THE NEW Miscellaneous Talk That Doesn't Deserve A New Thread Thread Thread (Part 2)

Would that scale only apply from awakening onward?

2 Likes

I think so. It’s only relevant when there’s energy consumption technology, because it’s a measure of energy control for civilization after all. It’s hard to say this is below the awakening stage.
Below that, I think it would be correct to perhaps divide it between macrophage and awakening like they did in Spore but a little differently - level of brain/neuronal/intelligence development. Like a kind of milestones without completely removing the option for alternative gameplay.

3 Likes

Going from just below one on the Kardashev to like 1.1 would not significantly change the gameplay so I think it is actually a pretty bad way to split parts of the game.

5 Likes

I totally agree. Maybe a meter somewhere in the corner, could be fun, but the primary progression system? I doubt even ascension gate making players will be at 3 so it just feels badly calibrated.

4 Likes

Post ftl space stage can be its own stage and pre ftl space stage can be added to industrial.

Or actually, why do we have the space stage? You can enter space stage before industrial. With a grvity battery trebuchet.

Do we have a ā€œcolombian stageā€ for discovering a new continent? No. Austronesians discovered the americas at a very early date. In fact, it was their sail designs that were adopted by the arabs and later the europeans[1].

And you may not even have another continent in your planet. It may be pangea. Similarly, you may not have other planets in your solar system.

Why have a stage, based on ā€œwhere you areā€ rather than ā€œwhat you doā€?

There needs to be a difference between before and after a stage starts. For example, when you make a cell colony, you now have a brand way of interacting with the world. You can design the placements of the cells rather than just the insides. In macroscopic, you interact with tissues. Later you start to move, control multiple organisms and then a city.

Nothing changes when you pass the karman line. You still have rockets. They go a bit further. You have a special rocket that takes you to to another continent that has a different gravity and the air is toxic. And there are these asteroids that have no gravity.

I think, lets have a ā€œfusion stageā€ rather than a space stage. The stage names depend on the energy type you use.

Industrial stage is when you make fossil fuel powered machines. Coal is much more cheaper, plentiful and more energy dense than wood. It makes machines more cheaper. Instead of a plow, you can have a tractor. Instead of handweaving, you get machine workers. Everything changes.

You would need to exit this stage before the fossil fuels end. Or earlier than they end.

The stages can be

society stage, industrial stage, solar stage, fusion stage

In the solar stage, you would discover solar panels. You need to have electricity before that. If you stay in your own planet, you would replace all the roofs with solar panels, eventually, all the surface of the planet, and you would replace farming with glucose and amino acid factories, that create these substances using electricity. If you go to space, you would start building a dyson swarm.

In fusion stage, since fusion is always 20 years away, it can happen anytime now. You would be able to turn hydrogen into free energy. If you use all the water in your planet, you would need to turn your population into machines. If you go to space, gas giants woukd be the primary places of settlement, because the hydrogen is easier to extract than stars. You would start mining the star after the dyson swarm is completed.


Now I will talk about a completely different topic. Banning doubleposting was a bad decision. It was an overreaction to fluffposting. Post boundaries are better, visually, than the line boundaries.


While a neural network can implement an optimizing system like evolution or gradient descent,

If there is an explicit ā€œgo faster geneā€, it is easy for it to get selected. If it depends on the tradeoff between different limb placements, if you need to move a limb slightly, make one muscle larger, another smaller; here, there are lots of variables. In fact, a variable number of variables.

1)The max number of limbs should be capped.

The function you want to maximise isn’t clear. You can’t do hill climbing, gradient descent, heavy ball, etc. Wikipedia has too many formulas and algorithms. Testing each step is expensive. It would need a physics simulation. But it might look intuitive to a human.

An AI needs to record the results of all that, and the next time, it should say which direction to go. If you make a simulation in someone’s computer, and if thats forgotten, not used as guidence for simulations happening in other computers, then it would be vain.

  1. Someone who has a slow computer can log onto the internet and his game can be simulated in other computers?

  2. Record all the previous simulations and feed it to an AI

  3. Updates to Thrive physics engine should be gradual, so that the AI can slowly adapt to it, and the AI’s from the previous versions can still be used, you don’t need to train it from scratch.

  4. Make AI’s within AI’s, and don’t try to help them

For example, lets say that you have an AI that gives you better legs for running faster, by tweaking the bodyplan you already have. For any type of body, the guesses it gives are better than random changes. But maybe, you don’t need to run fast. You need to carry weight or jump high instead. There can be a general AI, that looks at the circumstances you’re in, and choses which of the super specialised AI’s should be used.

Maybe you don’t know whether you want to run fast or jump high. Maybe you need to maximise a completely different thing. Leave that up to the AI as well. Keep it a black box. Let it decide how many sub-AI’s it wants to train.


And the biggest factor in the environment, that makes it different, is the creatures living in it. There are creatures adapted to climbing trees. An organism shaping the directions other organisms can take. Niches aren’t static and empty, needing to be filled. Your shape can completely change the niche map.

All the body shapes and other information about the species that were in that patch in the previous generation, or migrated there, should also be given as an input to all the AI’s.

If species tried to compete with each other by predicting what the others are likely to evolve into, that would be so cool.

  1. The max number of organisms in a patch should be capped.

  2. Organism editor can give hints to the player, ā€œthe AI suggests you to make this into thisā€

There can be a ā€œfind the missing speciesā€ AI. You give all the bodyplans of all the species in a patch, all except one, and you ask it to guess the missing species. And it is rewarded the closer it gets. This AI can later even be used in real life, ā€œthis tree has this type of nut, so the missing species should be this bird with this beak, and the bird should be this size because this predator does this and this.ā€

  1. Tell everything to the AI. The more it knows, the better it can get.

And it can also be used for game debugging. It can give intelligent human like responses to why a certain glitch happens in the physics engine. Or at least it can point out to that.

Imagine in the future, if an AI can give your species a score for ā€œhow fast you can avoid diving birds and collect berries from the top branches of the shrubs by jumping from the ground, revealing yourself, and changing your direction using a biologic rocket that ejects water and recharges before each jumpā€ and you get an achievement for getting it to 80/100, with 100 being the best thing AI ever designed and 0 is a very heavy thing that barely fits the criteria.


  1. wikipedia : They also established vast maritime trading networks, among which is the Neolithic precursor to what would become the Maritime Silk Road. ā†©ļøŽ

From my understanding you’d most likely need to achieve orbit in order to reach space stage…

1 Like

Yes. Trebuchets use gravity batteries. You lower the counterweight, and in return, it sends a projectile into the air.

You can connect multiple gravity batteries to a weight. Lets say you want to send 1 tonne to space, you make a ramp, 1 kilometer, whatever

you connect all the weights to the capsule, they start falling, you cut the ropes when the capsule passes by them and they’re on the ground

there isn’t an upper scale to the speed you can reach. you need a aerodynamic capsule and you need to reload lots of weights.

1 Like

I sure wonder what will Hhyyrylainen say about the idea of reaching space stage in society stage…

1 Like

3 Likes

Not sure if you can reliably achieve orbit with this…

1 Like

The stage is defined by the act of launching rockets to space. So basically doing space travel.

So the definition already fulfils this condition:

4 Likes

By ā€œrocketsā€ here could it mean ā€œvehicles capable of operating and moving around in spaceā€ (as I’m not sure if things like project orion spaceships are considered ā€œrocketsā€ and it’s not impossible to have a civ launch such a spaceship into orbit first, as in without doing so with ā€œrocketsā€ first)?

1 Like

you can go to space without rockets and you can have missiles with chemical propulsion that can’t go to space. aah is too fast.

2 Likes

Missiles are pretty much obligate industrial stage tech. Also aren’t missiles just militarized rockets?

2 Likes

I just looked it up and you need to reach 40 000 km/h to escape earth orbit. If we assume that you can’t vacuum everything, the terminal velocity of a rock is something like 100 km/h. Which is why, the ropes should be connected with pulleys, multiplying the speed.

400 rocks, fall at the same time, one rock, go to moon. I think it makes sense. You just need good ropes. They don’t need to withstand this more than once or twice.

To land on the moon, if it has an atmosphere, you can use a parachute. Once you’re down, you have gravity again, you can construct the same thing and return.

If the moon doesn’t have an atmosphere, you can use a skyhook to land down. Take out a second capsule from the first capsule, make them rotate around each other, and make the speed and distance so that, one of them becomes stationary with respect to the moon, when its at the lowest point, during which, the crew can just walk outside, and fall a little or maybe don’t fall. This could have been used in our moon landings as well.

No, wait, it couldn’t, because the moon is tidally locked

2 Likes

Wouldn’t the crew of such a ā€œrockā€ burn up in the atmosphere through?

They would be going at mach 33. If they can’t find a heat resistant material for that, they can let the outer surface melt away and reach space with a smaller ship.

If they bring lots of water, they can boil it to keep the center cool.

If the ship is covered with aerogels, molecules move very slowly in an aerogel, they can release the water into the aerogel, from the inside, it would take quite some time for it to go to the outside.

The heat covering can get to high temperatures. They can cool it to 100 degrees while inside the atmosphere. When they’re in space, they can cool it down to 0 degrees with water.

I recently learned that KSP2 was received negatively and that a ghost manager is still receiving money from the game being sold on Steam.

2 Likes

There is something called a ā€œghost managerā€? I don’t remember learning this in my business courses… Unless the Ghostbusters are doing a bad job recently…

Edit: Never mind. I learned it as a ā€œVampireā€ manager.

2 Likes

I’m so sad about it i’m a huge ksp fan, but ksp2 never got nearly as good as the original, super sad.