How realistic do you want science to be?

[Ruinsage]

First off, I would post this under suggestions and discussion, but it somehow won't let me make a poll that way, so if it being here is an issue, move it to development instead.

I don't like the way science works at the moment. I get that the old way, where you just send a probe and gather all the available science by sending it back by repeated transmission wasn't optimal, but the current model doesn't really work either, at least not for me. It greatly incentive's landers, but at the cost of probes and stations, which are both completely useless at the moment. But all of this is just my opinion, and I want to know what the community wants. So, as far as I can see, these are the options:

1: Science is fine as it is

2: Probes can scan from orbit, which can be transmitted for full science value (Like the kethane mod).

3: Rovers can take surface samples, which can be processed by the rover for almost full science value (like the mars rover). Kerbals can bring back surface objects, like small rocks and the like, which are hard to find from orbit, but a part can be used to tell if you are near one (which would make rovers useful, as they are a fast way to get around the surface). These would be worth a lot of science.

4: Stations gather science over time (let me finish), by gathering science from what is left of the total science pool in a given sphere of influence. So if it's in orbit over kerbin, it would gather all the remaining science from all the experiments in the sphere of influence. So this would NOT be a source of infinite science, or even a separate source of science, instead it's a way to gather the science you have somehow left out, because you didn't bring the science junior in orbit just above kerbin. Of course you can simply set this up initially to gather all the science, but that would take a long time, and it would be much more efficient to simply leave it in the background while you do something else. (it should continue gathering science while on rails)

5: A mix of the above

[boomerdog2000]

I don't like the way science works at the moment. I get that the old way, where you just send a probe and gather all the available science by sending it back by repeated transmission wasn't optimal, but the current model doesn't really work either, at least not for me. It greatly incentive's landers, but at the cost of probes and stations, which are both completely useless at the moment.

I disagree. A small station with a science lab can make it so you only need one lander instead of one per biome, and probes are for use in places where you don't feel confident enough in your abilities to get a lander like Eve or Tylo or you don't feel you have the tech to get a manned landing there right now. Plus, odds are probes will be cheaper than a manned lander and can be useful for that little bit of extra science you need when you don't have the funding for a full mission.

[Alymon]

They should have it so the transmission of research isn't such a negative with some caveats. So maybe a straight transmission from Mun to Kerbin would net 10% of the science value that returning it would. But then you could have satellites that the transmission could relay through to boost the value (up to a limit). You could also have stations with labs that would provide a bigger boost (due to the lab), again up to a limit. Maybe there's a total cap on the percentage to still encourage returning vessels, but at least this method would encourage setting up a satellite network and stations to make gathering science easier.

[r4pt0r]

Here is what i would like:

-Satellites to map biomes so i can more effectively see biome boundaries and plan my missions better(but not necessarily give science in and of itself)

-More Science tools (Aerogel wafers for collecting spacedust, unmanned surface samplers, some sort of highmag Telescope, an Impactor and a reader to register the ejaculated matter; 2nd part could be built into probe bodies, a small sealed pod with plantlife in it to see how things grow in space, I could go on and on...) Obviously not all of those will ever be stock, but lets throw out ideas for more varied SCIENCE-ing.

-No science over time. i will spam it and timewarp if that works.

-Finish up the biomes on all bodies, add a few more science gaining tools, and increase the point requirements for the late research center nodes.

-Non-Physical experiments should be 100% return for transmitting; Add more Experiments that require a return trip. that would lead the player to try more return missions.

thoughts?

[KerbMav]

Playing stock or with mods there are still good reasons to use stations and probes (apart from "playing the game!" and "because we can!"):

Uses for stations:

- home port at Kerbin for a reusable interplanetary ship to refuel and re-stock on equipment like landers, probes etc.

- ladingport for surface to orbit transports to drop of resources and interplanetary transports to pick them up (Kethane)

Uses for probes:

- scouting missions to pick a good/flat landing zone

- get science from places hard to reach/get away from

- independent mapping (SCANsat), comm relays (RT2)

I agree that the science lab lacks another perk, as only Kerbin and its moons have biomes for now, the ability to make science equipment reusable is not worth much on the other planets, and even then you could just bring the sample home, as the lab needs to be manned anyway, so it will most likely be a return mission.

They should have it so the transmission of research isn't such a negative with some caveats. So maybe a straight transmission from Mun to Kerbin would net 10% of the science value that returning it would. But then you could have satellites that the transmission could relay through to boost the value (up to a limit). You could also have stations with labs that would provide a bigger boost (due to the lab), again up to a limit. Maybe there's a total cap on the percentage to still encourage returning vessels, but at least this method would encourage setting up a satellite network and stations to make gathering science easier.

There is no reason for data loss due to distance, the transmitting system can check for errors and the damaged part of the transmission can be sent again if needed - so maybe changes to data rate and energy requirements depending on distance to KSC would be more reasonable.

Giving an incentive to use something is always better and more rewarding than punishing the player for not doing so.

Maybe giving the lab the ability to evaluate the data - have an "experienced science officer" go over the astronauts (EVA-, crew-)report and sensor readings or just have the guys go through their field notes and write a proper report ... something along these lines - and add a small bonus before transmitting?

Relay stations could still be used to alleviate the complications from long range transmissions though.

Here is what i would like:

-No science over time. i will spam it and timewarp if that works. agreed

-Finish up the biomes on all bodies, add a few more science gaining tools, and increase the point requirements for the late research center nodes. agreed, but make them interesting, not just another thermometer clicky

-Non-Physical experiments should be 100% return for transmitting; Add more Experiments that require a return trip. that would lead the player to try more return missions. agreed, but these should be logical

The old logic of 100% transmittable data after a few repeated transmission might still be used for thermometer, barometer ... anything dealing with variable numbers.

Switch EVA and crew report: EVA being the "we got here", and the crew report being "this is what we saw" - especially in regards to orbiting above different biomes it makes more sense to have different reports for each biome from the pod using its instruments (e.g. cameras) then "I am in space in a suit next to my spaceship and I see trees ... now water ... mountains ... trees again ..."

Edited April 16, 2014 by KerbMav

[r4pt0r]

i agree KerbMav, it would be nice to stay inside the ship so i can read "Crew report in orbit over minmus slopes"

i dislike going to eva to know if im in a new biome or not

[Ruinsage]

While I'm glad to see that the vast majority of the poll (so far at least) agree with me in science right now needing an overhaul, i'm a bit puzzled to see people wanting more incentive to use landers that can return back to kerbin. Right now, landers are the only choice for gathering science, unless a return trip is almost impossible. The vast difference in bringing back all data, and transmitting it back makes it practically useless to do anything but bring it back, and that is simply neither realistic, nor fun (at least not to me). I like the design aspect of the game, so spending all my time designing landers, and literally nothing else, is boring. I want to make deepspace probes, manned mun bases, duna space stations and more. And right now none of those are useful, except for being interplanetary fuel depots (I say again, BORING).

I am not saying that manned lander missions should not be the most rewarding missions, they absolutely should, but right now they are more or less the only option, as everything else just provides so little, because the absolutely largest source of science is soil samples, and everything else give so little when transmitted.

Also, as it seems I have not been clear enough on the issue, my suggestion for the space station mechanic would not gather science from the surface, so you would still need landers.

[regex]

I like how those of us who think science is fine right now are obviously having a screaming fit in your poll.

Anyway, I think the best way to improve science is to allow for more complex and interesting experiments like the impactor from KSPI.

Science over time seems like a lazy "something for nothing" mechanic to me and I don't even really like your idea above. Basically anything that incentivizes me to not play the game is bad. I'm also not into probes being able to do what Kerbals do since Kerbals are harder to get places and you generally want to return them; there's an investment there ovver throwaway probes. Anyway, my opinion on the matter.

[Kerbonautical]

I'd like to see the science expanded upon really heavily. Fundamentally it is the ground works to a whole bunch of possibilities for the career mode but currently it's being used simply as an asset to unlock nodes in the skill tree. Obviously Space Stations should have significantly more use scientifically if we're to ever have a use for them, otherwise we may as well just build unmanned fuelling stations. There is a little way to expand upon it for career that I would like to see exploited however, for instance we could be tasked with a mission to perform particular experiments at a certain difficult to reach location perhaps even with a specified vehicle type.

For that to work to it's fullest though there are a lot of extra parts we could do with, for instance our rovers have no means of performing tests outside of the science jr, mystery goo and the 4 remainders, whereas with a few more parts and some creative design we could excavate ground samples, water ice samples etc. I'll be surprised if this sort of thing makes it into the stock game personally, but this is the route I would really like to see Squad take. I mean let's be realistic here, the game is never going to be for most casuals anyway due to it's inherit complexity.

[Syphus]

The lack of having to publish scientific data in a respectable peer reviewed journal really takes me right out of the game.

What I really want is not quite to have it changed yet. But I want there to be more things to do on and around planets. What are those? I am not sure.

[Ruinsage]

I like how those of us who think science is fine right now are obviously having a screaming fit in your poll.

Anyway, I think the best way to improve science is to allow for more complex and interesting experiments like the impactor from KSPI.

Science over time seems like a lazy "something for nothing" mechanic to me and I don't even really like your idea above. Basically anything that incentivizes me to not play the game is bad. I'm also not into probes being able to do what Kerbals do since Kerbals are harder to get places and you generally want to return them; there's an investment there ovver throwaway probes. Anyway, my opinion on the matter.

But space stations are not "something for nothing", and should absolutely not be in the game either. At some point or the other, keeping kerbal in space indefinitely is going to become expensive, and keeping the space station powered would require a massive solar array. There are lots of ways to ensure that building space statons are not the easiest way to gather science, but maybe the most interesting.

[cantab]

I think the current system is more or less along the right lines. There are things that can only be done by a Kerbal, which is right. There are things that give full return by transmission, and other things that need returning for best results, which is also right. Now which things are which can be fairly quibbled - I agree that stuff like thermometry and gravimetry ought not to incur a transmit penalty - but the general idea is right.

The issue is with the biomes, which make the science grindy. We need a way to get a message popup about a biome crossing so we're not resorting to clickspam to pick them up. I also think there should be diminshing returns to encourage going to new places rather than just spamming Mun and Minmus science. The simplest way to do that would be to make fewer experiments biome-sensitive; better I think would be diminishing returns so the first biome or two you hit are worth plenty and after that there's less in it.

"Auto-Science" would take the load of the player, but removes the feeling of involvement. With messages about biome crossings (and include stuff like SOI changes and low/high space changes) there's no need for the game to actually run the experiment for you, you know when it needs doing.

That said, a couple of experiments that require time to run would be good, rather than everything being click, done, return. I believe KSP Interstellar has a telescope that does something like that, though I think the stock game's version should offer finite science. There could be a requirement to keep the experiment fully powered throughout, or you have to restart, so there's an extra challenge of needing to think about your power supplies.

[KerbMav]

i'm a bit puzzled to see people wanting more incentive to use landers that can return back to kerbin. Right now, landers are the only choice for gathering science, unless a return trip is almost impossible. The vast difference in bringing back all data, and transmitting it back makes it practically useless to do anything but bring it back, and that is simply neither realistic, nor fun (at least not to me).

I like the design aspect of the game, so spending all my time designing landers, and literally nothing else, is boring. I want to make deepspace probes, manned mun bases, duna space stations and more. And right now none of those are useful, except for being interplanetary fuel depots

Just do so and build them?!

There are mods with parts that can grab a surface sample to do an automated rock and dust return mission.

It might be that KSP is a game that gets a lot of depth from modding, customizing the game to your liking, instead of a precooked product.

I am not saying that manned lander missions should not be the most rewarding missions, they absolutely should, but right now they are more or less the only option, as everything else just provides so little, because the absolutely largest source of science is soil samples, and everything else give so little when transmitted.

Also, as it seems I have not been clear enough on the issue, my suggestion for the space station mechanic would not gather science from the surface, so you would still need landers.

So, you do not want stations to generate science on their own, and not for siphoning science from the surface - what do you want them to do for you?

And we agree in so far that the game could use more options to do science.

[Baron_von_Fluffy]

I want actual research based science. Not "land here and come back" science. What I mean is, you learn new things about rocketry from experiencing how those things happen, you don't unlock a nuclear engine because of some dirt from the moon, get it?

unlock massive rocket engines by crew reportes and instrument readings on how they perform, by setting up orbital microgravity test labs to direct research points towards a specific topic, get it? it would be cool, bbecause then you could not only unlock engines, but potentially upgrade existing parts for better thrust, higher gimbal, better ISP.

edit: basically this would mean make the tech tree more like a tech tree, rocketry branch would contain instruments to assess the performance of rockets, orbit or extra-Kerbal labs could be given a single research branch, etc to develop "rocketry science" or "electrical science" -- landers still relevant, to develop other forms of science, just not for the same tech branches

Edited April 16, 2014 by Baron_von_Fluffy
revising it

[samiamthelaw]

Regarding realism, it makes sense that some experiments are much more valuable if you return them home. This is pretty well reflected with transmission efficiency, though I don't understand why things like temperature readings wouldn't transmit at 100% value.

Right now, landers are the only choice for gathering science, unless a return trip is almost impossible.

I disagree with this. For the ~2500 delta v it'd take (from LKO) to land on the Mun and return, you could instead send a probe to Eve or Duna or Dres, or with some decent maneuvering and aerobraking you could manage a swing by most of the Joolian moons. Maybe people do opt for repeated Mun missions, gathering samples from every biome,* but you can get a decent way up the tech tree without going that route.

* - The sheer amount of science available from the Mun is a balance issue. It might make sense to scale back some of the Mun experiments from per-biome to global, if the Mun is going to have quite so many biomes.

[Pecan]

At the moment career mode is almost only about doing science so you can get parts, which is somewhat the wrong way around (although better parts partly means the ability to get more science, of course, it's iterative) which is why I use mods or imagination to create non science-based programmes (sets of missions, within projects, within "why we choose to do these things"). Since it is very much still being worked on I'm happy to wait to see what Squad do with it, until then I stick to sandbox.

[Xavven]

Regarding realism, it makes sense that some experiments are much more valuable if you return them home. This is pretty well reflected with transmission efficiency, though I don't understand why things like temperature readings wouldn't transmit at 100% value.

* - The sheer amount of science available from the Mun is a balance issue. It might make sense to scale back some of the Mun experiments from per-biome to global, if the Mun is going to have quite so many biomes.

This is a great idea! Maybe Science Jr. and Goo should be once per surface. Reduce EVA near-space to once (not biome dependent) but allow EVA to be biome dependent on the surface. Surface samples should stay biome dependent.

[r4pt0r]

unlock massive rocket engines by crew reportes and instrument readings on how they perform, by setting up orbital microgravity test labs to direct research points towards a specific topic, get it? it would be cool, bbecause then you could not only unlock engines, but potentially upgrade existing parts for better thrust, higher gimbal, better ISP.

The "tech tree" is a means of getting new players to use base tier parts to understand basic orbital mechanics without overwhelming them with 100+ parts, so they can use more advanced designs effectively in future.

this allows the "casuals" (as someone called them previously) to get into the game and also learn.

also i disagree with "upgrading parts". all parts should have the same stats for everyone. this is why we all have the same in-game universe, so players can relate to each others games. there would be designs on the spaceport that require certain upgrades to get equal performance with the creator.

plus giving your rocket essentially a "commando pro" perk so it can burn longer on less fuel is silly.

[Tw1]

Your poll and your thread title are really asking different things. I was expecting discussion about how close to real science kerbal experiments should be. Instead, there is a list of possible experiments, some of which exist as mods, and/or may be added into the game in future.

[KerbMav]

i agree KerbMav, it would be nice to stay inside the ship so i can read "Crew report in orbit over minmus slopes"

i dislike going to eva to know if im in a new biome or not

Go to \Kerbal Space Program 23.5\GameData\Squad\Resources

open ScienceDefs.cfg

scroll to and edit these parts to look like this:

EXPERIMENT_DEFINITION
{
	id = crewReport
	title = Crew Report
	baseValue = 5
	scienceCap = 5
	dataScale = 1

	requireAtmosphere = False
	situationMask = 63
	biomeMask = 23

Will now give crew reports anywhere as usual but only differ by biome when landed, splashed, flying low and orbiting low.

EXPERIMENT_DEFINITION
{
	id = evaReport
	title = EVA Report
	baseValue = 8
	scienceCap = 8
	dataScale = 1

	requireAtmosphere = False
	situationMask = 63
	biomeMask = 3

Will now give EVA reports anywhere as usual but only differ by biome when landed and splashed.

"Problem" remains that crew reports cannot be stacked in the pod. For a crew report the pod functions just like the thermometer, it can hold only one result at a time. You can though pick up the report and store it in the pod, it then frees up the function of making a new report, but you will have to leave the capsule for that.

One more mod: Using ShipManifest you could place a part capable of storing science reports and transfer the data without leaving the pod.

[Mr_Brain]

I think this suggestion should be implemented.

In addition to this, there should be more incentive for rovers: Currently, rovers are the same as traditional landers, unless they are landed very close to a biome border or you are willing to drive for a very long time to get more science. A solution like in the RoverScience mod would be great, but with more experiments possible at each science spot. Instead of just a generic "Analyze science", it might be possible to take a core sample, take a picture with a camera and use an instrument like Curiosity's ChemCam, or some science spots may only work with a certain instrument (for example, at one spot the soil seems interesting, so you'd want to take a sample of that, but your ordinary camera wouldn't get you science there.

I support the idea that for example a thermometer only grants a small bit of science in a single reading, but many readings would get you a lot of additional science. After all, looking at a thermometer once doesn't tell you what a place's climate is like, you need to take many readings over time.

[Baron_von_Fluffy]

The "tech tree" is a means of getting new players to use base tier parts to understand basic orbital mechanics without overwhelming them with 100+ parts, so they can use more advanced designs effectively in future.

this allows the "casuals" (as someone called them previously) to get into the game and also learn.

also i disagree with "upgrading parts". all parts should have the same stats for everyone. this is why we all have the same in-game universe, so players can relate to each others games. there would be designs on the spaceport that require certain upgrades to get equal performance with the creator.

plus giving your rocket essentially a "commando pro" perk so it can burn longer on less fuel is silly.

I apologize, I assumed that the TUTORIALS were for TUTORIAL purposes, not the career mode... I suppose a tech tree is more of a tutorial tree. Then perhaps that is what my input is suggesting be revised. And it is not a commando perk. The point of science is to improve upon something, or perform the research so that we might do so and invent bigger and better rocketry that is more efficient, is it not? It's not a "commando pro" perk, whatever that is, it's "more efficient engines, fuels, and parts"...

edit: I was sort of unclear about what I meant by upgrading, I meant rather than flat out going from a tiny 215kN engine and jumping to a giant 2.5 meter 700kN+ engine, there could be some modular upgrading inbetween. Rather than swapping engines, why not swapping precooles, bells, fuel types, combustion chambers? It would require work, but it could be a potentially cool system for advanced players. It wouldnt have to be messed with by new players who don't understand it

Edited April 16, 2014 by Baron_von_Fluffy

[Creature]

I've never really understood the "tech tree is a tutorial so it's not the focus of gameplay" -thing. If you'd make career more about progressing along the tree, making it a tad harder (or at least longer) and more interesting, it wouldn't make it any less of a tutorial~ish for the new players. On the contrary, they'd spend more time refining their skills, getting some actual challenges and they'd have to think a bit about what they're gonna unlock next and therefore learning about what the parts really do.

I remember Harvester saying that tech tree is meant to be quickly completed for advanced players so they can get to the good parts as soon as possible. But isn't that what sandbox is for? If you don't want to play the unlocking game, then don't.

But science as it is is maybe a bit too simple system. I'd like a system where I have to actually plan my science mission in the VAB properly and make it a focus of the mission. Right now it's just slapping on all the experiments I have and making sure the vessel can get back. Only interesting part is designing (and flying) the craft to get the instruments to their destination.

[regex]

I've never really understood the "tech tree is a tutorial so it's not the focus of gameplay" -thing. If you'd make career more about progressing along the tree, making it a tad harder (or at least longer) and more interesting, it wouldn't make it any less of a tutorial~ish for the new players. On the contrary, they'd spend more time refining their skills, getting some actual challenges and they'd have to think a bit about what they're gonna unlock next and therefore learning about what the parts really do.

The reason career doesn't feel like progression right now is that it isn't finished. The tech tree is only one piece of the puzzle and, as it stands, is used to introduce newer players to the parts in a fairly controlled fashion. That could stand changing to some extent, but the currently released information we have is that science is only one of three "currencies" that will all be interchangeable. So once you unlock all of the tech tree there is still incentive to get more science because it will provide you with additional "income". Subject to change according to the dev's whims, or course, but that's what we know right now.

[Mystic]

I think the current system is more or less along the right lines. There are things that can only be done by a Kerbal, which is right. There are things that give full return by transmission, and other things that need returning for best results, which is also right. Now which things are which can be fairly quibbled - I agree that stuff like thermometry and gravimetry ought not to incur a transmit penalty - but the general idea is right.

The issue is with the biomes, which make the science grindy. We need a way to get a message popup about a biome crossing so we're not resorting to clickspam to pick them up.

I agree with science being mostly good the way that it is. But I can't get on board with a message indicating when you are crossing into a new biome, because I would find that to be gamey. The way I look at it is this: in real world science you don't know in advance whether or not you are going to get a big science return from going to any given spot on Mars, for example. You just take your best educated guess, based upon what you can see from a distance, and then cross your fingers that you find what you are looking for (or better yet, what you never would have thought to look for). I think the current system provides some emulation of that, in that it doesn't tell you whether going to a given spot for science is just going to give you the same results you got halfway around the same moon on the previous mission. We just say "that area looks like a plateau ... there might be a different biome there! Let's go find out!" ... and I like it like that, including the part where you get there and find out it was for naught.