in thrusters, you basically have max speed. The ship can be any shape or size, it doesn’t matter. I think it would be much better if smaller, denser ships could achieve higher speeds. Because they’re more difficult to design, and probably include quality items, so this should be rewarded. Instead, right now they can only achieve the same speeds as giant bricks full of empty spaces.

It makes me wonder why they even added the “no holes” mechanic, if the number of tiles matters so little in the end. Just aesthetics? Resource costs? Neither of these things matter at all in the long run.

[IsaacOscar]

Yes please! This surprised me so much when I discovered it, I wasn't sure if it was a bug or not. so I reported it 122717, unfortunately I didn't get confirmation as to whether it was intentional and why.

Also cigars are not the optimal shape, see 123935

[skaudus]

Also cigars are not the optimal shape, see 123935

I didn't mention this since I believe thruster stacking is an exploit, and such designs are even more egregious

[mmmPI]

The issue is that currently, the optimal shape is a vertical line.

I don't think this is an issue as i think the horizontal lane is worse. ( viewtopic.php?p=636365#p636365 )

By making mass more important, it allows for other shapes, such as triangles, or irregular organic shapes.

To me that's not compelling, because the optimal shape would still be a horizontal lane the same way as currently it is a vertical lane imo.

This design has a very low mass but a wide scoop up front to get asteroids. Despite it being a skeleton, it's much slower than an equivalent cigar shape.
I know it's a lot slower because there was a ship race a month or so ago, and every top scoring ship was a cigar.

I've seen that design in the discord, but i don't expect it to win a race, the aerodynamic profile is not good. I think it is more "intuitive" even if scientifically incorrect to have the ship designed for speed to be thin, and ships designed for collecting ressources to be wide.

I would be ok to alleviate the impact of width, and increase the one for mass, but i still think there shouldn't be such drastic change as to invert the way those 2 works.

[mmmPI]

I completely agree with you, but I think “cigar ships” are actually the smallest problem this causes. To me, the most damaging aspect of the way speed works now is the fact that there’s no reward to building space-efficiently.

There is a reward for building space-efficently, it's just not very good, it serve to determine which ship is the fastest between 2 of the same width and same thruster coverage.

[adve-nyan]

Current optimal shape assuming enough production of ammo and fuel is brick
Don't matter vertical or horizontal so long you have all thrusters placed

Tbh I hate this

[zig1000]

After thinking about and playing around with this, I would propose something like the following formula for speed (in km/s, mass in tons, thrust in MN):

Code: Select all

max_speed = 500 + 200_000 / (mass + 200)                  # +200 avoids low-mass asymptotic speed silliness.
drag_efficiency = thrust / sqrt(thrust^2 + 5000*mass)     # ranges from [0, 1).
speed = max_speed * drag_efficiency

Not sure if that would necessitate acceleration changes too or not.

The idea is that this places a mass-dependent upper bound on speed, so that past a certain thruster count, squeezing all the mass horizontally has diminishing speed gains, depending on the scale of the ship. That also quietly quashes the thruster-stacking exploit, which AFAIK the devs wanted to kill too (though I'm sure that'd make some people sad).

See https://www.reddit.com/r/factorio/comme ... d/nxxxnub/ for more some numbers and more details.

EDIT: Here it is in desmos: https://www.desmos.com/calculator/l2fqk6pv7p

[eugenekay]

There is a mod which lets you alter the space platform acceleration any way that you like. I encourage you to try it and share the before/after results from changing the balance factors.

Max Speed and Drag are not defined separately; the usage of Speed and Width in this expression controls the point at which "Acceleration" is reduced to 0, giving a steady-state velocity for a given Thrust level.