Jump to content

World size questions


Recommended Posts

How big? Well, the stock planets will be the same sizes and in the same places as the original, but consider this: the Kerbol system is 1/10th scale; a light year is about 95 trillion km; the nearest star to the Sun is about 4.5 light years away and there are about 20 stars within a 10 light year radius. To make a 'realistic' interstellar journey in Kerbal scale, I would expect distances of between 20 and 100 trillion km between Kerbol and other stars.

How many stars? I would expect several new stars, maybe 2-4 to offer multiple destinations and each with its own solar system. Any less and it's boring, any more and they have to make many more planets and moons to fill them up, which would take time to make them interesting.

Link to post
Share on other sites

Let's speculate!

Let's imagine each planetary system, or moon system is its own puzzle, akin to a level in of itself.

KSP1 has 16 celestial bodies (5 planets, 2 dwarfs, 9 moons, excluding the star Kerbol). If they wanted to have ~100 planets and moons to explore, then there may be 10 visitable stellar objects.

There are 52 visible stars within 4 parsecs (13.05 lightyears) of Sol. Assume that the maximum velocity of a given spacecraft is 0.990c, and that timewarp still has the same scale delimiters as KSP1: 100 000x is maximum timewarp, giving a Kerbin year at 92 sec/year. At this sort of scale, it would take about 20 minutes running at maximum timewarp and maximum theoretical speed (ignoring acceleration phases) to travel from the center of this imagined world, to the most extreme edge. It would take 40 minutes to go from extreme edge to opposite extreme edge.

Even if the game is scaled down in terms of distance or number of stars, 500 places to visit sounds like a lot and 'only 50' seems like very few.

Although we have no evidence to do so, I think it is reasonable to assume the game will be within these sorts of bounds.

HabHYG15ly_LargeTB.jpg

Map from Atomic Rockets, retrieved 2020-11-14

Edited by starcaptain
Link to post
Share on other sites

I'm going to be conservative and guess around 3 systems worth of work. I can't see them pumping out too many systems up front when they can be made DLC later on.

1) Original system

2) Full new system near the original

3) More distant, strange system with less worlds but more unique/far fetched ones. 

4) About half a systems worth of other stuff. Maybe a rogue planet, machine world, black hole, or even a nebula with a utopia world hidden inside?

Link to post
Share on other sites

If KSP2 really is built to be easily mod-able, I would expect lots of mods designed to add new systems to the game will quickly pop up. Some such already exist for KSP1, but it sounds like KSP2 will be built with a foundation to make such additions easier; with a system for travel to other stars already included in the vanilla game, it'll just be begging modders to let their imaginations run wild and dream up new systems for us to explore. :)

Link to post
Share on other sites

Well, there are few mods that give us multiple star systems, but they all are quite heavy. With new game, and system already done, even converted from old mod, one should be just able to determine place in a galaxy, drop the system there and voila. No need to replace anything. After a while, there could be even an actual Kerbal Galaxy pack, with different star systems from different people all in one.

...with that said, it will still take me years to even get through the stock game.

Link to post
Share on other sites

Just as KSP showcased a variety of planet types in the kerbal system, I'm assuming, along with the increased scope of KSP 2, that we will similarly now see a variety of star system types.

Some of what I'm expecting or hoping to see would be:

  • The Kerbal system
  • A young system where many of the planets have rings since we heard Nate talking about that in one of the first dev diary episodes (I expect here is where we will find rask and rusk)
  • A system with a red giant star
  • A system with a red dwarf star
  • An all Jovian system
  • A binary star system

Possibly some more exotic I hope for:

  • Neutron stars
  • A black hole
  • A planetary nebula (recent supernova remnant)
  • A white dwarf
  • Rogue planets
  • Hypergiant with a dense binary partner with an accretion disk
11 minutes ago, kerbiloid said:

How much hard is to auto-generate KSP-1-like textures varying the coefficients?
Is it really needed to store them as bitmaps? Unlikely they are handmade pixel by pixel, probably they are made by moving several brushes..

what?

Link to post
Share on other sites

From a gameplay perspective the size of the planets and moons in the Kerbol system works fairly well.   And they are confirmed to be the same size in KSP2. 

Also, bigger celestial bodies have much more surface area, so that means a huge amount of extra work detailing them up if they are to look right.  I would therefore expect the 'average' sizes of  bodies in the new star systems to be about the same, but with quite a bit of variety overall.

Link to post
Share on other sites
4 hours ago, kerbiloid said:

To auto-generate numerous planets rather than store their textures in game resources as image.

This has already been asked before in interviews and said to be not the case with KSP2.

Each celestial body will be created with specific attention paid into making them like unique puzzles to solve, with their own interrelated environmental challenges. So each world is more like a level, with its own gameplay design. Think for example about the navigational challenges of the Eve system vs. the Jool system. Achieving this sorts of distinctiveness is quite intentional, and procedural generation of these sorts of natural features is impractically difficult.

That said, the surface level features and details themselves (such as rocks, trees and minutia of the texture) are almost certainly going to be procedurally generated. The levels' planetary-level map and biomes, will not.

Link to post
Share on other sites

I would want a 7- Star system extremely far away, with a bunch of hypothetical and rare planets and stars like quark stars, strange stars, quasi-stars, black dwarves, diamond planets, planets that are on the verge of being liquid but are still not solid, and some other strange stuff.

And don't forget a really disfigured red dwarf, and a contact binary of two gas giants with everything having a really strange color, shape, and orbit. And maybe it could be hidden in the tracking station.

AND MAYBE THEY COULD RELEASE IT ON MAC OS!  :mad: - :antinormal::normal::antiradial::radial::targetpro::targetretro:

Link to post
Share on other sites
On 11/14/2020 at 6:51 PM, starcaptain said:

Let's speculate!

...

There are 52 visible stars within 4 parsecs (13.05 lightyears) of Sol. Assume that the maximum velocity of a given spacecraft is 0.990c, and that timewarp still has the same scale delimiters as KSP1: 100 000x is maximum timewarp, giving a Kerbin year at 92 sec/year. At this sort of scale, it would take about 20 minutes running at maximum timewarp and maximum theoretical speed (ignoring acceleration phases) to travel from the center of this imagined world, to the most extreme edge. It would take 40 minutes to go from extreme edge to opposite extreme edge.

Even if the game is scaled down in terms of distance or number of stars, 500 places to visit sounds like a lot and 'only 50' seems like very few.

Although we have no evidence to do so, I think it is reasonable to assume the game will be within these sorts of bounds.

I would disagree with your assumptions. I would assume a maximum velocity of around 0.1-0.2c (although 0.5 may be achievable if they have antiatter rockets, avoiding velocity faster than this also has the nice side-effect of making it Ok to ignore time dialation). Given the general 1/10th scale, I would expect the nearest star to be 0.45 light years, not 4.5. I would also expect maximum timewarp ot be at least 10x higher. So assuming 1/10th the velocities, 1/10th the distances, and 10x the time warp, the 20 minutes comes down to just 2 minutes... which seems fine to me.

The standard KSP system is smaller with less bodies than our system. So if our system has 52 "nearby" stars, I would expect the KSP system to have less than 50.

plus, >50 is a lot of work...

So that's my speculation

18 hours ago, pandaman said:

From a gameplay perspective the size of the planets and moons in the Kerbol system works fairly well.   And they are confirmed to be the same size in KSP2. 

Yea... but it continues working fairly well up to about 4x, then the excessive mass and low tankage ratios really start to hurt.

18 hours ago, pandaman said:

Also, bigger celestial bodies have much more surface area, so that means a huge amount of extra work detailing them up if they are to look right.  I would therefore expect the 'average' sizes of  bodies in the new star systems to be about the same, but with quite a bit of variety overall.

Well, even in KSP 1, procedural generation was used to generate the finer scale detail of planet surfaces (including terrain height, not just the cosmetic ground scatter). I imagine that KSP2 will still use procedural generation for fine scale detail, even if globally it is a custom made piece of art.

So I hope larger scales work, because I plan on modding the KSP2 system from the start

Link to post
Share on other sites

The problem of KSP is that it is trying to cover two opposite gameplays and strategies.

1.
A rookie rocket, made of parts found next to the road.
Micromanagement, manual attachment of every buttery, cable, or pipe, manual piloting without autopilot. Micromanagement of mining and manufacturing. Life support, Kerbal health.

Requires as small rockets as possible.
Requires a Sun-Kerbin-Mun-Minmus world without other planets, because interplanetary flights dramatically increase complexity and requirements. Low delta-V, small planetary sizes.

So, the rocket construction is limited, and your way is to mine in your mine, process moonshine for fuel (oil for kerosene at the next progress phase), manually assemble every rocket, and the rockets would be very primitive, similar, and you're focused on flights and manufacturing rather than new parts discovery.

To make the progress interesting, the only two available targets (say, Mun and Minmus) should be much more detailed, because they should replace all other planets.
So, there should be one big moon of miracles (with air), one Moon-like airless moon with artifacts (resources, derelict ships), and a small hardly available mystic moon.

Then you visit various locations on these moons many times on similar primitive rockets, routinely building them from what you had looted in dirt.

I.e. first you have that barn, crops (greenhouse or garden), so you turn the crops into food for yourself and in fuel for your rover, Cessna, and primitive rockets, just to reach the orbit (Atlas-Mercury).
You keep searching for oil on your rover, Cessna,  and from the orbit.
Once you have found the oil near your place, you put a rig and start mining oil and producing kerosene. Et cetera.
On the kerosene rocket you can reach the moons. You need food (you grow it) for the journey, and oxygene for both you and engine.
You have  just several Kerbals (team? family? ), so you can be one of them (like now), and you can't populate large bases and other planets.
You need just a small, Salyut-like orbital station for 1..3.
You don't live on the Moon longer than you need, so you just visit the lunar base. On the other hand, you can reasonably micromanage it.

See also LBA/LBA2 (Little Big Adventure) as an example of very much differing world with numerous locations.

I.e. "pilot-miner-explorer" or "hillbilly" way.

2.
Colonisation.
Focused on interplanetary network, cargo traffic, macromanagement, progress in rocket parts (new and new types of engines, reactors, etc).

Almost opposite to the previous one in almost everything.
It looks silly to attach every pipe and wire when your parts are "tocamac module" and "universal habitat". So, the micromanagement  gets  useless.

You need moar ships, ISS-like orbital stations, and you can't pilot them all at once, so you need to automate their flight.
Usually you don't need a pilot, you pilot the whole ship.
Also it just looks silly to land a superfreighter manually, when its fusion reactor is controlled by computers.

I.e. "CEO" way.

***

The former way is limited from above, the  latter from below.
They require two absolutely different universes.

So, while the KSP engine allows to implement both ways, this should be splitted into phases, like in Spore.

***

In Spore  you're first an infusoria, then a rabid rabbit, then a savage, then a savage with bombs, then a highly developed interstellar being.

In KSP-like game there should be two phases, differing not just in KSC design. 

The first way need a masterpiece planet, manually made.
The second way requires procedurally generated planets, see No Man's Sky or Spore.

Edited by kerbiloid
Link to post
Share on other sites
49 minutes ago, kerbiloid said:

Snip

What you're describing seems like a mix of games with Stationeer as phase 1 and Surviving Mars as phase 2 but Kerbal focus isn't in what either those 2 games do, Kerbal core is in fact in what both of those game lack, orbital mechanics. The main gameplay loop of KSP is building a rocket/spacecraft to solve a specific problem or puzzle and that calls more for a mix approach, one in which those two phases are not one after the other but they proceed in parallel, and the player is not "forced" to do any transition, because sometimes you whant to build a 5 cores "Saturn Heavy" while some other times you may want to send a probe to Duna and get as much science as possible with only 0.625 parts.

Your distintion of "big rockets vs small rockets" is too much a "black and white" one, you won't replace a comnet sat launch, a exploration rover to a new planet or a mapping satellite in search for new resource deposits with a skyscraper-sized colonization ship, as you progress in the game you'll need and want both at the same time.

Let's use a "the Martian" mission profile as an example, yes, you use a big mothership to move the crew between Kerbin and Duna, but the robot reconnaissance for the landing location, the presupply missions, the HAB, the DAV will get there before the big mothership and using traditional "small rockets".

KSP is not Cities Skyline or Spore the point is not colonize, colonization is just an excuse to give you new and interesting rockets to design, just like the passengers and industries in OTTD (open source remake of Transporter Tycoon Deluxe) are just an excuse for you to play with train lines and a good one at that, you'll need to move several KSCs worth of mass to build your colonies and the ability to build new space centers on other celestial bodies will change your point of view for the next wave of missions (ever launched a Eeloo Apollo Style exploration mission or even a mere Bop commsat from Laythe?).

And then, when you're bored with the system you're in, you build a huge colonization mothership and get off to another system, maybe searching for a Kerbin-like  planet (huge boost in colony development and huge discounts on resource usage?) to colonize and start over.

That's coincidentally why I think that this game could go well even with "only" 3 or 4 additional star systems, it will never need more than what the Devs (and modders) can design manually, exploration shouldn't be "how far I can go with the strongest engine and 20 minutes at the maximum timewarp".

Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...