Science is pretty much stupid. Just get rid of it.

[Pulstar]

2 hours ago, hatterson said:

Also once we get colonies as theoretically if you wanted to put a surface colony on Duna, Eve, or Laythe, you'd want to land your stuff pretty close to attach it. I'd assume you'd need some sort of rover ability to move it around the surface a few hundred meters, but being able to land within 500m is a lot different than landing 20km away.

From the scarce info we have my impression is that colonies themselves will have a "colonial VAB" (think construction yard from command and conquer) as the first structure/core which will build the rest of the structures using some kind of input (resources are an obvious one, but I imagine we would be able to send pre-fabs from Kerbin?  I think something like that might have been implied or mentioned once, but who knows), so no actual attachment like we did in KSP1 with rovers, KAS cranes and docking ports would be required. If we can send pre-fabs I wouldn't be surprised if all it takes is landing within say 10 or 20km to be able to recover the payload, similar to how you can recover craft anywhere on Kerbin, that's something that could also be a difficulty setting. Then it would get added to the local resource pool as a prefab ready to deploy.

Anyway for me personally the drag is less of a problem doing targeted landings than the surface's rotation is when I plan maneuver nodes (I would love a predicted target location at impact/lithobraking time indicator). You can correct overshooting in atmosphere with wings, airbrakes, grid fins or parachute deployment to some degree so there's more wiggle room than one could initially think if you don't get the deorbit burn right.

Edited January 3 by Pulstar

[JoeSchmuckatelli]

Have not played for a week - but after a 'land at a place on the Mun' and gather science mission... docking crafts clones the science to both craft.

I'm guessing that once one ship 'turns in' the science both will be at zero - but won't know until the return trip is complete.

Currently, both craft = 144 Science in Samples and 80 Science in Data

Edit - Can confirm: once the ship that has science turns in science... the ship with the cloned data loses its science.

IOW - after separating the lander can from the return craft, both had equal science... after landing the return craft at Kerbin and turning in the science - the science on the Mun orbiting lander can has zero.

Edited January 6 by JoeSchmuckatelli

[Pulstar]

3 hours ago, JoeSchmuckatelli said:

Have not played for a week - but after a 'land at a place on the Mun' and gather science mission... docking crafts clones the science to both craft.

I'm guessing that once one ship 'turns in' the science both will be at zero - but won't know until the return trip is complete.

Currently, both craft = 144 Science in Samples and 80 Science in Data

Edit - Can confirm: once the ship that has science turns in science... the ship with the cloned data loses its science.

IOW - after separating the lander can from the return craft, both had equal science... after landing the return craft at Kerbin and turning in the science - the science on the Mun orbiting lander can has zero.

Yes, this also works with Kerbals boarding a different vessel in EVA. It's a very convenient system.

[hatterson]

7 hours ago, JoeSchmuckatelli said:

Have not played for a week - but after a 'land at a place on the Mun' and gather science mission... docking crafts clones the science to both craft.

I'm guessing that once one ship 'turns in' the science both will be at zero - but won't know until the return trip is complete.

Currently, both craft = 144 Science in Samples and 80 Science in Data

Edit - Can confirm: once the ship that has science turns in science... the ship with the cloned data loses its science.

IOW - after separating the lander can from the return craft, both had equal science... after landing the return craft at Kerbin and turning in the science - the science on the Mun orbiting lander can has zero.

When crafts dock their science data and samples are combined. When craft separate, each one gets a full copy of the science data/samples. Similar when a Kerbal boards a craft or leaves. He combined all his science with the craft when boarding and copies all science when he leaves.

The one thing you can't do right now (at least that I've seen) is manually transfer science from one craft to another or otherwise grab it with a Kerbal. So if you create a Mun rover that goes to all the Mun biomes collecting soil samples the only way to actually get that sample science is to return the rover itself to KSC or to have a seat on the rover so the Kerbal can walk up and officially "board" the vessel. You can also json edit to copy the science over, but obviously that's an out of game solution.

The process of copying data makes sense. Samples makes a little less sense since it's physical thing. You could obviously just cut the sample in half, but given the animation only shows them collecting a vial full, you couldn't do that too many times before running out. Either way, I think this is a decision they likely made to simplify the process. Instead of the manual tedium of copying or transferring science all over the place, you just have to worry about getting back to a recovery point and the game does the rest for you.

[JoeSchmuckatelli]

21 minutes ago, hatterson said:

this is a decision they likely made to simplify the process. Instead of the manual tedium of copying or transferring science all over the place, you just have to worry about getting back to a recovery point and the game does the rest for you.

Clearly - but that just adds to my impression that Science is merely notional in the game. 

[Pthigrivi]

1 hour ago, hatterson said:

The one thing you can't do right now (at least that I've seen) is manually transfer science from one craft to another or otherwise grab it with a Kerbal. So if you create a Mun rover that goes to all the Mun biomes collecting soil samples the only way to actually get that sample science is to return the rover itself to KSC or to have a seat on the rover so the Kerbal can walk up and officially "board" the vessel. You can also json edit to copy the science over, but obviously that's an out of game solution.

I suppose if you were trying to do a rover-based sample return you could also design it to dock with an ascent/return vehicle using clamp-o-tron jr's.  Again part of my problem is without biome maps in stock you can't pick an LZ thats in close proximity to 2 or 3 biomes so this kind of mission doesn't have much utility.

[Pthigrivi]

1 hour ago, JoeSchmuckatelli said:

Clearly - but that just adds to my impression that Science is merely notional in the game. 

What do you mean by notional? 

[hatterson]

30 minutes ago, Pthigrivi said:

What do you mean by notional? 

He means that science points are more like exploration points. There’s no in game connection between any experiment you do and any information or technology you gain.

Running a radiation scan at a location has no difference from collecting a surface sample, both just make the counter go up.

Personally I’m fine with that abstraction, but other want science to be more sciencey or to actually teach the player something.

[JoeSchmuckatelli]

1 hour ago, Pthigrivi said:

What do you mean by notional? 

 

45 minutes ago, hatterson said:

He means that science points are more like exploration points. There’s no in game connection between any experiment you do and any information or technology you gain.

Running a radiation scan at a location has no difference from collecting a surface sample, both just make the counter go up.

Personally I’m fine with that abstraction, but other want science to be more sciencey or to actually teach the player something.

Hatterson's answer is good.

 

If you want more:

Spoiler

Let me reiterate my complaint from way up thread: science in Science! feels more like getting a Mario Coin than actually doing anything.  They should literally just add the Mario Coin sound to every time you push the blue button.

And... my complaint does stem from wanting more out of the KSP2 experience.  Partly based on 'the promise' (or perhaps unique feature) of the original.  KSP did not only teach people about orbital mechanics.  It opened up space exploration in a novel and fun way that inspired a LOT of people.  That includes learning a lot about other worlds, beyond merely how to get there.  KSP2 has that opportunity - but it doesn't look like the Devs wish to follow up on that.

...

The more I think about it, I begin to realize that YouTube and Streaming has dominated the mindset of game development.  By which I mean if you were to categorize KSP videos, the vast, overwhelming majority are of the 'lets build a KrazyKraft and see what it can do' type rather than 'lets see what the rewarding single player experience is like.'   So, the game design in KSP2 (like Cities Skylines2) is optimized for the Sandbox / KrazyKraft crew rather than a single player experience.  I write this knowing that Science is currently in KSP2 simply to enable a progression system.  I like the progression system.

But there is the thing that made KSP special - and that's the educational aspect - that's what's missing.  I wrote it in my opening statement: the "I hate how clicky Science is" in KSP crew won the argument for Science! in KSP2.  There are certainly a lot of players who want a progression system and don't care about the science part - they just want to easily progress through the system and achieve... Sandbox.  That's what rushing through the progression system recreates.

I look at the game assets they've created: whole worlds, different biomes, things to discover.  Why not make discovering the things about the assets they've created both fun and educational?

 

[rjbvre]

1 hour ago, hatterson said:

He means that science points are more like exploration points. There’s no in game connection between any experiment you do and any information or technology you gain.

Running a radiation scan at a location has no difference from collecting a surface sample, both just make the counter go up.

Personally I’m fine with that abstraction, but other want science to be more sciencey or to actually teach the player something.

I'm fine with it too if that's the direction they go. I will add though that I've been reading most of the flavor text from experiments I run, and a lot of them hint toward or downright state what resources are present in different biomes. Orbital surveys also mention how difficult it might be to land there.

I get the impression science isn't just to unlock parts, but to transition players into the next steps on the roadmap: colonies and resource gathering. We're going to need info to decide where to build our colonies. It makes sense they're putting more dev time into the core gameplay than some niche science mechanics (I say this as someone who would enjoy many of them).

[JoeSchmuckatelli]

51 minutes ago, rjbvre said:

I get the impression science isn't just to unlock parts, but to transition players into the next steps on the roadmap: colonies and resource gathering. We're going to need info to decide where to build our colonies. It makes sense they're putting more dev time into the core gameplay than some niche science mechanics (I say this as someone who would enjoy many of them).

I'm getting the same impression.

The Colonies / Resource Management parts of the game seem to be what the team themselves is most interested in.  I'm also looking forward to those aspects of the game.

And I agree - they do need to keep developing the core aspects.  I'm just hoping that after all is said and done, Science! gets another pass, perhaps as a DLC (ala KSP).

[rjbvre]

1 hour ago, JoeSchmuckatelli said:

And I agree - they do need to keep developing the core aspects.  I'm just hoping that after all is said and done, Science! gets another pass, perhaps as a DLC (ala KSP).

Absolutely, that would be great. I'm hoping this is really just the beginning of science too.

I'm starting to wonder more and more how they will deal with the fog of war with the new star systems, and if it will ever be applied to parts of the Kerbal system.

 

[JoeSchmuckatelli]

21 minutes ago, rjbvre said:

fog

We had early suggestions (during development) that they only allow blobby images of the other worlds until players had visited them.  Would have been cool.  Probably a pain to code, though, given what little I've gleaned of how they've struggled with planets and performance.

[Vl3d]

After playing a lot, my conclusion is that the biggest innovation is the fact that the science parts force you to build unique craft adapted to their shape, size and weight.

[rjbvre]

1 hour ago, JoeSchmuckatelli said:

We had early suggestions (during development) that they only allow blobby images of the other worlds until players had visited them.  Would have been cool.  Probably a pain to code, though, given what little I've gleaned of how they've struggled with planets and performance.

Cool insight, I wouldn't have guessed that. I meant more in terms of when we learn basic info about them and how. Even just what bodies are there, orbits, volume/mass. There's gotta be some kind of progressive unveiling of those systems. It would be really disappointing if on day one you can just open up the map and see everything 

[Pulstar]

1 hour ago, JoeSchmuckatelli said:

We had early suggestions (during development) that they only allow blobby images of the other worlds until players had visited them.  Would have been cool.  Probably a pain to code, though, given what little I've gleaned of how they've struggled with planets and performance.

The complexity is orders of magnitude smaller compared to the planet terrain rendering itself. I imagine a lazy "correctly looking" implementation of an actual dynamic "blobby image" would need the game to render the planet mesh first to a texture, then draw that texture with a pilxelation+blur+other distortion shader applied. But I can't speculate how much of an issue fitting something like that into the current scene drawing code would be. I'm not that familiar with how to optimize rendering in Unity nor with the relevant KSP2 scene drawing code, all I do know from other engines is that using just one more shader at some point can break your optimizations/require a rewrite.

So it really depends if the scene rendering code already was written with something like this in mind or not. If not then the issue is not so much can it be rewritten and optimized to include it, but rather are there resources to spare at any given moment to do such a thing, including cleaning up any new bugs that the changes might cause.

 

 

[SciMan]

I'm gonna start off by saying that I haven't read this entire thread, so my apologies if I miss out on some critical details that have been gathered since I stopped reading, but I read a whole page that was mostly "the same stuff" and I have my forum settings showing me the maximum number of posts per page. So IMO this whole discussion has been going in circles, headed nowhere.
I think I have the key of what is missing from the science system that is irking some of you.

Flavor text, or lore dumps.
Text that serves no gameplay purpose (which is why it's not in yet, remember this is the FIRST iteration of the science mode), intended to convey information not critical to any of the gameplay loops in the game currently.
That's all that's missing. Just info not critical to the gameplay loops.
Heck, I can tell they're going to add that later, because some of the time when you do a science experiment it just spits a raw localization tag at you instead of a "generic you did the thing" message in your language of choice.

...and yet everyone in this thread seems to be eager to judge KSP2's science system by the same standards as the KSP1 science system.

Doesn't that seem at least a little bit unfair to KSP2? KSP2 barely learned to crawl, and you're already throwing hurdles and high-jumps at it.

My whole point is this: Give it time, this is only the first of many major updates to KSP2.

Even the version number of KSP2 tells a story. Version 0.2.x.x. If it was version 0.8.x.x or higher, I'd be right there with you complaining, because there wouldn't be much time left to change it and something would be clearly wrong with the core of the game.
But it's not. It's real early in the development of this video game. The fact that we're even able to have anything to play and give feedback to at all is incredible.

The gameplay loop itself? I love it. I don't have to click more than once to run all the experiments, I always get the full value of a science experiment, and they always transmit entirely or not at all. Perfection.

The only part I don't like about it is a game bug: ALL experiments that take time to complete will pause upon crossing a biome border, even the ones that are very pointedly NOT biome specific (such as orbital lab results).

What we have right now with the science system is a FRAMEWORK with a lot of PLACEHOLDER stuff in it. But it's a fantastic skeleton for what is to come.
When it gets fleshed out, I don't see any reason we can't have just what people in this thread are looking for, plenty of extra data about the celestial bodies in question, and probably the kind of data that would be useful for planning out where to put colonies on the various celestial bodies as well.
You're all correct, all of that is "not in the game". But you're forgetting the operative word. That word is "Yet". As in all of that is "Not in the game.... yet".

[hatterson]

1 hour ago, SciMan said:

The gameplay loop itself? I love it. I don't have to click more than once to run all the experiments, I always get the full value of a science experiment, and they always transmit entirely or not at all. Perfection.

Note that this isn't accurate. Experiments which collect physical samples also collect data. The data can be transmitted, the samples must be returned. So an environmental survey may be worth 100 science points total, 60 of which is samples and 40 of which is data. If you transmit you get the 40 for the data. When you recover you get the 60 for the sample. If you recover without transmitting you get all 100 at once.

[Bej Kerman]

1 hour ago, SciMan said:

If it was version 0.8.x.x or higher, I'd be right there with you complaining, because there wouldn't be much time left to change it and something would be clearly wrong with the core of the game.
But it's not. It's real early in the development of this video game. The fact that we're even able to have anything to play and give feedback to at all is incredible.

That is EXACTLY why people are choosing to complain now, because by 0.8.x.x the developers may not have the time or resources to overhaul science. Waiting until we're nearly at release before criticising one of KSP's most fundamental features does not track logically.

1 hour ago, SciMan said:

Flavor text, or lore dumps.
Text that serves no gameplay purpose (which is why it's not in yet, remember this is the FIRST iteration of the science mode), intended to convey information not critical to any of the gameplay loops in the game currently.
That's all that's missing. Just info not critical to the gameplay loops.
Heck, I can tell they're going to add that later, because some of the time when you do a science experiment it just spits a raw localization tag at you instead of a "generic you did the thing" message in your language of choice.

Believe me, there's enough flavour text. But KSP 2 isn't a command-line game and there's a little bit more irking people than bits of text you say has little impact on gameplay.

1 hour ago, SciMan said:

Doesn't that seem at least a little bit unfair to KSP2? KSP2 barely learned to crawl, and you're already throwing hurdles and high-jumps at it.

KSP 2 doesnt exist in a vacuum. We're judging KSP 2 as a game that's meant to take what KSP 1 did right, do it better, and not do what KSP 1 did wrong E.G. science being generic and offering little gameplay value. There's nothing exciting to do, and the devs have implied things like SCANSAT are not gonna see stock implementations that would have tangible benefits besides an abstract points system and a fairly generic tech tree. The developers had over half a decade of feedback towards KSP 1 to base the development of KSP 2 on when development started, and people are right to judge it that way. We're over a decade into KSP as a franchise and somehow the developers thought science boiling down to this is still appropriate.

[Pthigrivi]

5 hours ago, SciMan said:

Text that serves no gameplay purpose (which is why it's not in yet, remember this is the FIRST iteration of the science mode), intended to convey information not critical to any of the gameplay loops in the game currently.

And I would argue while flavor text is nice adding text that has no gameplay purpose will, unsurprisingly, have no actual effect on gameplay. This is the problem with this entire ill-named thread. These are atmospherics, not actual mechanics. And I hear what folks are saying about science points being an abstraction but... duh? Even if you split science into a bunch of different flavors it's still an abstraction because KSP is still a game. I always find it a bit funny when folks complain about being reminded that they're playing a game while they're attaching rocket legos and sticking little cartoon frogs in them. Yes. You are playing a game. This is not real life. Most all games have abstract currencies so this shouldn't really be so shocking. "Money" is an abstract currency. "Ore" is an abstract currency. Abstract currencies are not the problem and making KSP a better game will require more than flavor text. It should be unveiling information that does have a gameplay purpose. If we're playing Battleship and I call out "B7" and you tell me "the USS Indianapolis was sunk in 1945 by a Japanese submarine en route from Guam to the Philippines after delivering Little Boy to Tinian" that might be interesting knowledge that's somewhat related to the genre of game we're playing, but I'd probably rather you just told me whether I hit your submarine or not. 

 

5 hours ago, SciMan said:

What we have right now with the science system is a FRAMEWORK with a lot of PLACEHOLDER stuff in it. But it's a fantastic skeleton for what is to come.

After getting a bit deeper into the progression "skeletal" is exactly the word I'd use for it. And you're absolutely right at this stage I don't mean this disparagingly. I think what we have is a very solid foundation for science to operate. Look at all the complaints we aren't seeing. We are are seeing very few complains about grind. We're not right-clicking half a dozen parts hundreds of times. We're not seeing players easily exploit and clear the tech tree in 10h before they're incentivized to go interplanetary. Players are successfully combining science-for-science-sake and explicit mission goals and the general result is solid pacing, new environments, and steady progress. Thats huge progress over KSP1. In architecture they tell you the best stairs are the stairs no one comments on exactly because they feel comfortable and natural. Folks put a lot of work into building stairs so that you won't even notice them. Thats the real success of this structure, that it all flows pretty seamlessly and without grind or confounding distractions. The hard part going forward will be putting more and more flesh on those bones without throwing the balance off. 

All that said there's a lot of room for flesh on those bones. Folks who have dug in will know there are a series of plot-driven missions that move very aggressively from one precision landing to another without ever really supplying players with the experience and tools they need to learn how to do plane changes or vacuum landings let alone precision landings. We don't have trajectories or landing markers visible in flight mode. We don't have an in-game transfer window calculator. We don't have trajectories factoring drag when we get to Duna so we can land right where we need to without a dozen reverts. We can't look at a body in map mode and see what biomes are available or where we've already been. These are all solvable problems with some clever UI, it's just not there yet.  My only real argument in this thread is that this kind of valuable flight information should be integrated with the science system itself so exploration is the process by which it is revealed. The difference between a fuzzy red ball and the full-res Duna we see isn't actually any different from a gameplay standpoint. There is zero actual gameplay value in visualizing this this distinction. Again, without actual gameplay information like "where are the biomes" or "where are the nice flat places to land" or "where are the POI" the distinction is cosmetic and not actually important to the game. 

 

Edited January 7 by Pthigrivi

[rjbvre]

4 hours ago, Pthigrivi said:

We can't look at a body in map mode and see what biomes are available or where we've already been. These are all solvable problems with some clever UI, it's just not there yet.  My only real argument in this thread is that this kind of valuable flight information should be integrated with the science system itself so exploration is the process by which it is revealed. The difference between a fuzzy red ball and the full-res Duna we see isn't actually any different from a gameplay standpoint. There is zero actual gameplay value in visualizing this this distinction. Again, without actual gameplay information like "where are the biomes" or "where are the nice flat places to land" or "where are the POI" the distinction is cosmetic and not actually important to the game. 

 

One of the devs stated goals is making it so you can just look at the bodies and intuit where and what the biomes are. It's not perfect, but it's much better than it was and I'm having no problem figuring them out from orbit. All that's really needed is the science checklist from ksp1 to keep track of your experiments: no doubt this will be added.

A lot of people are forgetting that a lot of important game design won't apply to us, it'll be for first time players (or for us veterans in the new star systems). Making it a "fuzzy ball" absolutely changes the gameplay. Imagine a first time player planning their first trip to the Jool system and wanting to land on Laythe. Sure we know that's it's primarily an ocean world with a substantial atmosphere, but a new player would have no clue (wiki articles notwithstanding). Currently they can just look at it in the tracking station and see all that without ever going there (or studying it in some other way). If it's a fuzzy ball at first without much data available they'll have to study it (go there and do science) to figure out how to plan their mission. They also get the reward of discovering all the cool details about the moon.

KSP is no doubt primarily a rocket building/flying game. At close second it's about exploration and discovery. Some of gamers' favorite experiences are the first time they get to explore a game world whether it's Subnautica, Skyrim, BOTW, or whatever you're personal favorite is. Sure discovering the Kerbal system might be wasted on us veterans, but that's still a huge part of the game for newcomers. And the gameplay decisions made for the starting system will apply to the new ones, which is where we can get our "playing for the first time" experience.

 

[MarcAbaddon]

I am also critical of KSP 2 (but also the KSP 1) implementation of Science, and I agree that you do not need to produce a solution in order to be allowed to criticize as some seem to suggest. 

Generally, I think criticism of KSP 2 science falls into two areas for me:

Theme

I don't get a strong feeling that I am doing actual science, it feels like pressing buttons. With surface samples you at least get a nice animation. Generally, I would argue that this is just slightly better in KSP 1 since you at least confronted you with some flavor texts, though that lost it's appeal without mods quite fast once you went biome hopping on your first moon. Things I could see that help here is:

• More descriptions
• Make science generate reports about other bodies you can look at, that can be informative. Extra points for graphs
• Have a feeling of time passing for some of the experiments 
• Give a better impression of what kind of experiments is actually performed

Gameplay

Science is a new gameplay mechanics, and those should be judged on how well they integrate and interact with the existing mechanics. For example, just adding pieces to chess wouldn't improve the game, unless their moves mesh well with the existing ones. Interactions are also especially rewarding if in the long run they have positive feedbacks loops. Consider Civ where you build cities, cities allow you to build things, which allows you to acquire more cities (either by settling or conquest).

So what kind of interactions does science have? The good thing is that the most essential interaction is already there and works: Science allows you to get to new places and getting to new places allows you to get more science.

But the problem I think here is that the interaction is a bit weak in one regard: you really just need to go to a place. There's no setup or any interactions with other game mechanics involved in actually conducting the experiment. To give an initial example of an experiment that requires some effort: how if you have to run an experiment on the Mun in the same place for 3 months and the experiment require a constant supply of electricity? That would require you to account for being on the dark side for some of the time, and not being able to simply slap two solar panels on.

One example of science done right

While KSP 1 was also quite weak here in my opinion, I think a good example was the SCANSAT mod. You needed to construct vessel with the required experiments (some bulky, requiring fairings), you needed power and you needed to get in a polar orbit for which otherwise there was little incentive in the base game.  It also took time and there was an incentive to use probes instead of Kerbals since that meant you did not have to return to Kerbal and you didn't have a busy Kerbal for a long time (given the supply was limited) if you managed the reward both looked scientific (different maps of the body) and was useful, e.g. to locate anomalies or resources.  That felt like meaningful science to me.

[hatterson]

2 hours ago, MarcAbaddon said:

I am also critical of KSP 2 (but also the KSP 1) implementation of Science, and I agree that you do not need to produce a solution in order to be allowed to criticize as some seem to suggest. 

Generally, I think criticism of KSP 2 science falls into two areas for me:

Theme

I don't get a strong feeling that I am doing actual science, it feels like pressing buttons. With surface samples you at least get a nice animation. Generally, I would argue that this is just slightly better in KSP 1 since you at least confronted you with some flavor texts, though that lost it's appeal without mods quite fast once you went biome hopping on your first moon. Things I could see that help here is:

• More descriptions
• Make science generate reports about other bodies you can look at, that can be informative. Extra points for graphs
• Have a feeling of time passing for some of the experiments 
• Give a better impression of what kind of experiments is actually performed

I think this ties into @JoeSchmuckatelli's stated idea of an in game Kerbolopedia that gets filled in. To me that's absolutely a valuable addition, but I also totally understand why it's not in this version. Almost all of the currently relevant data is decoration text and in version 0.2 of an early access game that naturally has less priority. The stuff that might not be filler text (e.g. "your scans show that the Mun's Craters have an unusually high concentration of unobtanium, might be a good idea to set up a mining outpost there") isn't in the game yet, so wouldn't make sense in a Kerbolopedia yet.

That also provides a place for the random stuff that I'm sure the design team has produced in house for atmospheric modeling (density charts, chemical makeups, etc.) to get put.  Not stuff related to colonies or resources, but a bunch of stuff that currently resides in the wiki or similar for KSP 1. Purely informational stuff, but still fun to have in game for lore/interest reasons. As mentioned before though, this obviously has a lower priority.

 

2 hours ago, MarcAbaddon said:

Gameplay

Science is a new gameplay mechanics, and those should be judged on how well they integrate and interact with the existing mechanics. For example, just adding pieces to chess wouldn't improve the game, unless their moves mesh well with the existing ones. Interactions are also especially rewarding if in the long run they have positive feedbacks loops. Consider Civ where you build cities, cities allow you to build things, which allows you to acquire more cities (either by settling or conquest).

So what kind of interactions does science have? The good thing is that the most essential interaction is already there and works: Science allows you to get to new places and getting to new places allows you to get more science.

But the problem I think here is that the interaction is a bit weak in one regard: you really just need to go to a place. There's no setup or any interactions with other game mechanics involved in actually conducting the experiment. To give an initial example of an experiment that requires some effort: how if you have to run an experiment on the Mun in the same place for 3 months and the experiment require a constant supply of electricity? That would require you to account for being on the dark side for some of the time, and not being able to simply slap two solar panels on.

One thing I want is longer term experiments. Give me a year long orbital survey that requires full power all the time so I have to design a space station that can house Kerbals, generate electricity via solar, and store it via battery for when on the dark side. Give me deployable science like the Breaking Ground DLC had that takes a long time to run or that requires me to crash stuff into the body to get impact results. In fairness, I totally understand not having these in the initial version of science and several of them tie well into the introduction of colonies as they're designed around long term use.

 

[Hanuman]

The issue that I am seeing is that with these “tiers” of science, it gets goofy and in a hurry.   Are there enough biomes on the Mun and Minmus to support  the purchase anything in tier 3?   Or even just fully fill in Tier 2?  Is it possible using every available biome on Kerbin itself to even make it to unlock a Tier 3 science experiment,  so you can at least revisit biomes?  Why?   What about players that enjoy making atmospheric craft more than rockets? 

Science should not work that way.   Science is not a dang quest reward, and thinking of and implementing it like that is a mistake.  You gather science by exploring the universe around you, not by completing your curated mission string.  Use your missions to push and test player abilities and skills sure, but science should not be following the same progression.  A player that has spent hundreds of hours in game charting and exploring every inch of Kerbin and the Mun shouldn’t be that far behind a different player that went to Duna in a tin can in 5 hours science wise.   Mission wise?  Sure, of course.   Tell your game story there, but not in R&D.   Otherwise you get something silly, like there not being enough science on your entire home planet to learn to build a truck.

[cubinator]

17 hours ago, rjbvre said:

Making it a "fuzzy ball" absolutely changes the gameplay. Imagine a first time player planning their first trip to the Jool system and wanting to land on Laythe. Sure we know that's it's primarily an ocean world with a substantial atmosphere, but a new player would have no clue (wiki articles notwithstanding). Currently they can just look at it in the tracking station and see all that without ever going there (or studying it in some other way). If it's a fuzzy ball at first without much data available they'll have to study it (go there and do science) to figure out how to plan their mission. They also get the reward of discovering all the cool details about the moon.

This is the way I think science should be in KSP2. It should involve actual discovery and provide useful information for the player to plan their next steps. I wrote a whole long post in suggestions & development that I think is related to this discussion. It describes a very different version of science than what we have now: 

 

A few examples of my ideas for science:

• Planet terrain and details would not be visible, at least to probes, until photos and RADAR/LIDAR surveys are performed over multiple orbits. The player chooses which sections to take images and scans of and must plan ahead if they want to scan a whole planet in detail.
• Perhaps there is a place for signal delay in this game, where the player records a sequence of commands to be sent to a probe instead of controlling it in real-time. If you haven't mapped the terrain of a planet, your probe will need to be able to measure its own height above the ground or be able to reach landing speed miles above the ground to ensure a good chance of a safe landing. Of course, things like terrain maps of planets at various levels of detail could be unlocked for easier difficulty modes or returning players already familiar with the planets who just want to fly around.
• Magnetometers orbiting bodies can map magnetic fields, and a network of solar observation satellites would be the only way to detect incoming solar storms ahead of arrival. Perhaps the "monuments" would appear as anomalies in maps generated by surveys. They should NOT appear as a mission or text box appearing out of nowhere telling you to go visit them.
• The relationship between the tech tree and the collection of science would be expanded beyond the simple currency system. The direction the player chooses to go in on the tech tree would be related to the missions they choose to perform, where they would create some sort of "directive" for the space program. The science they collect would influence the directive of the space program, for instance discovering a water resource on Duna could provide incentive to increase development of resource extraction and long-term crewed mission technologies, or discovering an exoplanet could incentivize development of larger space telescopes and instruments that can characterize planetary atmospheres.

Basically, I want science in KSP2 to function as sensors that gather data about the environment, not as currency dispensers that unlock new parts as a reward for traveling someplace new with limited technology. I think the place for "tech tree"-like progression lies more in the missions and the player's own directive to perform increasingly complex missions around the destinations of their choice, with science acting as an influencing force on those directives and mission goals but not directly acting as the sole source of the currency for unlocking parts.