Science needs to actually be science.

[Whirligig Girl]

This is essentially my response to THIS, and the ideas in this thread are somewhat based upon the reddit post.

As we all know, the science system is somewhat flawed. I've never really found it extremely tedious (because I care more about going to more places than gaining science), but it's not that fun. It's often something I overlook when building crafts later in career mode (Because I have enough parts it starts to feel like sandbox mode in my head).

So here's a bold idea for a rework to the science system. And it makes science actually be science.

First off, atmosphere science needs to be situation-specific. No cold goo at re-entry. Different speeds at each altitude yields different results.

The current Science currency is replaced with Knowledge. Knowledge is gained from Crew Reports, EVA Reports, Mystery Goo, and perhaps Camera Pictures. Knowledge gain is automatically changed to Reputation as it is released to the public and they see the awesome pictures/results/cool-stuff that KSP has discovered. How does it interact with the Tech Tree?

Well, the Tech Tree is now split into Seven different separate trees. Command, Propulsion, Control, Structural, Aerodynamic, Utility, and Scientific. To advance in any technology, you have to actually do sciencey things to parts in these categories. Researching each category is done differently, but one thing is constant: All categories except for the scientific category can be somewhat researched whenever a part fails (explodes). IRL, when a part fails in a certain situation, it gives valuable information as to how to improve it. Part tests can also give plenty of research into those subjects. For instance, launching rockets will research propulsion, and flying at high altitudes with a basic jet will add research into aerodynamics. You can also simply divert Funding directly into R&D for ground testing. And remember, Knowledge is directly connecting to Funds.

What about experiments? In the Scientific Category in the Tech Tree, there's several separate paths which represent the different types of experiments. Knowledge isn't the only thing gained from some experiments. Using Barometers, Thermometers, Gravioli-Detectors, and Seismic Accellerometers will fill out the in-game planet info tab. Materials Science and Goo research and things of that nature will help to develop structural parts, and help to develop MORE science experiments. Sample studies are unlocked in the tech tree, including Atmosphere Sample, Surface Sample, Liquid Sample, and Rock Samples. These help develop exploration-type tech, such as rovers. Finally, Crew and EVA reports can help to develop Command type systems.

And what about Reputation? Shouldn't it be meaningful? Well, negative reputation causes a slow decrease in funding over time. (Don't timewarp too much!) Positive reputation causes a slow gain in funding that tapers off at a certain point. Not doing new missions or doing cool new things will also eventually lower reputation. (Staying in LKO for decades (or in Kerbal Timescale, months/years), not doing anything and just timewarping for months to Eeloo or Jool. Not doing missions will only make reputation go as low as 0, not negative.

Note I'm not suggesting an implementation exactly like this, but something like it. What do you guys think? Would this be more fun?

[Laie]

My modest ship to Minmus consisted of a Mk1 pod on top of some fuel and an engine, on which 4 detachable "science blocks" were strapped. A material Bay, some Mystery Goo and a thermometer, each attached on a radial decoupler. Here's a summary of the mission:

Orbit, transfer, land on Minmus

Right click on the pod, click crew report, click keep report

Click EVA, right click Kerbal, Click EVA Report, click keep

Right click on the pod, click take data, F to board

Right click Goo Cannister, click observe Mystery Goo, click keep

right click Science Jr, click observe material bay, click keep

Right click thermometer, click log temperature, click keep

Click EVA again, bring my Kerbal on the surface

Right click on my Kerbal, click EVA Report, click keep, click Surface Sample, click keep, click Plant Flag

Go next to the science modules, right click on the Goo, click take data, click yes, right click Science Jr, click take data, clic yes, right click thermometer, clic take data

Get back in the pod, change biome

Every 2 biomes, jettison a pair of science blocks

Repeat. This is so much fun ! 33 clicks for only 3 science modules, to repeat 4 times.

Well, ridding us of that confirmation dialog on Goo and materials bay would already help.

Basic intelligence for Kerbals ("collect science from all instruments in range") would help a lot.

There have been entirely different proposals, like actually having to set up a seismic experiment by placing sensors a couple hundred meters apart from each other and then crashing something on the Mun. You know, actually *doing* science rather then just clicking instruments.

Personally, I think the approach of "doing" science will become just as grindy when you have to do it twelve times over. I'd settle for a interface that is considerably less clicky. We'll get not only piloting skills, but engineering and science as well. I have great hopes that the latter will do more than just boost science points.

[Tuareg]

well, i could be the "i've already told this when they have announced science" guy and... well, i've already told this. dont wait anything intuitive from squad. they know it well they have already harvested most of the money from this game so they just hint some crap (like the new refueling) claim that the game is complete and abandon it. i follow this game 2 years ago and bought 1.5 years ago but it knows nothing more than in those time. squad had a good idea, they have built the base but they dont want to invest into it any more. it progresses like a garage project and it will end exactly like one.

[Master Tao]

We'll have to see how SQUAD updates the Science system once KSP is in Beta. So many features, especially aerodynamics, were added just to make sure all the parts functioned together. Many of them should be updated before the final release.

[Whirligig Girl]

well, i could be the "i've already told this when they have announced science" guy and... well, i've already told this. dont wait anything intuitive from squad. they know it well they have already harvested most of the money from this game so they just hint some crap (like the new refueling) claim that the game is complete and abandon it. i follow this game 2 years ago and bought 1.5 years ago but it knows nothing more than in those time. squad had a good idea, they have built the base but they dont want to invest into it any more. it progresses like a garage project and it will end exactly like one.

They're not abandoning the game after 1.0. KSP just exits early access then. They're not in it soley to make money, they're in it to make a good game. And they haven't harvested all of the money yet either. Sales are continuing to progress higher and higher for profits every day, more people are buying KSP. Heck, they even put KSP on sale practically every other week! You have to remember that this game is something a person and his co-workers are making, and has put years of continuous effort into. They're working extremely hard to get this update out, and have been working hard on updates since 2011. They're not even saying that it's complete. They're just saying that it will have all of the things in it that it needs to have to be the Kerbal Space Program, in it's most basic form. Beta is game balance and improving/rebuilding features so that they are no longer placeholders. It's making it totally stable. 1.0 will be when the game is a complete game, not when the development is complete. They will continue to develop KSP long after 1.0.

[woodensen]

well, i could be the "i've already told this when they have announced science" guy and... well, i've already told this. dont wait anything intuitive from squad. they know it well they have already harvested most of the money from this game so they just hint some crap (like the new refueling) claim that the game is complete and abandon it. i follow this game 2 years ago and bought 1.5 years ago but it knows nothing more than in those time. squad had a good idea, they have built the base but they dont want to invest into it any more. it progresses like a garage project and it will end exactly like one.

I sincerely hope you are wrong though. I've already put too much hope to SQUAD to make this game as spectacular as minecraft.

And yeah, im new at this game and the first glaring flaw that i see very clear is science and administration building. Except, ive never bothered to do all of these clickfest because i can just grind simple "part testing" contract and make 2.5k science out of it, which is a problem too.

Tried hard to raise my point in here, and hopefully the devs see it and consider rebalancing it.

[Ravenchant]

Maybe seven different branches are a bit too much. But I think KSP would profit from making a few features of KerbalEdu part of the main game.

[Tw1]

I have similar thoughts about the science mechanic. It doesn't inspire, it doesn't feel like doing real science. The split the tree thing does make sense, IDk if you could still make it so you get all the parts you need to operate different vessels in a logical order. But that really comes down to the details of how it's balanced, and set out.

My own thoughts were more derived around how science doesn't really encourage you to run a space program with any complexity, doesn't make surface exploration fun, and stuff like that.The tech tree, I'm not fussed about.

I had a similar concept of Knowledge, but separated that from points, those became tech points:

I still think science as points you collect was a bad idea. Once you've unlocked everything, science points have no purpose of their own. Collecting the points plus seeing a little flavour text does not provide an experience. Plus, a set up where points must deplete limits the potential role of stations, probes and bases,* and gives a large element of the game a finite lifespan.

Doing science could be modified, made into three (or more like two) subsystems, to make this critical part of the space exploration experience a lot more engaging, and rewarding.

Solar System Knowledge- Science where you collect Data to slowly build up an accurate picture of each planet, maps, atmosphere, mineralogy, etc, and the solar system, magnetic fields, radiation, solar wind, etc, which then could be accessed in the Editor to inform designs. Data is collected much like "Science" is now, through transmitting, and returning. It would be shown in the vessel recovery dialogue box in a similar way.

Research Levels, where you deploy probes, stations, bases, etc, to do on going experiments. This would be a point scoring system, but rather than a tally, these would be levels which rise and fall. Getting, and then keeping Research Levels high would be a boost to Reputation, (and through that, a boost to your budget, and available contracts).

These Levels would be dependant on how much your space program is currently contributing to science through on going experiments, on going monitoring (think solar wind measuring satellites, etc,), as well as the more instantaneous Data collection.

Ongoing Experiments would be done in a science lab. These would not go forever- some would have limited lives, or need resupplying. Different ones would be available, and cost money to run. The cost would offset one advantage- the penalty for having multiple experiments in the one place would be less steep, letting you build big research complexes. It is possible having certain science parts on your ship would be required to do some experiments though.

Monitoring equipment would be cameras, magnetometers, the thermometer, that sort of thing. These could break down, or need to be upgraded (technically, this would just be a replacement-upgrade would be the in-game reason for why the equipment is no longer contributing as much research) Having multiple monitoring systems on the same ship generally, would not be useful. To get more research happening, they should be in different biomes/orbits. Research will be higher when you first arrive at a new planet, and fall with time, but never become insignificant.

Monitoring would happen in the background, all the player needs to is activate them, land them, or put them on the right trajectory, leave them.

The parts used for the scientific Monitoring and Data collection would overlap where logical.

Research Levels could be subdivided into different types- Physics, Chemistry, Biology, etc, as well as the solar system Knowledge. However, unlike the others, the Knowledge one wouldn't rise and fall, only rise-providing a permanent boost.

Getting these to their full levels should be hard. Admin building strategies could be used to boost them.

Lastly- Experiment Contracts- Much the same as we have now, but also including contracts for experiments that aren't possible under the other systems. E.g- transmitting a radio signal while on the other side of the Sun to Kerbin to test Albert kermanstine's theory of relativity-

Where appropriate, the results from these would contribute to Knowledge, and temporarily boost the appropriate Research Level.

How would the tech tree fit into this?

The current science point thing works fine for unlocking that tree. It could stay the same, but the points renamed "Tech Points" or something like that.

They'd only be shown in relevant places.

What about asteroids?

Contribute to knowledge until you've found all types. Grabbing and capture could give boosts to Research Level. Something that might be useful for a contract.

TL,DR version:

I'd break science into three things,

Knowledge- building up a picture of the solar system.

Research- Ongoing testing and monitoring.

Experiment contracts- for other, special experiments.

Giving you something to complete, something to ongoingly work on, and tasks you can follow. That would offer a much greater experience than the current, superficial system. Tech points would be earnt as a separate thing, similar to Science points now.

After all, a fun experience is all a computer game can offer.

*It is true this is rectified to an extent by contracts, but they still don't offer much of a science experience, plus there's the odd discrepancy of being asked for data that is apparently not valuable...

Something with ideas in the original post of this thread, and the quoted section just above, could be a really awesome game.

Edited December 5, 2014 by Tw1

[technicalfool]

"Not doing new missions or doing cool new things will also eventually lower reputation."

*cough*Punish The Lazy*cough*

Actually not done any work on it in a while. But, feel free to use it and see how you like it.

[CaptainKipard]

This thread could potentially get very long, and very few people like reading walls of text. Could everyone in here maybe keep a running record of bullet points or tldrs for the purpose of writing a summary later? Also format your text a bit; use headings and lists and colours. Break things up a bit. You're more likely to grab and keep Squads attention that way.

Edited December 5, 2014 by Cpt. Kipard

[Scoundrel]

I too have pondered this question, and had several ideas similar to the ones GregoxMun and others have suggested. Then I realized that KSP has long since evolved from a game about building a rocket and getting them into orbit, into an actual exploration game with some groovy rocketry stuff. With that realization, when I look at KSP through that lens, I realize that KSP has virtually nothing for exploration. There's nothing to discover!

It is, in fact, the biggest piece of gameplay missing from this game, and Squad doesn't actually seem to realize this!

So, what does that have to do with science?

Science Points only role is to serve as a rather wince-worthy and clumsy interface with an advancement tree that doesn't actually represent any form of technological advancement. Once we've unlocked everything in the tech tree they no longer serve a purpose, and all we are left to do is pursue contracts, which don't actually *do* anything for late game gameplay, let alone exploration.

I do agree with parts of the proposal in that I think that science should be part of the exploration mechanics, and we should do some real science wherever possible, but that would require repurposing Science Points and figuring out what exploration mechanics best fit KSP... something that should be for another discussion, IMHO, after beta has come out and we can see what direction Squad intends to go with fleshing out their game.

Therefore, I humbly submit we get rid of science points and the tech tree and replace them with building advancement; let the buildings (and their advancement) determine which parts we unlock; and politely request some actual exploration gameplay that goes beyond planting a flag someplace.

[Wallygator]

I'm beginning to feel lately that the science dynamic should be replaced also. It seems to me that there are several types of information gathered as a result of conducting various missions - observations (reports), data (sensor readings), physical samples, and technical validation (testing). These different types of information might be tracked separately and then combined by the player (according to a formula) to allow unlocking of parts, improvements in reliability, increases in reputation, or perhaps unveiling additional targets for discovery missions. I haven't yet put my thoughts together yet on this, but let be enough to say that I support the ideas and discussion in this and other similar threads.

Just like in real life, We don't need reasons to explore, just better mechanics to make it worthwhile.

[Wanderfound]

Agree with the thread title, not sure if I agree with any of the proposed solutions so far. My thoughts:

* Moar scientific instruments. Lots more. Scientists use a vast diversity of gadgets, not just a couple, and this is a good way to motivate repeated missions and stretch out the tech tree in the endgame.

* Science tied to specific times and places. Except for the very early days of spaceflight, we don't send a mission up to just vaguely wave a thermometer around; we have specific targets. We did this in ye olden days as well; Cook observing the Transit of Venus etc. "Have instrument X in place at location Y at time Z".

* Science should actually involve some intelligence and discovery on the part of the player. Not just something you collect, but something you do. I have no idea of the best way to implement this in-game, however.

Perhaps subtle visual cues indicating science bonus points, with a slight random factor and transience involved in their location to prevent wiki spoilers. This might also encourage players to walk around a bit instead of just doing touch and go landings.

Edited December 6, 2014 by Wanderfound

[Pthigrivi]

I'm not certain I see how many of the suggestions in this thread really improve upon the current system. People are already not super thrilled by the part-test missions, I'm not sure making science more like them will improve things. I actually think the current system is quite a good foundation. To my mind all that's really missing are more involved experiments, and surface features at which to carry them out. I like the idea of placing surface sensors and impacting probes to do a seismic test. There could be others as well integrated with the new kerbal skills system. I also know surface features like geysers and volcanoes and resource deposits are planned, and planning a mission around investigating specific sites really would make things more exciting and challenging. The game has a long way to go still, they're just not there yet.

[Bill Phil]

I'm thinking more like instruments that you deploy in a place, but only a kerbal can deploy it. By deploy I mean "place down " on ground". So, an instrument package, like, say, a a collision seismometer. So if something then crashes, you get science. But it can only be placed and activated by a kerbal.

[peadar1987]

I still like my idea of interacting with some sort of terrain scatter for soil samples. You get given a picture of what you have to bring back to Kerbin, and when you land on the Mun, or Minmus, or whatever, you have to find that specific item for bonus science. That way you get a bit of a treasure hunt instead of just a click fest, and it adds a point to having a rover, as being able to cover more ground is a huge advantage if you're looking for something that may actually be several hundred metres away.

[Kegereneku]

Be all aware that complicating needlessly the system won't solve any problems one see with it.

You ought to be frustrated if the game was taking you for moron, but KSP wouldn't exist if that was the case.

Yes, the current Science-system is clicky, however it is fast and it can be balanced easily.

What I'm personally worried of is grindiness. Having a new science-system where it take more time to reach the same result wouldn't be an improvement over the current one regardless if someone consider it "more logical" or even "more intuitive". There's also the common discovery than someone's intuitive system look completely counter intuitive to other.

Said system should both be easily predictable and reasonably be neutral in what the player do. Each player will try to play along different strategy and would be frustrated if "equal" amount of work didn't gave equal amount of result.

Personal analysis :

- We (will) have A LOT of biomes to cover

- Not all of them should be explored to finish the tech-tree

- Some experiment can become quickly redundant / boring (ex : checking temperature everywhere)

- Manned and Unmanned mission should have the same appeal

Thus, an improvement could be to automatize the research gathering for a category of "basic science" while creating some sort of "complex science".

The idea is a science at two speed :

- Basic science give little but is done automatically if the sensor is activated (sending data is still manual), it would reduce click fest

- Complex science give lot of science but require more action from the player, it should give the fulfilling gameplay.

To be noted : Complex&fulfilling science would not be limited to manned mission, even if probes are cheaper to send all around.

[Tw1]

Then I realized that KSP has long since evolved from a game about building a rocket and getting them into orbit, into an actual exploration game with some groovy rocketry stuff. With that realization, when I look at KSP through that lens, I realize that KSP has virtually nothing for exploration. There's nothing to discover!

It is, in fact, the biggest piece of gameplay missing from this game, and Squad doesn't actually seem to realize this!

So much this. The same realisation inspired me to write the blog I quoted earlier.

We have good gameplay systems for building the spacecraft, and a good system for getting it there, but not a lot for once you've got to your destination.

Stock KSP isn't really suited for missions other than apollo style short visits.

You get given a picture of what you have to bring back to Kerbin, and when you land on the Mun, or Minmus, or whatever, you have to find that specific item for bonus science.

This could be cool, and is something that could be done through contracts. But I'd love to be able to cruise across Duna or Laythe without knowing what I'll find, and have sensors pinging way. Drill in one place and get one reading, and drill a bit further on, and get something slightly different. Exploration with nuance, that would be my ideal.

What I'm personally worried of is greediness. Having a new science-system where it take more time to reach the same result wouldn't be an improvement over the current one regardless if someone consider it "more logical" or even "more intuitive". There's also the common discovery than someone's intuitive system look completely counter intuitive to other.

To be noted : Complex&fulfilling science would not be limited to manned mission, even if probes are cheaper to send all around.

Agreed, but I don't think making it harder and more grindy is what anyone here has suggested so far. Some things that take time would be cool, but should be warpable. (Like how kethane mining is.)

The earn points to unlock parts system does its job well, (though it can be a bit grindy) but doesn't really live up to being called 'science'.

It's not really about the tech tree. I'd like there to be something beyond the tech tree. Ideally, something that isn't going to be used up.

Edited December 7, 2014 by Tw1

[Unabled]

I love these threads! Some really great ideas in here, it might not be a bad idea to try summarizing some of them.

... I realize that KSP has virtually nothing for exploration. There's nothing to discover!

It is, in fact, the biggest piece of gameplay missing from this game, and Squad doesn't actually seem to realize this!

So, what does that have to do with science?

Science Points only role is to serve as a rather wince-worthy and clumsy interface with an advancement tree that doesn't actually represent any form of technological advancement. Once we've unlocked everything in the tech tree they no longer serve a purpose, and all we are left to do is pursue contracts, which don't actually *do* anything for late game gameplay, let alone exploration.

I do agree with parts of the proposal in that I think that science should be part of the exploration mechanics, and we should do some real science wherever possible, but that would require repurposing Science Points and figuring out what exploration mechanics best fit KSP... something that should be for another discussion, IMHO, after beta has come out and we can see what direction Squad intends to go with fleshing out their game.

Therefore, I humbly submit we get rid of science points and the tech tree and replace them with building advancement; let the buildings (and their advancement) determine which parts we unlock; and politely request some actual exploration gameplay that goes beyond planting a flag someplace.

I really like this idea of linking the tech tree with building advancement/upgrades. In addition to the ideas Squad already came up with for what upgrading certain building would grant, having tech unlock with certain upgrades would possibly make more sense and simplify some of the things that will need to be unlocked/upgraded.

The way I look at it, they already have size tiers of parts, so why not link them to building tiers?

Start off with all your tier 1 buildings at KSC and upgrading the VAB gives you the upgrade up from 1.25 size rocket parts.

Upgrade the SPH to unlock the different Space Plane parts topping out with the new MK3 parts.

Upgrade the Science Labs to unlock maybe different power sources like solar panels, and batteries as well as more science experiment parts.

Makes sense to me, plus it gives players a bit more of an option to move down the tech tree as they see fit.

As far as Science/exploration goes? I think it can be a big problem to solve, and I hope that Squad focuses on it after .90.

Adding more interesting features/terrain/ whatever to planets and moons will be number one. Exploring flat terrain isn't going to be very interesting no matter what they do. The Mun is a great example of a more interesting place with the Rilles and Craters everywhere, but its just one step.

Adding more experiments that reflect the Kerbal Universe I think are very important: For example the Temperature and Gravitron Parts...You can activate them and watch as they change depending on where you are and what you are doing. You can actually observe change in your environment and get active feedback on that. For me, it is quite a different experience than being told what is happening to the Goo by a text box.

Now sure, you do get some different text based on where you are, but I think it would be better if it were a bit more 'visual'

Lastly, I think making the contracts a bit more specific might help some too. Something like: Kerbals Scientists are interested in the Munar Poles, and would like you to investigate and explore (Specific location/Biome). Even listing specific experiments to conduct there.

I think this could even provide a bit of a narrative to the experience as well if done right.

Speaking of narrative... After playing Buzz Aldrin's SPM, I realized how neat it would be to have a News Ticker of sorts to highlight the public's thoughts on your progress and accomplishments/failures.

I very much like the idea of having Reputation link to the Public's interest in your space program. (As Technicalfool Posted) Having Rep degrade due to inaction or a lack of new things being accomplished could be a way to make the world come to life so to speak.