Realism in KSP

[NathanKell]

Also, you should read that post of HarvesteR's that Rowsdower linked. His original proposal was a 2D game without real orbital mechanics, because real orbital mechanics are too tough for people.

And yet here we are.

[r_rolo1]

Yeah, that post could be also used to say that the game has been steadily progressing in direction of more realism

P.S On a tangent , what has gone to the game developers these days for them to so frequently assume that their players are dumb as a rock or to atleast heavily underestimate their average inteligence ? In other example, there is a quite seasoned game developer of a certain famous 4x strategy game that said some years ago that the fifth instalement of his game has removed any randomness of their combat mechanics because he believed that the players ( note : remember, strategy games, that kind of games where people play to exercise their mind ) didn't had the mental width to deal with probabilities. He continued saying that people would end saying stuff like " why did I lose this 3:1 battle? Three is such a big number and one is so little ... "

Yeah :/

Edited August 29, 2014 by r_rolo1

[Xyrus]

There is a balance between realism and fun. Too little and it becomes basically a kid's toy. Too much and you end up with a very limited player base. No one want's a "press a button and your on the moon!" game, but at the same time no one wants to play a game that requires a Ph.D in physics and engineering just to be able to get a rocket off the launchpad or plane off a runway without it falling apart or exploding.

Squad has been trying to find that balance that appeals to the largest audience with the stock game, something not easily done with a game like this. In my opinion they're making decent progress towards that goal. But there's no pleasing everyone. Some are going to want more realism others will want less. Some will want more realism in only certain aspects of the game, and less in others. For example, some may want more realistic aerodynamics for planes, but don't necessarily want to deal with life support issues. In those cases, if people want more or less realism, then they can create and/or use mods to do so.

It's Squad's game, and I'm pretty sure they aren't ignoring the user community. If they think a particular modification will enhance it's appeal to a broader player base (i.e., SP+) then they'll add it. If they don't, then they won't.

[Jouni]

I think there are several different classes of realism features:

1. Improved physics simulation usually makes sandbox games more fun, if the overall complexity of the game doesn't increase too much. Better aerodynamics and reentry damage would be good additions, while n-body physics would change the game completely.
   
2. Features involving abstract game mechanics are much harder to get right. Having life support in the game could be fun, but it could also end up being just another layer of tedious micromanagement.
   
3. Some features increase complexity without making that much difference in gameplay. We've used many different rocket fuels in the real world, because we've had many different concerns to consider. How many of those fuel types would be relevant in a game that models only a small fraction of those concerns?
   
4. Some features affect game balance. Bigger planets would make multi-stage rockets more relevant, but rocket launches and other routine tasks would also take more time. Reducing engine efficiency would make staged rockets relevant without increasing launch times, but it would also make everything else harder.
   

[regex]

Some points:

Features involving abstract game mechanics are much harder to get right. Having life support in the game could be fun, but it could also end up being just another layer of tedious micromanagement.

I don't think I've ever "micro-managed" life support under any mod in KSP, so I 'm not sure what you're talking about. Is that something to do with the base-building mentality where someone is worried about having to ferry life support resources around?

Reducing engine efficiency would make staged rockets relevant without increasing launch times, but it would also make everything else harder.

Most engines in KSP are ridiculously underpowered or inefficient compared to real world examples, I'm not sure why they'd need to be nerfed further...

[Vanamonde]

If a few numbers were changed to make something more realistic, would you even notice? Would anyone unaware of the change? Except for having played many, many hours, in particular, I'd have to say the answer would be "no."

Maybe I'm misunderstanding, but this makes it sound as if you consider the changes you yourself want to be arbitrary and unrelated to gameplay. In which case, why bother with them at all? It seems to me that many of the arguments in favor of rescaling the game are simply aesthetic objections. Some people want the game's numbers to be closer to the real solar system's numbers, just for the sake of the idea of "realism," proposing parallel changes to keep the experience of playing the game much the same. In which case, again I must ask, isn't that an awful lot of work to ask of the devs for an abstract final result that is merely more aesthetically pleasing to your own personal tastes?

[juanml82]

That doesn't address my post at all. Your claim was that a 500+t to orbit rocket is unrealistic because our largest IRL rockets lift far less. I made no mention of jet engines or the type of transfer an interplanetary mission would use, just that using lack of historical example as evidence of lack of realism is not accurate.

Well, you did mention SSTOs, which in KSP often use jets, and interplanetary missions, which in KSP are often done via slow Hoffman transfers.

There is no known working engineering solution in the real world capable of lifting 500T. It's not known if it is even possible. Sure, it seems fine on paper, but there is a big difference between concepts and reality. So I'll call unresolved engineering challenges unrealistic. That doesn't mean they shouldn't be possible in the game.

Microgravity is a tough one, because most solutions to it I've seen involve centrifuges of one sort or another. In KSP as it is now, that would preclude the use of on-rails timewarp and make IP transfers incredibly long. Maybe an animated centrifuge part could be used, or maybe the microgravity issue could be ignored completely as real life astronauts have survived over a year of being in orbit without significant long term effects.

They do face significant long term effects years later. Microgravity is not ignored in real life for the planning on any manned interplanetary mission.

Basically, sending a bunch of Kerbals to Jool, returning them and putting them back into another ship like nothing happened isn't realistic.

To borrow from the anti-realism crowd, do kerbals get cancer? Their world is tiny, they might have evolved for higher radiation levels. Honestly, anything related to Kerbal physiology is far easier to handwave away, unless you have some working model of Kerbal medical science (must remember to capitalize or spell checker changes it). It's enough for the stock game for them to say that kernels need 1 unit of life support per day (food/water/air/scrubbing-all inclusive), and that long durations trips might need a special "long duration module" (retexture the hitchhiker or something). The LDM would include whatever it is that kernels need of longer trips that is in addition to consumables (exercise equipment, recreation, a "storm shelter" whatever you want to imagine).

Realistic mods can mess with those to make them specific to nominal human issues if they wish. Radiation is a tangent, frankly, and I don;t care one way or another about it. Having LS alone is great, and the plus of "radiation" is that it adds another novel set of design limitations.

Radiation shielding is part of life support. It can be discarded by the game designers and the in-universe explanation might very well be "Kerbals are immune to radiation"

[tater]

Maybe I'm misunderstanding, but this makes it sound as if you consider the changes you yourself want to be arbitrary and unrelated to gameplay. In which case, why bother with them at all? It seems to me that many of the arguments in favor of rescaling the game are simply aesthetic objections. Some people want the game's numbers to be closer to the real solar system's numbers, just for the sake of the idea of "realism," proposing parallel changes to keep the experience of playing the game much the same. In which case, again I must ask, isn't that an awful lot of work to ask of the devs for an abstract final result that is merely more aesthetically pleasing to your own personal tastes?

No, they'd not be arbitrary at all. Arbitrary would be making them intentionally unrealistic for no reason. Realism means that new things added later make sense. There is no need to balance "realistic" physics, they are what they are. When you make stuff up as a dev/modder, you have to keep making everything up arbitrarily "for balance."

It's not an awful lot of work to reset a scaling factor. People do it in mods, people might do it this evening over a beer. Much has already been done from whole cloth (FAR, etc) already.

I have proposed nothing of the sort you suggest, BTW. Do you not think Isp should work as the math has it actually work?

Seriously, the anti-realism folks seem wedded to what, KSP as it is this instant? 0.24.2? Are any changes good?

My comments on the user not knowing the difference are not that the ships would behave identically, it's that the USER EXPERIENCE would be identical. You'd still get to the Mun, but the specifics would be very slightly different. The user experience, exactly the same. You build a rocket, you fly it the same (different values to numbers of trajectories here and there, but the act of flying is the same). Shape orbits the same way, your maneuver node will just have slightly different burn lengths, etc.

In other words gameplay not hugely changed from the perspective of actually playing, though what craft are idea for a certain mission design likely will change.

Edited August 29, 2014 by tater

[tater]

There is a balance between realism and fun. Too little and it becomes basically a kid's toy. Too much and you end up with a very limited player base. No one want's a "press a button and your on the moon!" game, but at the same time no one wants to play a game that requires a Ph.D in physics and engineering just to be able to get a rocket off the launchpad or plane off a runway without it falling apart or exploding.

Squad has been trying to find that balance that appeals to the largest audience with the stock game, something not easily done with a game like this. In my opinion they're making decent progress towards that goal. But there's no pleasing everyone. Some are going to want more realism others will want less. Some will want more realism in only certain aspects of the game, and less in others. For example, some may want more realistic aerodynamics for planes, but don't necessarily want to deal with life support issues. In those cases, if people want more or less realism, then they can create and/or use mods to do so.

It's Squad's game, and I'm pretty sure they aren't ignoring the user community. If they think a particular modification will enhance it's appeal to a broader player base (i.e., SP+) then they'll add it. If they don't, then they won't.

I disagree with the premise entirely, that there is a balance between "realism and fun" in a game. Take modeling going to the bathroom, the guy mowing the grass at the Space Center, etc off the table. We're talking about the mechanics of movement, and a few relevant technologies within space flight/rocketry/orbital mechanics. What determines fun has exactly nothing to do with the physics, arbitrary and wrong physics is no more inherently fun than non-arbitrary real physics.

No one is suggesting that you be a physicist to play. As it is, the UI is good enough that the physics is unseen by the player anyway. People are arguing that the unseen black box be arbitrarily lousy at math because if the players knew the math was right, they'd apparently decide that it was less fun. The user experience would not change. Would the trajectories change? Sure, but not the user experience.

[tater]

Radiation shielding is part of life support. It can be discarded by the game designers and the in-universe explanation might very well be "Kerbals are immune to radiation"

Fine, as I said, who cares about the specifics when it can be abstracted a little? The only reason I thought radiation was cool is that it ADDS TO GAMEPLAY. Anythign that adds a new problem to be solved has that potential. That's worth looking at from a balance standpoint. "Do we want hostile environments in space or on worlds that will require specialized missions or parts?" If that looks to not be fun, then "no." No biggie.

I'm not a fundamentalist, I do not think that the Book of Jebadiah, verse 0.24.2 is the holy word, inerrant and immutable. The career is pretty "meh" right now. The same craft, or slight variants will get you near anywhere. I saw the radiation suggestion and thought, "there's a cool way to make missions to certain worlds more difficult in a novel way." If they are immune to radiation, so be it. They presumably need to eat. Of course as of 0.24.2, they are effectively immortal anyway, they can only be killed by time compression in my experience.

[smart013]

You forgot one:

5. Lobby the devs to hold KSP to a higher standard, to make KSP better.

Since it seems that the orbiter devs did most of the things you hold dear right, why not lobby them to do some (small) changes and become the "better ksp"?

[Jouni]

I don't think I've ever "micro-managed" life support under any mod in KSP, so I 'm not sure what you're talking about. Is that something to do with the base-building mentality where someone is worried about having to ferry life support resources around?

An example of micromanagement would be flying routine resupply missions to a space station orbiting Kerbin. Building a resupply ship and demonstrating that it can get its job done is fun. Having to do it a few dozen times while trying to do interplanetary missions gets boring.

Most engines in KSP are ridiculously underpowered or inefficient compared to real world examples, I'm not sure why they'd need to be nerfed further...

It's part of the game balance issues with planet sizes. Reasonable aerodynamics make reaching orbit from a stock-sized Kerbin too easy. There are currently two solutions to the problem. You can fix that by increasing the size of Kerbin, but then the player has to spend more time on routine stuff. The other solution is to reduce engine efficiency further, which has the side effect of making everything else harder.

[tater]

An example of micromanagement would be flying routine resupply missions to a space station orbiting Kerbin. Building a resupply ship and demonstrating that it can get its job done is fun. Having to do it a few dozen times while trying to do interplanetary missions gets boring.

This is an absolutely valid argument, unlike many others. I agree completely. Part of managing a "space program" might be such boring, resupply runs. How about SOME automation as a better solution? You have these excess astronauts, after all, sitting around while Jeb does the heavy lifting. Want if the management would task a few of these for "resupply." In addition to the kerbs, you'd have to have a saved rocket type to assign them, then a frequency of delivery (once every other month or whatever). The resupply would appear docked at some point, then later it leaves and the guy is back in the pool, waiting for his next run. Such an abstraction could provide the source of stranded kerbals, etc. (yeah, adds random stuff, but it's NOT things the player is flying… there might be a small % chance (based on "stupidity") of a mission failure, which would then generate a "contract" for the player to rescue the ship/kerb. Win-win, gameplay wise.

[regex]

Since it seems that the orbiter devs did most of the things you hold dear right, why not lobby them to do some (small) changes and become the "better ksp"?

Because they don't do the other things I hold dear. As I noted, they don't have a native Linux client, they don't have a VAB/SPH and Lego-style craft building, they don't have a career mode which I actually find kind of fun, and they don't have Kerbals. KSP is much closer to the game I'd love to play. It needs some work, but I still love it and want to see it become better. Besides, I freely admit that, while I want to see a real scale solar system in stock because I think it's better and more fun gameplay overall, it doesn't really belong in stock. Many other changes do, however, such as improved aerodynamics, thrust correction, and reentry that matters.

An example of micromanagement would be flying routine resupply missions to a space station orbiting Kerbin. Building a resupply ship and demonstrating that it can get its job done is fun. Having to do it a few dozen times while trying to do interplanetary missions gets boring.

Sure. How do most mods handle that? How do you think SQUAD would handle that, given that they don't like things that are tedius or not fun? Life support can be a wonderful early game limiter that can become irrelevant through player advancement, and I think that is how it should be in stock.

It's part of the game balance issues with planet sizes. Reasonable aerodynamics make reaching orbit from a stock-sized Kerbin too easy. There are currently two solutions to the problem. You can fix that by increasing the size of Kerbin, but then the player has to spend more time on routine stuff. The other solution is to reduce engine efficiency further, which has the side effect of making everything else harder.

It stands to reason that, if the delta-V requirements to orbit are the same, the time to orbit should be pretty similar as well. I don't think this is an issue to worry about.

[smart013]

Would it be possible to continue this otherwise interesting thread without tagging posters as "realism mongers" or "the anti realism crowd"? Maybe the "aggressive tone" the mods were warning about would soften a little.

On topic: i myself would draw the line for a "kerbalised reality" somewhere around FAR and deadly reentry. RSS elements would in my opinion sacrifice too much of the playfulness of ksp with his learning curve from "explosions are fun" to "a really efficient encounter with moho is fun" for a lot of unhappy people on all sides. Something like the "snacks" mod would be fitting for ksp, something like tac ls would not.

Of course *my* perfect ksp would have a lot of things included, some from mods, some from my preferred playstyle, some from my shortcomings in math But in contrast to a lot of posters in these threads i am not believing that *my* way of things is the one everyone else should adapt. And with this i feel quite at home at ksp, whereas orbiter and his community ... not so much.

Just my 2c.

Edited August 29, 2014 by smart013

[juanml82]

Would it be possible to continue this otherwise interesting thread without tagging posters as "realism mongers" or "the anti realism crowd"? Maybe the "aggressive tone" the mods were warning about would soften a little.

On topic: i myself would draw the line for a "kerbalised reality" somewhere around FAR and deadly reentry. RSS elements would in my opinion sacrifice too much of the playfulness of ksp with his learning curve from "explosions are fun" to "a really efficient encounter with moho is fun" for a lot of unhappy people on all sides. Something like the "snacks" mod would be fitting for ksp, something like tac ls would not. Just my 2c.

The catch with life support supplies is that you can have, early in the tech tree, a ultra-efficient waste and air recycler, meaning only food (ie, snacks) run off

[tater]

Given that things like "deadly reentry" are perfect fodder for the new difficulty settings we have been told about, there is no reason not to have such stuff in stock as a setting.

From a modding standpoint it's then something to tweak, instead of something to invent from whole cloth. I'm not up on modding this game (yet ), but in other games I've modded this can make it easier to "play nice" with other mods. I might be wrong in the case of Unity, I'm sure a modder here can say if this has any truth to it.

[JebNeedsHelp]

As someone who has only been playing stock KSP .23.5 for a few months, has yet to try version .24.2, and only has a small gray bar underneath his name, here's my take:

I would be fine with increased physics realism, as long as my run-of-the-mill computer could handle it. Right now my computer already counts MET clock seconds as slowly as two or two and a half seconds long during liftoff of some of my bigger ships, and I can't even land a Kerbal on Tylo yet (my last frontier, if you will), nor have I returned any Kerbals from anyplace except the Mun and Minmus. If I had the .craft files of some of the monstrosities I've seen built by the really advanced players here, I'm not too sure my computer wouldn't burst into flames. The people I know personally who have heard of KSP and play it from time to time are not die-hard players with top-of-the-line PCs, and don't even bother with the forums. I'm guessing those casual players wouldn't even notice (much) if the physics was tweaked, unless their games started really slowing down.

As for life support realism, though, I'm in the "less realism is better," camp. I enjoy doing things we can't do in real life, like starting colonies all around the system. It's already easy enough to kill a Kerbal without asphyxiating him or giving him radiation poisoning. Make life support too realistic, and long-duration colonies become death traps. Eventually I might try the Snacks mod, when I'm good enough, but with TAC LS every Kerbal I've ever sent somewhere beyond Kerbin's SOI would be dead, and where's the fun in that? (As an aside, it would be cool if your off-planet Kerbals had a way to break down the local rocks into the air they need to breathe and whatever it is Kerbals consume to survive. I'd be totally on-board with that. Right now I just role-play that they've got a machine for that tucked away on every ship for now, anyway. And maybe absorbing radiation is like giving haute cuisine to a Kerbal. No wonder they all want to become Kerbonauts! )

So I'd crank a physics slider as far as I could, while I'd leave a life-support slider where I wasn't killing off all my colonists.

My two centiKredits, or whatever it is they use for money.

JNH

[O-Doc]

I don't get it. There's already realism mods in the game covering everything which has been described plus more. Squad has already indicated that the admin building will have settings and there's still big features yet to roll out. This game is still alpha. Squad has a project plan and schedule in place. Why not be little patient and wait for the project to be more finished before having a bash on what this game should or shouldn't be? In the meantime: There. Are. Mods. That. Supply. The. Realism. You. Are. Discussing.

OK, I get it, the community feels a sense of ownership of the game and wants it to be seen well to strangers and newcomers alike. The community takes pride in the game and their achievements in it and doesn't want the game to potentially disappoint their peers. However, I trust Harvester's vision because:

1. I've had a ball playing this game and have gotten my monies worth many times over
   
2. I classify this as an old school game. Which is, something unique that no-one else has done.
   

That takes some serious talent to achieve. I don't want a game that everyone wants to play and feel special about. That's not achievable. I want the game Harvester wants make.

[Vanamonde]

It's not an awful lot of work to reset a scaling factor.

It has been suggested here that increased warp factors should be introduced so that the longer distances are covered in the same real-time interval. But higher warps introduce inaccuracies in the simulation of trajectories affected by gravity, as the ships move through the gravity well in fewer ricks. That certainly would not be realistic, and would require the developers to devote time to ironing out this problem, which need not be introduced into the game in the first place. I am dubious that the art of planet surfaces could be arbitrarily blown up by a simple numerical factor without ruining their appearances, either. It's not as simple as you make it out to be.

... it's that the USER EXPERIENCE would be identical.. The user experience, exactly the same... In other words gameplay not hugely changed from the perspective of actually playing...

You keep insisting this is a necessary change while also arguing that it would make no substantive difference even if it was implemented. If it doesn't change the gameplay experience, then it is arbitrary, and a waste of effort to go full circle and end up where the game started off.

Fine, as I said, who cares about the specifics when it can be abstracted a little?

I agree. Let's abstract the planet sizes and distances between them so as to keep the scale of the game convenient and fun to play. Which is the situation we have now.

[regex]

Why not be little patient and wait for the project to be more finished before having a bash on what this game should or shouldn't be?

Discussing the shortcomings of the game in no way signifies impatience and, since we have no idea when or if SQUAD will ever address any of these shortcomings, now is as good a time as any to talk them over. If there were something to indicate what SQUAD had in mind for the future, like a roadmap of planned features for instance, we probably wouldn't be talking about this right now.

I want the game Harvester wants make.

The game Harvester wants to make (or rather, wanted make in the first place) is covered in a link to the Orbiter forums that Rowsdower posted earlier. Quite frankly I'm glad he got talked out of it because it wouldn't have been anywhere near as successful and we certainly wouldn't be having this conversation right now. KSP is pretty good as-is, and does a fair job introducing orbital mechanics in an intuitive manner, but it needs some work on key parts of the simulation. Whether those fixes come now or later isn't really of importance; whether they come at all is.

[Rainbowtrout]

No matter how you slice it, if NASA comes to you and says they'd like to work with you, you take it. It's an extremely rare opportunity for a game like ours to work with a big time governmental space agency *and* be able to create relevant new content for the game at the same time.

I detailed some of that in a blog post a while back, and I thoroughly agree. It's the NASA. Even if a cleaner asks you to show them KSP, you do it. Not now because it's a nice thing to do, but also because it's NASA. We haven't gone back to the Moon in about 40-something years, without them it could be a question of we haven't been to the moon. If you also get to develop the game that you love, it's a win-win. NASA gets some of the dirty work of raising awareness done for them, you push out a new update with ASTEROIDS.

#0.02$

[O-Doc]

Discussing the shortcomings of the game in no way signifies impatience and, since we have no idea when or if SQUAD will ever address any of these shortcomings, now is as good a time as any to talk them over. If there were something to indicate what SQUAD had in mind for the future, like a roadmap of planned features for instance, we probably wouldn't be talking about this right now.

I was mostly referring to the aggro which has sprung up around this topic. I'm pretty comfortable with whatever road map Squad has for the game. I haven't invested much time into modding because we don't really know what the game is going to be in its release state. I'm not really a big fan of realism. If I want realism I'd have studied astrophysics or aerospace engineering. What I do want is immersion so that I feel like I'm running space agency. That means all the design, logistics and navigation that comes with it. I suspect this is more about what this thread is over actual realism.

Having said that, one of the fundamental and recognised appeals of this game is big firey explosions when your absurdly built rocket tears itself apart.

[Linear]

I was mostly referring to the aggro which has sprung up around this topic. I'm pretty comfortable with whatever road map Squad has for the game. I haven't invested much time into modding because we don't really know what the game is going to be in its release state. I'm not really a big fan of realism. If I want realism I'd have studied astrophysics or aerospace engineering. What I do want is immersion so that I feel like I'm running space agency. That means all the design, logistics and navigation that comes with it. I suspect this is more about what this thread is over actual realism.

Having said that, one of the fundamental and recognised appeals of this game is big firey explosions when your absurdly built rocket tears itself apart.

That's your opinion. Others have their own opinion(s). All have valid arguments.

The only aggro that's been caused has been from people not respecting and just plain arguing about these opinions.

[cantab]

On the station supply problem with life support: Add a "commercial supply port" part. Put it on any ship and it converts money into life support provided said ship is in a closed stable orbit or landed, cost depending on situation. (LKO pretty cheap, surface of Tylo eye-wateringly expensive.)

That's just one idea. The wider point is that when a change to the game, eg life support, might have negative effects, eg needing to fly repetitive supply missions, there's always going to be plenty of ways to fix the negatives without losing the positives. It just takes a little imagination to come up with them.