How does everybody feel about the current set of stock experiments?

[mcwaffles2003]

18 minutes ago, James M said:

But you CAN score a goal WITHOUT the stick. It's just doesn't make a whole lot of sense to do it that way.

Enjoy  :

Spoiler

 

 

21 minutes ago, James M said:

Regardless I get what you're saying >.< Also. If KSP had a "real" goal, it wouldn't be KSP anymore. It'd be more like a Bethesda game. You have some main mission that you're deliberately ignoring at the time cause' there's so many other things you're always interested in but the game always comes back to remind you, "Hey, don't forget! You need to do this thing to progress the story. Make sure you do that!" And even when it doesn't remind you, there's always that annoying thought in the back of your mind like "I feel guilty cause' the devs put in the work to make this campaign thing, but I'd rather just not. Ugh." I guess my point is, I don't think KSP would be the same if it had an "Ending" or a "Goal". It would actually turn off players once they achieve it. 

Doesn't have to be that way. A goal doesn't necessitate a plot line, they can be more vague and arbitrary. What if the goal was "catalog the universe" and you filled out a catalog? Any travel to a new body is furthering that goal until you have explored the whole thing and that always seemed like intention of the game to me. Just there no nice resource to summarize your progress and point out the limits of your exploration thus far. I dont think this would be anywhere near what something like a bethesda game is like with hand holding and dragging you along a story.

[James M]

48 minutes ago, mcwaffles2003 said:

What if the goal was "catalog the universe" and you filled out a catalog? Any travel to a new body is furthering that goal until you have explored the whole thing and that always seemed like intention of the game to me. Just there no nice resource to summarize your progress and point out the limits of your exploration thus far.

So basically what we have right now, but with a couple progress bars saying, "This is what you haven't done yet."?

49 minutes ago, mcwaffles2003 said:

Enjoy

 Alright fine you got me. 

[mcwaffles2003]

1 hour ago, Brikoleur said:

• Split science into disciplines with different effects.
  • Engineering would let you buy more advanced spaceship parts.
  • Astronomy would let you discover exoplanets, asteroids, comets, and other space things to land on/interact with, also see "simulator" below.
  • Chemistry would let you buy more advanced resource extraction parts.
  • Construction would let you buy more advanced outpost parts. And so on.
  • Different types of activity would contribute to different disciplines of course.

The tech tree does this already a little by having separate branches:

Spoiler

But I agree there should be more to it with more nuances. As for new planet discovery I'm on the build telescope satellites to find them instead of opening a new horizon directly through a tech tree

 

1 hour ago, Brikoleur said:

• Add a simulator that lets you design, refine, and practice missions. The accuracy of the simulated bodies depends on how much you know about them: to start with you just know the basic physical characteristics (diameter, mass, orbital features), and they appear in the simulation as featureless spheres. As you gather more science about them, you'll be able to simulate the atmosphere, topology, and eventually surface features.

That... would be awesome

1 hour ago, Brikoleur said:

• Bind particular science experiments to resource extraction. Make surveying bodies for resources a real thing, not just running an orbital scan once.

I think a SCANSat touch to stock would be very welcome.

 

I can see a general path where:

1. Send probe to new planet to understand its general characteristics
2. Simulate new probe with discovered characteristics
3. New probe sent to planet to accurately discover and map out high density regions of resources
4. Simulate and run a crewed mission to establish resource gathering infrastructure in regions with desired resources
5. Establish colony
6. Repeat

4 minutes ago, James M said:

So basically what we have right now, but with a couple progress bars saying, "This is what you haven't done yet."?

No, something more useful that displays the data you have recovered thus far about planetary bodies and stuff so you have a reference to access while figuring out how to address interacting with that celestial body in the future. I have a more detailed and fleshed out explanation on my other thread "A re-purposing of science"

 

[James M]

8 minutes ago, mcwaffles2003 said:

No, something more useful that displays the data you have recovered thus far about planetary bodies and stuff so you have a reference to access while figuring out how to address interacting with that celestial body in the future. I have a more detailed and fleshed out explanation on my other thread "A re-purposing of science"

I know I know xD I responded in that thread myself not too long ago.

[Guest]

The problem is that science is part of a 3 currencies system with interchangeable conversion rates (strategies), if one of those currencies becomes useless very early in the game (Yes, Kerbin SOI is early game, especially with interstellar travel) the other two will become trivial to obtain too (science lab in orbit and converting science points to funds and reputation).

Basically when you reach Duna/Eve you are in sandbox mode.

Yes, they could add more parts that costs even more points, but you're not solving the problem, you're just moving it around. Having more costly nodes would either make the game too grindy or require more efficient ways of getting science points, making the starting experiments obsolete.

They should totally depart from the simple "science points as a currency to buy tech" to something else.

A random example could be having science as a prerequisite for tech research:

• You need 100 atmospheric science to unlock the "Atmospheric flight" tech tree (yes, let's use the "different kind of science idea")
• Science is not a currency, it's knowledge, you don't expend it to unlock nodes or exchange it for other currencies.
• When you reach 100 atmospheric science you have to do the research needed to unlock the nodes
• The research will consist in running determined lab experiments (options here are manned, unmanned, consuming resources, requiring life support, being in determined places/conditions) and technology demonstration missions (that should replace test part contracts).
• Every node in the tech tree has its own "knowledge" requirements and its own required missions / research / resources to unlock.

 

No, I'm not considering science as a mode on its own because I'm convinced it wouldn't make sense in KSP2, removing money makes little difference when your laboratories are in colonies that require a costant supply of resources.

Edited August 13, 2020 by Guest

[mcwaffles2003]

5 minutes ago, James M said:

I know I know xD I responded in that thread myself not too long ago.

You sure about that? O_o 

Cause I don't see it

3 minutes ago, Master39 said:

The problem is that science is part of a 3 currencies system with interchangeable conversion rates (strategies), if one of those currencies becomes useless very early in the game (Yes, Kerbin SOI is early game, especially with interstellar travel) the other two will become trivial to obtain too (science lab in orbit and converting science points to funds and reputation).

Basically when you reach Duna/Eve you are in sandbox mode.

Yes, they could add more parts that costs even more points, but you're not solving the problem, you're just moving it around. Having more costly nodes would either make the game too grindy or require more efficient ways of getting science points, making the starting experiments obsolete.

They should totally depart from the simple "science points as a currency to buy tech" to something else.

A random example could be having science as a prerequisite for tech research:

• You need 100 atmospheric science to unlock the "Atmospheric flight" tech tree (yes, let's use the "different kind of science idea")
• Science is not a currency, it's knowledge, you don't expend it to unlock nodes or exchange it for other currencies.
• When you reach 100 atmospheric science you have to do the research needed to unlock the nodes
• The research will consist in running determined lab experiments (options here are manned, unmanned, consuming resources, requiring life support, being in determined places/conditions) and technology demonstration missions (that should replace test part contracts).
• Every node in the tech tree has its own "knowledge" requirements and its own required missions / research / resources to unlock.

 

So instead of points to unlock tech, you get points as prerequisites for experiments and performing those experiments unlocks the tech?

[James M]

4 minutes ago, mcwaffles2003 said:

You sure about that? O_o 

Cause I don't see it

I'll have to go look. I know for a fact though that I read every comment in it while I was at work and could've sworn I also commented on my thoughts. I dunno. I know you've attempted this topic multiple times and I've read through quite a few of the threads.

[mcwaffles2003]

Just now, James M said:

I'll have to go look. I know for a fact though that I read every comment in it while I was at work and could've sworn I also commented on my thoughts. I dunno. I know you've attempted this topic multiple times and I've read through quite a few of the threads.

Honestly, this time I tried to talk about the experiments themselves specifically since it seemed no one wanted to talk about how science worked but now people are talking about science in a thread not about it... so yay lol

[James M]

2 minutes ago, mcwaffles2003 said:

Honestly, this time I tried to talk about the experiments themselves specifically since it seemed no one wanted to talk about how science worked but now people are talking about science in a thread not about it... so yay lol

Well to be fair the two go hand in hand. (Most of the time with the exception of Kerbal EVA reports and Surface Samples) Anyway, sorry about that. Back to experiments?  I just started using mods for the first time about 2 weeks ago and hands down ScanSat has been my favorite mod. It's provides all of the pertinent information I need to determine where I want to go on a celestial body and why. I feel if you're going to have an experiment be in the game it should serve a purpose. For example if magnetometer booms were used at different planetary bodies, one could progress research on planetary magnetospheres and eventually develop technology for active radiation shielding. Just as von Braun always wanted 

[mcwaffles2003]

10 minutes ago, James M said:

Well to be fair the two go hand in hand. (Most of the time with the exception of Kerbal EVA reports and Surface Samples) Anyway, sorry about that. Back to experiments?  I just started using mods for the first time about 2 weeks ago and hands down ScanSat has been my favorite mod. It's provides all of the pertinent information I need to determine where I want to go on a celestial body and why. I feel if you're going to have an experiment be in the game it should serve a purpose. For example if magnetometer booms were used at different planetary bodies, one could progress research on planetary magnetospheres and eventually develop technology for active radiation shielding. Just as von Braun always wanted 

Have you tried kerbalism yet? Cause it puts in a magnetosphere you can analyze.

Edited August 13, 2020 by mcwaffles2003

[boolybooly]

KSP was primarily designed a physics sim and not as a resource game. Its appeal lies in doing what NASA does. Resources are a good addition to the gameplay. You risk smothering the physics sim if you get  carried away with additional mechanics but where they contribute to the feeling of being like NASA I think its worth doing.

I think science is indispensable as a mission objective in KSP. The way that provides tangible results to the player is open for discussion. Though I suspect the mechanics for KSP2 are already specced, there is the potential for mods to adapt the gameplay and there is always KSP3 !

The nature of research in the real world which begat NASA is curiosity, which asks questions and experimentation, which answers them. People dont really do this because they are paid, but they get paid for it because it is valuable. Also you dont get better engines because you examined a moon rock in the real world and that is one of the shortcuts in the KSP model.

When science first became part of KSP I have to admit I felt it was not what I was expecting. It takes a resource model and applies it directly to research points which is a fun gamification because it gives you mission objectives and encourages exploration.

If we were going to be realists with KSP then IMHO you could set it up with a 4X style research directorate which uses funds to commission research which takes time to complete. You would use the mission test sytem to test less optimal prototype parts and accelerate the process of researching and perfecting them and personally I would also like to see iterative research as opposed to innovative research, where you can improve the specs of an existing part like improved engine ISP or lower weight or better thermal protection or even scale it up or down, which is what prototype testing would represent only to reach a preset design, thereafter you could improve a part spec with more iterative research.

(EDIT you could even have an engine testing microgame where the player adjusts intermix to develope an engine with particular thrust/ISP attributes which can then be added to the engines tab in the VAB like a regular engine.)

I also think the research tree could be structured with tech development in mind with lines of cockpits of different scales and crew capacity for example following on from each other in a distinct line of research as opposed to being bundled in with other parts. Each part type requiring iterative improvement to gain better  attributes like thermal resistance or bigger batteries etc.

It would also be nice to see technology research which would represent breakthroughs like lithium batteries or LED lights, applying across all relevant parts and improving charge capacity or  reduce weight or reduce power use for parts which store charge or create light respectively.

In that scenario, the current research resource model could be repurposed as location research for colonisation support and used to unlock the potential for establishing successful colonies in a given location, which in turn would allow the completion of more adventurous missions in that location which would earn money which could then be used on researching engineering parts.

2c

 

Edited August 13, 2020 by boolybooly

[Guest]

4 hours ago, mcwaffles2003 said:

So instead of points to unlock tech, you get points as prerequisites for experiments and performing those experiments unlocks the tech?

You get points with experiments, just like in KSP1, but they're not an active currency, just a "knowledge level indicator".

Each node has its "knowledge" requirements, when you meet those requirements you can start the research to unlock the node, the research consist in some of technology demonstration missions and/or some "research programs" to be run in laboratory module at specific conditions (ex. in 0g, on a planet/moon without atmosphere, at a specific minimum ambient pressure...).

That would give the player some missions to do in the short term (just like the contract system) and a reason to build and use early station/outposts while giving the same feel of real-world space technology research that has a lot of pure technology demonstration missions (just like the martian heli on its way to Mars right now) and research done on the ISS.

 

 

Edited August 13, 2020 by Guest

[linuxgurugamer]

10 hours ago, mcwaffles2003 said:

Not sure if you read my other thread, but in your opinion, as an experienced programmer, how difficult do you believe it would be to develop an in game wiki that unveils progressively dependent on what science has been gathered as I described in my other thread?

Asking to know if learning to code would be worth it or would the amount of coding required seem to be a bit much for a novice.

Not a novice task at all.  As it gets built, you will find new issues, interactions, etc which are unexpected.

But learning to code is always worth it, regardless of what you write.

[James M]

10 hours ago, boolybooly said:

If we were going to be realists with KSP then IMHO you could set it up with a 4X style research directorate which uses funds to commission research which takes time to complete. You would use the mission test sytem to test less optimal prototype parts and accelerate the process of researching and perfecting them and personally I would also like to see iterative research as opposed to innovative research, where you can improve the specs of an existing part like improved engine ISP or lower weight or better thermal protection or even scale it up or down, which is what prototype testing would represent only to reach a preset design, thereafter you could improve a part spec with more iterative research.

(EDIT you could even have an engine testing microgame where the player adjusts intermix to develope an engine with particular thrust/ISP attributes which can then be added to the engines tab in the VAB like a regular engine.)

I also think the research tree could be structured with tech development in mind with lines of cockpits of different scales and crew capacity for example following on from each other in a distinct line of research as opposed to being bundled in with other parts. Each part type requiring iterative improvement to gain better  attributes like thermal resistance or bigger batteries etc.

It would also be nice to see technology research which would represent breakthroughs like lithium batteries or LED lights, applying across all relevant parts and improving charge capacity or  reduce weight or reduce power use for parts which store charge or create light respectively.

Duuuude. Yes. 

Although I guess there would have to be a line drawn eventually so as not to complicate the development process to the point that players not familiar with all the technical jargon would have to open a wiki page every time they wanted to do something with their tech? 

That or you could just do tutorials. You know, those things new players only play like 30% of then rush into the game thinking they know what to do xD

Edited August 13, 2020 by James M

[James M]

10 hours ago, mcwaffles2003 said:

Have you tried kerbalism yet? Cause it puts in a magnetosphere you can analyze.

I got Kerbal Health and Snacks for now. I guess to me it's like Kerbalism Lite. Not as immediately life threatening and prepares me for my inevitable doom once I do finally download it. 

[mcwaffles2003]

Just now, James M said:

I got Kerbal Health and Snacks for now. I guess to me it's like Kerbalism Lite. Not as immediately life threatening and prepares me for my inevitable doom once I do finally download it. 

There is a kerbalism science only config as well

4 hours ago, linuxgurugamer said:

Not a novice task at all.  As it gets built, you will find new issues, interactions, etc which are unexpected.

But learning to code is always worth it, regardless of what you write.

I guess when the game comes around I'll see what they mean by "improved modability"

 

[Draco T stand-up guy]

On 8/12/2020 at 8:31 PM, Master39 said:

It's a pain to interact with every single experiment

That's why I run the ForScience mod. It really only works when you've got a scientist aboard but, then, that's we're paying them for.

 

[WelshSteW]

This is still an unfinished idea, so apologies if it gets a bit rambling, repeats what people have already said, or goes beyond the scope of what this thread was intended for. Also, I'm still fairly new to KSP, I only started playing it a few months ago, and I haven't yet left the Kerbin SOI with a manned vessel, so some of this may just be inexperience. I've also only ever played career mode.

 

It seems to me that science is only there as a way to restrict the tech available to you. Doing science doesn't really get you anything more than points to use to unlock the tree. You don't learn anything from it. I have no idea what the point of science is in a sandbox game.

What I would like is for science to actually result in knowledge. And for that knowledge to be needed to progress Or at least, make it easier to progress.

 

For my games I use a mod called Research Bodies. It hides celestial bodies until you find them. You can find them by scanning space with a telescope, or by accepting contracts to search space via the observatory - a new building it places in the KSC. The first option, actively looking for new planets with a telescope, is a pretty good idea. It's the active option, you actually have to point the telescope, and click to take a picture. It's not implemented superbly, it's a bit random, but it's along the lines of science resulting in knowledge. You scan the skies, and gain knowledge of new planets.

The second option, finding new planets via the observatory, costs you money. Which again, is along the lines of what's being said here by some people. You're investing in your scientists, and gaining knowledge. This is the passive option. You accept the contract, then sit back and wait.

 

So to progress, you have to put in a bit of work. The science is actually giving you knowledge.

 

If this concept was developed, you could get something really amazing. You'd have to put in work to be able to progress. And finding the planets would just be the first step. Once you've found them, you should have to set up experiments to track them, so you can calculate how far away they are, how big they are, what the orbits are like...

And you could have a system where the longer you've studied a planet, the more you know, and the more accurate the knowledge would be. Only just started studying Duna? Then you know it's further away from Kerbol than you, and you can guess that it's between 400km and 900km in diameter - for example. You wouldn't know what it looked like, what it's made of, whether it has moons.... until you study it more, or send probes etc.

Some knowledge could be such that it's only available via experiments performed on location. Or from long term experiments rather than single run, snapshot type experiments.

 

Also, and I may lose some of you here, but if the system you're playing in had a bit of randomness, you also wouldn't know how much delta-v you needed to reach the planets, orbit, and land. I like the suggestion above, from Brikoleur, about having a simulator to help plan missions. And if that simulator had only the knowledge you'd gained from your observations and experiments, it could give you an estimation of the delta-v required. Further simulations would be improved when more knowledge is added. But you wouldn't know for sure until you actually go to each planet, and each moon.

As I said above, I'm yet to leave Kerbin's SOI with a manned vessel - and I'm actually yet to orbit every planet with an unmanned vessel - but I know how much delta-v I need to land on every planet and moon already. I'm not going to learn anything from actually doing it. There's no risk to the Kerbals I send. I know exactly what's needed to get there and back. So what's the point, really? It's just a long version of going to Mun, or Minmus. (That's an exaggeration for the sake of making a point, sorry)

If the systems were randomised, you'd have the exhilaration of exploring the unknown. You wouldn't be able to go online and see what delta-v you needed, or what resources where there, or what the planets looked like. You'd actually have to go there with a vessel and see it for yourself. And there'd be risk in doing that. You might underestimate the requirements, and strand you Kerbals there. You might miss entirely, and not get an encounter. You may have thought the atmosphere was thinner than it was, or vice versa. You might have thought the surface was solid enough to land on, but when you get there you find out it isn't. You might have been relying on there being a resource you can mine to provide fuel for the return journey...

And of course you'd have the thrill of being the first person to see what the planet or moon actually looked like, up close. All of that is lost with a fixed system to play in - or at least it will be, once the game has been released for a few weeks.

[Jacke]

I dislike simplistic tech systems that industrialize R&D into grind out points, put points into machine, get tech back.

Real world tech improvements also need existing tech to be used to find out what works and what works better.  I want that in KSP.

The contract system and its missions give a mechanism to do this.  I've wanted to put out a mod that would do this for a while.

[afafsa]

Perhaps a split is necessary between science and engineering. Science would be driven by a desire to learn more about the places you're trying to visit, collecting atmospheric samples to learn atmospheric composition for instance. Whereas engineering would be driven by a desire to push limits. Setting altitude records, speed records, acceleration records, completing missions to places you've never been, etc. Contracts could include rewards that have rewards for both science and engineering. For instance, perform a supersonic retro-propulsion at an altitude of above 30km.

You could go further and tie requirements for particular parts to completing particular challenges rather than using an abstract point system, but it difficult to walk the line between simple and arbitrary on one hand and overly complicated on the other.

[catloaf]

Here's my opinion:

Tech progression should not be based on science. New tech should cost money and time. If it is improving existing tech rather than adding new tech it should cost less money and time. In addition the tech tree should be split into various smaller tech tress rather than one large web. Also, tech that iterates on other tech would require the previous tech. So for example, let's say that the hammer engine requires the flea and requires little time and money to unlock because it's just a big flea, but the reliant engine, the first liquid engine is in a new branch so it requires more money and time. Also, the tech tree should have few large nodes and many smaller, cheaper nodes. So the reliant, swivel, and terrier engines would all have their own nodes in the liquid engines part of the tech tree. The way science fits into this scheme is by unhiding certain nodes. Basically, certain nodes are hidden at the beginning of the game, and must be unhidden to be unlocked. So for example, at the beginning of the game you launch a probes with a Geiger counter into a high orbit of Kerbin, and it discovers the Van-Allen belts and deep space radiation. This knowledge then unhides the nose with radiation shielding, which then must be unlocked with money and time.

Science progression outside of tech progression should be discovering about other planets like others have discussed. I also like the idea that some science can't be transmitted, and transmitted science/knowledge is worth it's full value. A possible improvement would be that transmitted science is worth full value, but some experiments must be analyzed before becoming science. So let's say you wanted to learn about the rocks on Duna. You would have to send a drill and a lab. The smallest and cheapest lab would be uncrewed and could tell you how much of certain abundant materials there are, but nothing more, and resources that make up <1% of the total ore content are not detected. So with that lab you may learn that the ore is 80% silicate, 5% metallic, 6% ice, and not very radioactive. All other resources are not abundant enough to be detected. The next tier would be recovering the rocks and analyzing them at Kerbin. Analyzing at Kerbin shows you resources that make up 0.5% or more of the rocks and shows you more about each category, like what metals are in the rock and what percentages of those metals are in the rock. The final tier would be a crewed lab and would tell you everything about the rock. This system would remove another arbitrary feature of science in ksp and give labs an actual use besides being an exploit added by the devs on purpose.

 

[Cosmonaut15]

On 8/13/2020 at 12:18 PM, mcwaffles2003 said:

The tech tree does this already a little by having separate branches:

  Reveal hidden contents

But I agree there should be more to it with more nuances. As for new planet discovery I'm on the build telescope satellites to find them instead of opening a new horizon directly through a tech tree

 

That... would be awesome

I think a SCANSat touch to stock would be very welcome.

 

I can see a general path where:

1. Send probe to new planet to understand its general characteristics
2. Simulate new probe with discovered characteristics
3. New probe sent to planet to accurately discover and map out high density regions of resources
4. Simulate and run a crewed mission to establish resource gathering infrastructure in regions with desired resources
5. Establish colony
6. Repeat

No, something more useful that displays the data you have recovered thus far about planetary bodies and stuff so you have a reference to access while figuring out how to address interacting with that celestial body in the future. I have a more detailed and fleshed out explanation on my other thread "A re-purposing of science"

 

I'm totally agreed with the idea of exploration in the ScanSat mod. It is the basic concept of why we send probe before manned mission. It also fits well with the science branch that people are discussing about. The more you explored, the more understanding and unlock new toys for more advance exploration.

The engineering tech, on the other hand, should be researched and funded over time. Additionally, I'd love to see a game mechanic that let the tech advancement accumulate based on how much we use each prerequisite tech. Such as, you cannot unlock the research of a big booster unless you have already use a small booster to some extent. The accumulated score can be scaled similar to the existing planet multiplier, and even scaled with time so that the advanced tech cannot be unlocked too soon.

Edited August 19, 2020 by Cosmonaut15

[mcwaffles2003]

14 hours ago, Cosmonaut15 said:

The engineering tech, on the other hand, should be researched and funded over time. Additionally, I'd love to see a game mechanic that let the tech advancement accumulate based on how much we use each prerequisite tech. Such as, you cannot unlock the research of a big booster unless you have already use a small booster to some extent. The accumulated score can be scaled similar to the existing planet multiplier, and even scaled with time so that the advanced tech cannot be unlocked too soon.

I agree that money should be required for R&D to act as a money sink along with time. Equipment use should be required as well though I doubt it will even be considered. 

All that said I hope that experiments still play a role in R&D still and a culmination of these all would, in my opinion, lead to a more fulfilling design for tech advancement.

[Warrior_Me]

I think the basic system of spending science as a resource is fine. Apart from the player's natural curiosity it is the only thing incentivizing going to space in the first place and is at the center of the game's progression system as a whole. If progress could be made with just funds and time, the whole tech tree could be unlocked by just repeating the same contracts again and again.

I agree though that the current experiment system is a bit too simple. I think the different experiments need to have more specific requirements and challenges to yield results. That's why i think the Science Jr. is probably the most interesting stock experiment, it is very bulky and heavy compared to the other ones, so it is much harder to return it for full value.

Another example would be the impact experiments added by some mods that require you to bring a second craft along and crash it.

To then incentivize players to utilize all the experiments despite these challenges, maybe different experiments could contribute differently to each tech tree branch e.g. barometers would advance aerodynamics tech much more than seismic readings etc.

[SciMan]

I was writing a big post about how you should need science of some sort to get the ideas for new part concepts, and then it takes money and time to do the R&D to actually produce parts from those concepts, but now I'm thinking we don't even need a science points system.

The reason is simple. NASA doesn't send experiments to distant worlds to learn how to make better rocket engines. That's backwards. What NASA does is build better rocket engines so they can get science experiments to distant worlds.

So how do you get better rocket engines? You invest in science research RIGHT AT HOME. The reason for this is that the laws of physics, and therefore the results of your experiments, don't change based on where you are, and when you're designing a rocket engine the experiment results that matter are the ones from materials science. Barring some formerly undiscovered stable chemical element with different properties to anything we already know, what we're left with is figuring out what ratio to mix elements in to get metals that are strong at high temperatures and pressures. That right there is what gets you better rocket engines.

What you DO send experiments to other planets for is (for one) to learn if they were ever capable of supporting life, and to learn if they currently possess the resources needed to create a self-sustaining colony there.

In other words, you need science experiments to figure out where to put your colonies, not to get a better rocket engine to use.

Heck we already know that it should be possible to build a fusion powered rocket, the problem isn't the science, it's the ENGINEERING. Which for rockets is mostly plumbing. Engineering is what makes science results into something you can hold in your hand. And 99% of the science has been figured out long ago.