Jump to content

KSP2 Science System - The limits of science points


Challyss

Recommended Posts

Hello everyone.

Following a reddit thread, let me post here some thoughts about the current science system and what I'd love to see in KSP2.

Currently, experiments give you science points. Some missions or achievements also give you a small amount of points. You use these points to unlock nodes of a tech tree. Nodes give you parts. Upgrading the RD Building allows you to reach further in the tree. You may (according to your career settings) need money to unlock each different parts.

What I like about it. It's a familiar concept in games. It's easy to understand. Fancy parts are harder to get.

What I dislike about it. All the experiments are equivalent (1). It encourages grinding science (2). Experiments aren't connected to unlocked technologies (3).

(1) Science points are science points, no matter where they come from.
(2) I feel like it's a lot of "go that high, go that far, bring thermometer, barometer, goo, come back"
(3) I know the temperature of the North Pole of the Mun, my scientists can finally develop supersonic air intakes !

A system that I think would be interesting (I will most likely purchase KSP2 in any case but hey, my pride would be infinite if I can contribute) would be to..... Have missions to unlock technologies or parts.

Yeah, revolutionary idea, right ? 

For instance, launching the first uncontrolled rocket "that high" would unlock basic control. Then the first suborbital flight unlocks higher vacuum ISP engines. The first orbit unlocks basic RCS, while the first rendezvous unlocks docking ports. And so on. You get the idea.

EDIT : I think I got misunderstood at some point. I didn't mean that the "missions" or "milestones" would come one after the other in a predetermined sequence. You could have a tree as well. And chose how you want to progress through that tree. My point is just that instead of having science point requirement to unlock technologies or parts, you'd have to do specific stuff, or go to specific places. But not in a specific order. /EDIT

Why do I think it would make sense ?

Well, right now, you can basically have almost everything unlocked before you send your first interplanetary craft. Not so challenging. So it would be more.... less..... well, I don't know. It just doesn't feel right.

But moreover, and especially with colonies and interstellar flight, it makes even more sense. Because... imagine making your first colony and.... you already have unlocked fancy habitat techs. Nah.... The first offworld colony must be a dusty sh*thole. Barely habitable. But from that experience your scientists would learn sustainability. And so on.

That being said, a science point system can still be useful to either upgrade parts (cheaper, lighter, harder, whatever-er) or to unlock some better technologies. Like the first Orbit unlocks small RCS. Science points unlocks bigger RCS blocks or increase the ISP, or allows for LFO, ion or other kind of technologies.

 

You've read so far. I'm flattered. What do you think about it ?

Edited by Challyss
Grammar / Precisions
Link to comment
Share on other sites

1 hour ago, Challyss said:

That being said, a science point system can still be useful to either upgrade parts (cheaper, lighter, harder, whatever-er) or to unlock some better technologies. Like the first Orbit unlocks small RCS. Science points unlocks bigger RCS blocks or increase the ISP, or allows for LFO, ion or other kind of technologies.

Upgrading is a good idea. Make the part cheaper by spending science points.

1 hour ago, Challyss said:

Why do I think it would make sense ?

Well, right now, you can basically have almost everything unlocked before you send your first interplanetary craft. Not so challenging. So it would be more.... less..... well, I don't know. It just doesn't feel right.

Add a science generating slider so you can have extremely difficult advancements for veteran players, and noobs won't be scared of the difficulty and turn into veterans.

Link to comment
Share on other sites

I agree with your criticism. In real life, most progress in space flight technology seems to come not from scientific data collected by the space programs but from practical experience. There are some exceptions, e.g. measuring radiation levels or atmospheric density helps prepare for missions to these places, but overall measuring Saturn's gravity field won't help you design a new engine. What will help is using and improving the old engines. So it doesn't improve immersion when you need that Gravioli detector data to unlock Mainsail.

On the other hand, "railroading" technological progress through a series of mandatory missions is not something people will like for a sandbox game. So I hope KSP2 will find a way to avoid science grind and to reward players for interesting, daring missions.

Link to comment
Share on other sites

In another post someone suggested different points (science, engineering, medical and so on) with different requirements for every part. I would love a semi-scripted requirement test mission for each node you unlock, immagine having to build and fly a Bell X-1 analogue to have the supersonic parts or a X-15 for Hypersonic flight.

Link to comment
Share on other sites

24 minutes ago, Master39 said:

In another post someone suggested different points (science, engineering, medical and so on) with different requirements for every part.

I like the extra flavor of that.

32 minutes ago, garwel said:

On the other hand, "railroading" technological progress through a series of mandatory missions is not something people will like for a sandbox game. So I hope KSP2 will find a way to avoid science grind and to reward players for interesting, daring missions.

It's already somehow the case in the structure of the tech tree. Plus, the career aspect is kind of a deviation from pure sandbox. I mean, there is an actual sandbox mode after all, without technological restrictions.

Link to comment
Share on other sites

5 minutes ago, 5thHorseman said:

So once you've landed somewhere (mun) and gone somewhere (Sun orbit) what will still need unlocked, and how will it get unlocked? How does "Land on Duna to unlock the NERV" make any more sense than "These temperature readings at Duna gave me the NERV!"

Bringing an RTG in solar orbit to mesure the effects of solar wind on radioelements makes more sense, for instance, than a sismic measurement on a moon.

Putting a certain mass in orbit could unlock high thrust engines.

Landing on Duna and measuring the temperature and pressure could on the other hand unlock techs related to land habitats or off world resource processing.

Of course, something like unlocking an aerodynamic nosecone with a fly--by of a moon makes no sense....

Link to comment
Share on other sites

54 minutes ago, 5thHorseman said:

So once you've landed somewhere (mun) and gone somewhere (Sun orbit) what will still need unlocked, and how will it get unlocked? How does "Land on Duna to unlock the NERV" make any more sense than "These temperature readings at Duna gave me the NERV!"

I agree, it's still pretty arbitrary. I also agree with the point that it should avoid railroading, which would very much be counter to the spirit of KSP. 

 

Perhaps, rather than earning science points from experiments, they could be earned by testing predecessor parts in certain environments. 

For example, the Lv-t30, with its lack of vectoring could perhaps be considered a predecessor to the lv-t45. Perhaps the spark could be a predecessor to the even smaller engine. Just examples, not actually suggesting this as a progression. 

If you understand how a part functions in different environments, you can build a better, more versatile one. Perhaps some combination of harder to get to environments, and using more "advanced" engines would result in you earning more points to wards the next part in that line. 

 

My concerns with this type tech tree are:

-is it just getting in the way? I hope it would lead to a natural progression. It allows people the possibility of designing specific missions to test tech, should the so wish. 

-will it be too limiting? It's more restrictive to which parts you get next, but makes a lot more sense, and could increase immersion. 

-does it make enough sense? Would people understand what to do? 

-possibly less modular as a piece of game design. The specific nature of it would make it harder to sort parts into groups, and harder to integrate mod parts. But I think it could still be done based on parts shared features. 

 

I would very much like to see the current experiment type science devorced from the tech tree system. Science could just boost funding, reputation, and be there for its own sake. 

I've never been a huge fan of the tech tree. It is too prominent a game play mechanic imho, distracting from the actual space stuff I came to do. 

I'm fine with most parts being unlocked before you do much interplanetary stuff. I think there's better ways for the game to measure your success. 

Edited by Tw1
Link to comment
Share on other sites

47 minutes ago, Tw1 said:

I agree, it's still pretty arbitrary. I also agree with the point that it should avoid railroading, which would very much be counter to the spirit of KSP. 

Perhaps, rather than earning science points from experiments, they could be earned by testing predecessor parts in certain environments. 

For example, the Lv-t30, with its lack of vectoring could perhaps be considered a predecessor to the lv-t45. Perhaps the spark could be a predecessor to the even smaller engine. Just examples, not actually suggesting this as a progression. 

If you understand how a part functions in different environments, you can build a better, more versatile one. Perhaps some combination of harder to get to environments, and using more "advanced" engines would result in you earning more points to wards the next part in that line. 

True, the game should use time and money to fund Research, not science points. Science should only be exchanged for reputation and funds and contracts.

Link to comment
Share on other sites

I also think that part unlocking should be made independent of Science, as collected on scientific missions. IRL, everything comes down to money, more or less. That's kind of the point of having such a thing as money, if you think about it. Doing science should be one of the ways to earn money. In the career mode, money should be the primary mechanic of the game. There's no need for a separate currency for part unlocks. Doing science should be tied to contracts, with "science of opportunity" giving small monetary gains and reputation.

Additionally, research takes time. With life support involved, there would be more to time-based mechanics than just "timewarp until done", though early research would inevitably involve that. Indeed, IMO, waiting is better than grinding, because the former can be skipped with warp, and the latter can't. You should have to pay facility maintenance, so timewarping while doing absolutely nothing is not the optimal solution. Building on that, perhaps "Science points" could be equivalent to "lab time" from the Kerbal Academy of Sciences. You give them what they need for their research, and they give you more lab time that you can spend to improve your rocket tech faster. Maybe also discounts, but I'd tie that to reputation.

Ultimately, I don't think we need "Science points" at all, or at least not as a mandatory component of the career. You may want to run a space tourism business, not a scientific program. KSP was originally envisioned as sort of "Rocket Tycoon". KSP2 should take a good, hard look at the economic component, since it's one thing Squad managed to botch completely, and I haven't seen a mod that would truly fix that.

Link to comment
Share on other sites

I suspect I'm in the minority here, but I actually quite like the science system as it exists now - it's the one part of career that actually works, as far as I'm concerned. The gamplay loop of "go place get science to get parts to go more places to get more science..." is simple, but it does feel quite rewarding, at least until the end of the tech tree. As it stands, it's still the only reason to go anywhere in KSP1, outside of "just 'cause".

That said, it obviously has some flaws, like collection could do with some streamlining. In particular, I'd like to see most instruments be "passively" collecting rather than the player having to guess when they're in a new biome or situation or whatever, so you just get a chunk of science at the end of mission, with only the big one-use experiments needing any user input. And the tech tree could do with some reworking, maybe more nodes with fewer items in each or something? You could also have different types of science points, which you might need a mix of for certain nodes - like "rocket science points", "plane science points" and uh, "science science points", just to make it slightly more appropriate for certain parts.

But yeah, the core system I think works rather well, and doesn't particularly need to be gutted like most of career mode. If anything, I think it'd work better in KSP2 - half the problem in the orginal is that you've probably unlocked most things before you go interplanetary, so there's not much to "get" by that point. In KSP2, it looks like there'll be multiple levels of technology, so the tech tree can keep going much longer, without the feeling of missing parts that you should just have like in KSP1.

Link to comment
Share on other sites

One thing that’s always bugged me about science is the whole biome thing, and I think that’s part of the reason why you could unlock interstellar stuff before even leaving Kerbin’s SOI. There are just too many biomes in this area and that increases the amount of science you can get dramatically. They definitely need a nerf, regardless of what science is used for.

I get what the biomes are trying to do, encouraging exploration and all, but I don’t think the sheer amount of science each biome can provide makes sense. I think biomes should be more the icing on the cake of science; they add some extra information with repeated experiments, but it’s just doing the experiment the first time that really counts. To achieve this, that experiment which gives you that 100 science could be separated from biomes, and instead of each biome giving you another 100 science, it would only be like 10. Maybe with some bias so more interesting biomes give more bonus science. This way, the first time you do something still gives the big burst of 110 science (100 for doing it + 10 for the biome you are in), but repetition in other biomes is much less significant than before, giving only 10 science instead of 110 again and again.

For example: you land on the Mun and take the temperature for the first time. Finally knowing the temperature of the Mun gets the science community interested, so 30 science comes from that. But you were also in the midlands specifically, so this one experiment is worth 5 science more, for a total of 35. You move over to the lowlands and take more temperature readings. Everyone already knows the temperature of the Mun, but not the Lowlands, so this is a fairly minor discovery and will only provide the 5 biome science as the 30 science for doing a temperature reading landed has already been achieved. But if you move to the poles temperature gets a little more interesting, so maybe that’s worth 10 science instead. Still, each biome individually would not be as major as the very first experiment. But despite that, if you visit all biomes, that could easily be an extra 50 science you wouldn’t get otherwise.

I think that could still encourage exploring the biomes without making the biome grind too rewarding, which means you have to actually do something outside of Kerbin before you get the best stuff there is.

Edited by pschlik
Link to comment
Share on other sites

26 minutes ago, pschlik said:

One thing that’s always bugged me about science is the whole biome thing, and I think that’s part of the reason why you could unlock interstellar stuff before even leaving Kerbin’s SOI. There are just too many biomes in this area and that increases the amount of science you can get dramatically. They definitely need a nerf, regardless of what science is used for.

I get what the biomes are trying to do, encouraging exploration and all, but I don’t think the sheer amount of science each biome can provide makes sense. I think biomes should be more the icing on the cake of science; they add some extra information with repeated experiments, but it’s just doing the experiment the first time that really counts. To achieve this, that experiment which gives you that 100 science could be separated from biomes, and instead of each biome giving you another 100 science, it would only be like 10. Maybe with some bias so more interesting biomes give more bonus science. This way, the first time you do something still gives the big burst of 110 science (100 for doing it + 10 for the biome you are in), but repetition in other biomes is much less significant than before, giving only 10 science instead of 110 again and again.

For example: you land on the Mun and take the temperature for the first time. Finally knowing the temperature of the Mun gets the science community interested, so 30 science comes from that. But you were also in the midlands specifically, so this one experiment is worth 5 science more, for a total of 35. You move over to the lowlands and take more temperature readings. Everyone already knows the temperature of the Mun, but not the Lowlands, so this is a fairly minor discovery and will only provide the 5 biome science as the 30 science for doing a temperature reading landed has already been achieved. But if you move to the poles temperature gets a little more interesting, so maybe that’s worth 10 science instead. Still, each biome individually would not be as major as the very first experiment. But despite that, if you visit all biomes, that could easily be an extra 50 science you wouldn’t get otherwise.

I think that could still encourage exploring the biomes without making the biome grind too rewarding, which means you have to actually do something outside of Kerbin before you get the best stuff there is.

IMO it would be great for science to work the same, but gatekeeping not being on upgrading the science lab with cash but making "new discoveries", around planets/moons etc. That way, if you'd grinded 100% science on Kerbin, you'd still get the science points, but to get metallic hydrogen/HE2 fusion drives etc, you'd need to reach mun/duna/jool/eelo to unlock that "tier" of the tech tree.

Link to comment
Share on other sites

1 hour ago, Technical Ben said:

IMO it would be great for science to work the same, but gatekeeping not being on upgrading the science lab with cash but making "new discoveries", around planets/moons etc. That way, if you'd grinded 100% science on Kerbin, you'd still get the science points, but to get metallic hydrogen/HE2 fusion drives etc, you'd need to reach mun/duna/jool/eelo to unlock that "tier" of the tech tree.

Personally, I don’t like such arbitrary gates to progression. It doesn’t seem right to force you to do things in a rigid order in a sandbox simulation, where you should be free to choose whatever path you want. And that’s why I simply suggest making science harder to grind for-it slows things down without directly saying “nope, no more tech for you.” If someone wants to grind Kerbin/Mun/Minmus for the science to get fusion drives, so be it. But I want it to at least be hard to do that. Not outright impossible, but that is what an arbitrary restriction would do...make it impossible.

Regardless, this is also the game where arbitrary milestones determine colony growth (I do not like the ‘baby boom’ idea) so I guess I can’t expect many open gates. They have said that the ‘Progression Mode’ thing will reward exploration (science mode half) [and completing contracts as the career mode half] which might just mean gating tech behind going places.

Link to comment
Share on other sites

11 minutes ago, pschlik said:

Personally, I don’t like such arbitrary gates to progression.

Well, there are in a way that the tech tree is arbitrary.

My suggestion didn't mean a fixed sequence. It can be a tree too, and you can either chose which mission to start among the available missions, or just see what's available and chose how to progress. 

I edited my original post to clarify that point.

Edited by Challyss
Link to comment
Share on other sites

13 hours ago, GluttonyReaper said:

I suspect I'm in the minority here, but I actually quite like the science system as it exists now - it's the one part of career that actually works, as far as I'm concerned. The gamplay loop of "go place get science to get parts to go more places to get more science..." is simple, but it does feel quite rewarding, at least until the end of the tech tree. As it stands, it's still the only reason to go anywhere in KSP1, outside of "just 'cause".

I agree that the basic idea is fine. My problem with it has always been with how much clicking you need to do.

Killing kobolds has given you anything from better bartering abilities to cooking skill to increased disease resistance for literally my whole life, and in computer games for ALMOST that long. This is no different.

I'm not against a different way by any means, but I'm not against the basic construct we have currently. And whatever different way is proposed has to be better than what we have now (in my opinion of course) for me to get behind it.

I don't think a player should be required to go anywhere in particular to unlock tech. If the only way to get metallic Hydrogen engines is to aerobrake at Jool, then I'm going to hard pass on that scheme. Even if there are multiple ways to do it (For example aerobrake at Jool, get x km from the sun, set up a research outpost at Moho, or reach another star "the long way") then that's still restrictive and I don't want it. I want to be able to unlock the engine doing what *I* choose to do. I may have to do a lot of it, or a little of it depending on how hard it is, but I want the choice to be *mine* and not the game's. THAT is what KSP does now, and THAT is what I very much hope will continue in KSP2.

Edited by 5thHorseman
Link to comment
Share on other sites

Why not just keep the current science system and add milestone unlocks? It wouldn't be the first game that gave you the choice between earning some unit of progress and spending it on upgrades or venturing forth and finding them for themselves. You could also make it kind of like "Eureka's" in CIV VI; where a discovery/milestone reduces the science cost of a particular node (which to me would be more realistic). 

Link to comment
Share on other sites

17 hours ago, pschlik said:

Personally, I don’t like such arbitrary gates to progression. It doesn’t seem right to force you to do things in a rigid order in a sandbox simulation, where you should be free to choose whatever path you want. And that’s why I simply suggest making science harder to grind for-it slows things down without directly saying “nope, no more tech for you.” If someone wants to grind Kerbin/Mun/Minmus for the science to get fusion drives, so be it. But I want it to at least be hard to do that. Not outright impossible, but that is what an arbitrary restriction would do...make it impossible.

Regardless, this is also the game where arbitrary milestones determine colony growth (I do not like the ‘baby boom’ idea) so I guess I can’t expect many open gates. They have said that the ‘Progression Mode’ thing will reward exploration (science mode half) [and completing contracts as the career mode half] which might just mean gating tech behind going places.

As said. I'm not talking about making it arbitrary. But linking it to actual materials science/locations/launches/tests. Finding metallic hydrogen/checking the gravity to a gas giant/doing spectrometry with a telescope.

"Arbitrary" in the same way any system of not giving you 100% parts is also arbitrary. Because the game is not real life, we can only simulate/simplify it.

 

Quote

It doesn’t seem right to force you to do things in a rigid order in a sandbox simulation,

No, in the career mode. In the science mode. Not in the sandbox mode.

Quote

where you should be free to choose whatever path you want

There can always be multiple options (telescope for Jool, atmospheric test in Jool, science hab orbiting Jool) etc. If He2 or whatever is limited to Jool orbit, then no amount of "gating" science stops the need to get fuel from Jool before going interplanetary. If He2 can be made on the mun? Then you could launch an interplanetary on day 1 if you like... if you don't want future tech engines... but things like "nuclear fusion" don't fall out of an envelope by magic... it takes a LOT of testing and learning.

 

Quote

If someone wants to grind Kerbin/Mun/Minmus for the science to get fusion drives, so be it.

Why? If they want that, that is what sandbox is for. Career *is* a step by step gameplay system.

Edited by Technical Ben
Link to comment
Share on other sites

19 hours ago, Technical Ben said:

IMO it would be great for science to work the same, but gatekeeping not being on upgrading the science lab with cash but making "new discoveries", around planets/moons etc. That way, if you'd grinded 100% science on Kerbin, you'd still get the science points, but to get metallic hydrogen/HE2 fusion drives etc, you'd need to reach mun/duna/jool/eelo to unlock that "tier" of the tech tree.

This suggestion is logical, but isn't practical.

Remember, many people who download KSP2 will not be hardcore players. They will have no interest in going to Jool, they just want to crash some rockets onto a cheap colony on kerbin for fun science.

In terms of playability, the original system is more noob-friendly. As I say, leave the hardcore parts to the mods. The stock game could be a good bait to get your average Joe/Steve to play the game, arouse their interest in spaceflight, and as they learn to control mods and various aspects of spaceflight, give them the tools to make the game hardcore.:cool:

Or implement this as an option, but leave it turned off by default.

Edited by Xd the great
Link to comment
Share on other sites

2 minutes ago, Xd the great said:

This suggestion is logical, but isn't practical.

Remember, many people who download KSP2 will not be hardcore players. They will have no interest in going to Jool, they just want to crash some rockets onto a cheap colony on kerbin for fun science.

In terms of playability, the original system is more noob-friendly. As I say, leave the hardcore parts to the mods. The stock game could be a good bait to get your average Joe/Steve to play the game, arouse their interest in spaceflight, and as they learn to control mods and various aspects of spaceflight, give them the tools to make the game hardcore.:cool:

Wait. So if they just want to crash rockets, why are they playing Career?

(PS, and all this can be totally sidestepped by offering "singleplayer AI multiplayer opponents", which just gives you a second Kerbal space race company to play against, any of their "discoveries" become yours a little later, so even "noobs" get up the tech tree, because an AI opponent gets the milestones first, but unlocks them for you. :P ).

 

Edited by Technical Ben
Link to comment
Share on other sites

1 hour ago, Technical Ben said:

Wait. So if they just want to crash rockets, why are they playing Career?

That's a very fair point. 

On the other hand, the limitations of career early game can teach you a lot about simple things such as how important is multi stage rockets, why have fairing, TWR

If your first game is in sandbox mode, you have hundreds of parts to choose from. I remember being totally lost, then playing the tutorial for a few days, then starting a science play.

KSP is a hard game, and when you first start playing, you think you're going to make a space station in 30 minutes. Then.... well, then you go back to "how to leave the ground", learn about SRBs and variable thrust engines, decouplers, parachutes, staging and so one. 

To sum up I think a seasoned player wouldn't find too limiting to unlock through milestones, while new players would benefit a lot in term of learning how to do useful things. 

Link to comment
Share on other sites

1 minute ago, Challyss said:

That's a very fair point. 

On the other hand, the limitations of career early game can teach you a lot about simple things such as how important is multi stage rockets, why have fairing, TWR

If your first game is in sandbox mode, you have hundreds of parts to choose from. I remember being totally lost, then playing the tutorial for a few days, then starting a science play.

KSP is a hard game, and when you first start playing, you think you're going to make a space station in 30 minutes. Then.... well, then you go back to "how to leave the ground", learn about SRBs and variable thrust engines, decouplers, parachutes, staging and so one. 

To sum up I think a seasoned player wouldn't find too limiting to unlock through milestones, while new players would benefit a lot in term of learning how to do useful things. 

Seems about so. I'd also suggest to StarTheory, on having the option to start "early/mid/late" like you can in say CIV games or Simcity games. Then seasoned players could start with everything "unlocked", instead of starting pre-space age. I mean, a LOT of games go "are you sure you want to skip the tutorial" when you start. And to be fair, with saves/replays/reloads there is no real punishment for new players who just want to learn by failing. The game could always pop up a little "you crashed 5x in a row, do you wish to try the flying tutorial/orbital tutorial/awesome mun lander tutorial" etc. :P

 

Link to comment
Share on other sites

2 hours ago, Technical Ben said:

 I mean, a LOT of games go "are you sure you want to skip the tutorial" when you start. And to be fair, with saves/replays/reloads there is no real punishment for new players who just want to learn by failing. The game could always pop up a little "you crashed 5x in a row, do you wish to try the flying tutorial/orbital tutorial/awesome mun lander tutorial" etc. :P

 

Yeah, remind new players to play tutorials first, please. This will help keep noobs.

And someone please make a mechjeb mod/stock autopilot ASAP.

Link to comment
Share on other sites

×
×
  • Create New...