On Implementing Functioning Life Support Systems for Kerbals

[FlowerChild]

I use lots of probes in early career because the mass penalty (and cost of kerbonaughts) is a real one.

Serious question: do you actually play career or does the above just represent how you think it should play out?

Because the above makes very little sense to me from an in-game perspective. A player is naturally trying to progress in the game, especially in the early stages. They then choose a command part that allows them to progress *less* per mission than the other options available to them, because they cost/weigh less? Of course they cost/weigh less...you can't do half the stuff with them you'd want to do.

I understand you'd *like* to use probes earlier on. Me too

Later in career, we lack an appropriate lever to discourage sending Jeb on a one way trip to eeloo, so it's all about balance.

Why would you send Jeb (or a probe) on a one way trip to Eeloo? First of all, your science returns would be terrible, meaning that you could be making much more effective use of your time flying a return trip. Second, there's enough science available in Kerbin's SOI to max out the tech tree, maybe many times over, so I think the odds the player will actually make it to Eeloo before the progression is over to be exceedingly slim.

As for a mechanic to discourage sending Jeb on that one way trip in the first place: Life support would do that. Of course, you'd also have to be willing to kill Kerbals with it and appropriately penalize the player for that death.

But balance does not necessarily mean equal, and if we land in a place where the probe cores have a secondary, specialized role (but still a role), I'm fine with that.

You keep using that word. I don't think it means what you think it means

Edited May 3, 2015 by FlowerChild

[RoverDude]

Yes, I do play career (just started another one with 1.0.2), but also use life support, Which is why I send out a lot of probes. Both due to the risk, and due to the mass penalty, especially early in the tech tree. I am also quite aware of what the word balance means.

[Wayfare]

Yes, I do play career (just started another one with 1.0.2), but also use life support, Which is why I send out a lot of probes.

Well that's cheating Seriously though, maybe it's a misunderstanding but I think this discussion should be about the stock experience and how to improve it.

Edit

That's not to say modded play is invalid. If anything, it shows that there are workable ways to imrove the stock experience!

Edited May 3, 2015 by Wayfare

[tater]

I've had Mun landers tip over on slopes that didn't look steep enough for that to happen. 5degrees or so. Big tall landers, because they were for contracts that wanted a cupola+science laboratory, which necessitates a tall, top-heavy lander.

Tip over, and all that happened was a solar panel broke. Otherwise, the craft and crew survived. But that craft isn't going to launch from the surface again.

Now, to return them to Kerbin, a rescue lander can be built. However, that rescue lander is N days away.

And that problem of the rescue mission being N days away only gets worse when you get further out. Due to both travel time, and launch windows. If a lander on Duna has a problem, then a rescue mission is many, many days away. For some planetary bodies, it would mean that unless a rescue mission had launched only a couple days after the main mission, then rescue is impossible.

Life support changes the nature of Kerbal Space Program more dramatically than anything else, I feel.

It turns the stories that attract people to the game, of "whoops, now, how to sort this", into "Oh. Damn."

Changes the whole principle where, as long as a Kerbal survives the landing, you can always get them back, to one where surviving the landing isn't enough.

That's a bigger change than anything else that's ever been proposed, I think. Well, other than the time the Mun first appeared.

This is why LS is so very awesome, IMO. It is the only thing that makes KSP get actually more difficult farther out.

- - - Updated - - -

And that, Louella, is why I disagree with lethal life support as the default option.

Because in the event of an 'oops' on Laythe/etc., you're pretty much done. And 'oops' that could also be as simple as time warping a little bit too fast. I'd rather have the chance at doing that rescue mission given the opportunity

This is exactly why I prefer lethal life support, actually. I virtually never lose kerbals. Even in RSS configs (kerbol system, scaled up between 2 and 6.4X, mostly, I try many of them out). LS is basically the only way I can manage to kill any (and that is rare).

My solution to "oops" is to plan missions with redundancy. I sent fuel/habs ahead, for example. Any one who loses kerbals some other way should be fine with lethal LS.

Any stock LS solution should drop warp in advance of LS running out, perhaps with a player controlled slider---maybe per craft? IN that last idea, you'd set a slider for a given craft for when the game will kill any warp in progress, and warn you. If you are on Duna, you might set it to a reasonable travel time from kerbin to duna, for example. For a station, it might be set to default (default might be 15 days or something).

Edited May 3, 2015 by tater

[RoverDude]

Well that's cheating Seriously though, maybe it's a misunderstanding but I think this discussion should be about the stock experience and how to improve it.

Fair enough

But does go to show that when there's a penalty (even a light one - I use my own life support which is non-lethal), it does drive folks towards weighing the choices differently. In any case, not much to add here other than my belief that any stock life support system should be non-lethal by default, and with a very limited resource count (i.e. what I did with the one I built myself).

I also expect there are some very interesting mechanics that can be done with probes to expand their versatility outside of piloting, but I'll need to noodle that one over, as it's not really a topic for this thread anyway.

- - - Updated - - -

@Tater - I hear ya. I rarely lost them in my own saves except by cases of 'oops' (usually warping too fast). My only issue with lethality is that if you're twenty minutes late to Jool with that resupply craft, you're pretty much hosed. I've also seen folks have whole bases wiped out due to a keyboard error, so it may just be matter of scale (i.e. I usually see 20-30 kerbals wiped out at once - once was because a drill took all of the power and asphyxiated them. I saw another where a converter took all of the water... and no chance of a rescue ship in time)

[tater]

@Tater - I hear ya. I rarely lost them in my own saves except by cases of 'oops' (usually warping too fast). My only issue with lethality is that if you're twenty minutes late to Jool with that resupply craft, you're pretty much hosed. I've also seen folks have whole bases wiped out due to a keyboard error, so it may just be matter of scale (i.e. I usually see 20-30 kerbals wiped out at once - once was because a drill took all of the power and asphyxiated them. I saw another where a converter took all of the water... and no chance of a rescue ship in time)

I actually really like your new LS as something for Stock, frankly (and you know that my first exchanges with you on LS were fairly opposite ends of the spectrum---you changed my mind. I just find it funny that people who will have no issues with "LOL I BLEW UP JEB" have such an issue with LS possibly killing someone. I suppose WRT Jool (I always send probes first, and that is very much "end game" for me the way I play), I'd not ever send a craft that could not get to Jool, and at least hang a round a year or so---and I'd send resupplies ahead, and likely behind as well. Course my last Jool mission was in 6.4X, so I needed rather a lot of LS.

I like the "rescue" as a sort of goal, as well, as I consider unplanned stuff like that to actually be the most fun (which is why I'd like some sort of rare failures, but the current mod make stuff break constantly).

[FlowerChild]

Yes, I do play career (just started another one with 1.0.2), but also use life support

Ok...so you're not actually talking about stock probe cores being useful in any way on their own, and thus have been making a straw man argument. Good to know.

Which is why I send out a lot of probes. Both due to the risk, and due to the mass penalty, especially early in the tech tree. I am also quite aware of what the word balance means.

You sure man? I keep seeing you apply it to things that don't really make much sense, not just in this thread, but in others as well.

For example, in this thread you have been using it to describe how a part, with no viable use case, is somehow "balanced" vs another that is entirely viable, because it has lower mass and cost.

In another thread (I think it was related to the resource system) you said something along the lines of "that feature is 'balanced' because of <insert realism factor>" when balance and realism are entirely separate concepts.

My overall impression is that you may think you do, but you really don't have a good understanding of what game balance actually entails.

EDIT: And because I hate misquoting or quoting people out of context, I looked up the balance vs realism thing I was thinking of:

RE the mass of the equipment - I actually spent a lot of time reading NASA papers on ISRUs, looking at prototypes, etc. and feel the current weight requirements are fairly balanced.

http://forum.kerbalspaceprogram.com/threads/117073-Ore-is-Overpowered?p=1880703&viewfull=1#post1880703

Edited May 3, 2015 by FlowerChild

[Louella]

This is why LS is so very awesome, IMO. It is the only thing that makes KSP get actually more difficult farther out.

Yeah, it's literally a "game-changer", so a stock implementation has to be well-thought out.

any stock life support system should be non-lethal by default, and with a very limited resource count

I had a look for your life support mod, and yes, a single resource to manage is a good idea. It's easy to make things complex, it's difficult to make them simple.

which is part of my thinking behind the thing I suggested earlier. Psych support, to keep the Kerbals sane when confronted with the Terrible Secret of Space, and Scientists can rescue catatonic Kerbals.

[rodion_herrera]

Great ideas all around. I'm glad I resurrected this long-time needed suggestion/recommendation. Like some of you, I think it's imperative that it be added into the stock game.

[tater]

Flowerchild, realistic is automatically balanced. If all stuff relevant to gameplay was "realistic," it would be internally balanced. Gameplay is different, I'd balance that by messing with how much kerbals use since that's obviously a fiction anyway.

it is odd to me that many people who claim they want LS also want it to become a 100% efficient closed cycle---which means LS doesn't matter.

[FlowerChild]

Flowerchild, realistic is automatically balanced. If all stuff relevant to gameplay was "realistic," it would be internally balanced. Gameplay is different, I'd balance that by messing with how much kerbals use since that's obviously a fiction anyway.

That would imply that reality itself is inherently balanced as a game system. I suppose that could lead into an interesting philosophical debate about intelligent design and the overall nature of reality, but that's far beyond the scope of these forums.

I've also experienced enough un-fun moments in life to take issue with the overall balance if that were the case, think that the pacing sucks, think that the progression is rather messed up in that the longer and harder you work the more quickly your capabilities decline, think that dogs should get better and more diverse dialog as the barking gets old quick, and would like to suggest that my nipples are currently an underutilized and therefore extraneous system that really didn't need to be included in the first place

Honestly, I suspect if it were a well balanced game, we wouldn't feel the need to play games in the first place. So no, I disagree with the underlying assertion that realism equates to game balance, and given RoverDude was talking about a game system in that post, I do not think it reasonable to assume he was talking about anything other than game balance.

Edited May 3, 2015 by FlowerChild

[tater]

I said balanced, not fair or fun

Really I mean relative to each other in terms of balance.

[tntristan12]

I would like to voice my support for a "snacks" mechanic, which has levels of difficulty based on the consequences of not providing enough. For example, easy has no consequences, medium involves kerbals losing the ability to do their jobs efficiently (they behave like they have no experience because they're so hungry), and hard involves death. Some variant on these could also work. Just an idea.

[FlowerChild]

I said balanced, not fair or fun

If you want to broaden the form of "balance" we're talking about beyond the implied "game balance", then yes, I will agree that if the ISRU parts are placed on opposite sides of a scale to objects of equal mass, they are entirely and well balanced

[Wayfare]

I'll need to noodle that one over, as it's not really a topic for this thread anyway.

You keep saying that, you tease. I want to see that thread

[tater]

LOL. Sadly, it will launch to orbit with a tiny tank, and a 909, and can reenter directly from Jool with no damage.

[FlowerChild]

LOL. Sadly, it will launch to orbit with a tiny tank, and a 909, and can reenter directly from Jool with no damage.

Well, now you're talking game balance, which is an entirely different matter

[RoverDude]

I would like to voice my support for a "snacks" mechanic, which has levels of difficulty based on the consequences of not providing enough. For example, easy has no consequences, medium involves kerbals losing the ability to do their jobs efficiently (they behave like they have no experience because they're so hungry), and hard involves death. Some variant on these could also work. Just an idea.

Well option 1 is basically disabled, option 2 is what I would see as a default (or somewhere around there, the mechanics are

You keep saying that, you tease. I want to see that thread

Sure, soon enough Need to sort out some other bits first. With CTT wrapped up I have some tech trees to update

[Eric S]

I would like to voice my support for a "snacks" mechanic, which has levels of difficulty based on the consequences of not providing enough. For example, easy has no consequences, medium involves kerbals losing the ability to do their jobs efficiently (they behave like they have no experience because they're so hungry), and hard involves death. Some variant on these could also work. Just an idea.

I'd be good with that, as long as turning it off actually gets it completely out of the way. I'm just thinking something like the aerodynamics, where there's a way for a mod to say "Yeah, I'll be handling this, don't waste any computation on it". Really, anyone that feels the need for something more realistic in this is going to have different opinions about what matters, so they might wind up modding it regardless of what type of life support you put in. Just look at the number of different life support mod options that already exist.

[rodion_herrera]

I love the "snacks-mechanics" idea. I think it fits wonderfully into the Kerbal canon.

[rcp27]

In the end, life support is simply a matter of "add x additional parts to your ship to give you y time to complete your mission". At the moment, the game offers absolutely zero method for predicting the value of y. How long does it take to reach Duna from Kerbin if you leave at a pretty good transfer window? Once you get there, when will the next transfer window home come up? If, when I'm there, something goes wrong, when will the next transfer window for a rescue mission come round? How long will it take for that rescue mission to reach Duna? Without some kind of method for the player to get a handle on the answers to these questions, the only way of figuring out the value of y and therefore x is pure guesswork and dumb luck. That places life support in the same category as "random part failures" as a mechanic that is 100% guaranteed to enrage players and destroy their fun.

[allmhuran]

Whenever this topic come up I make the same two suggestions for any implementation of life support in KSP:

- Base life support on snack supply.

- Kerbals never die, but they get grumpy and stop working if they run out of snacks. In career mode this might mean they refuse to do experiments. Not sure what it might mean for sandbox.

[Tw1]

In the end, life support is simply a matter of "add x additional parts to your ship to give you y time to complete your mission". At the moment, the game offers absolutely zero method for predicting the value of y. How long does it take to reach Duna from Kerbin if you leave at a pretty good transfer window? Once you get there, when will the next transfer window home come up? If, when I'm there, something goes wrong, when will the next transfer window for a rescue mission come round?

There is a relatively simple solution to this- give the player the planning tools.

A launch window schedule, or maybe even a porkchop plot generator, would be a great addition to the Mission control building.

A simplified version of something like this would work.

http://alexmoon.github.io/ksp/

Re life support, I think it should be kept simple.

IMHO, it should

*Only have one, or at most, two, dedicated resources. Air and water can be recycled. Something harder to perfectly recycle is food, so it could be based of that, if anything.

*Be light, and last for a long time, so bringing it with you only starts being a design issue when you need many months' worth.

But not so big, that anything apart from a short stay on another planet is infeasible.

*Have some recycling mechanism, but not one that enables a complete closed cycle.

It should also be big and heavy enough that it's only worth bringing when you need very large amounts.

Eg, if it provides a 50% reduction in life support use, it should be at least as heavy as two large lifesupport containers.

*Have some sort of ISRU system to enable you to produce more life support. Also pretty heavy.

This wouldn't work everywhere. It would work best on Kerbin and Laythe, IDK about the others. An argument could be made you'll find necessary components to grow more food in many places, but if you can use it anywhere, even at a lesser rate, there's less incentive to be strategic about which planets you set up what bases on.

*Let kerbals last a considerably long time on EVA.

So rescue, and EVA orbital shenanigans are still possible.

At least one kerbin day, or even a week.

Edited May 6, 2015 by Tw1

[Scoundrel]

Aaaand the cycle continues! I think I've managed to reply to every life support thread since 2012... do I get a space cookie?

There only needs to be 1 resource (it was originally Snacks, but the Snacks guy took that so I'm changing my vision of Life Support to be called Munchies) that is consumed once per 4 hours (based on the mission clock), after which a kerbal gains hunger (meaning, Munchies!) over a period of 6 kerbal days. At the end of the 6 days the kerbal becomes uncontrollable and if the vessel has no kerbals that can be controlled, it is like a probe that has run out of energy. Transferring snacks via docking - or having a scientist give an unresponsive kerbal a snack - lowers the Munchies to the point where a kerbal is controllable, and a full 6 Snacks completely removes the Munchies. Kerbals cannot starve to death - they hibernate or something. Thus ships and missions and kerbals trapped on EVA can be rescuable, which fits with the game design principles of KSP.

Also it shouldn't require a "flush waste" or "change filter" button that needs to be pushed every 15 minutes (or whatever) or your kerbals die, which is nothing more than a means of cognitively rationalizing OCD being implemented directly into the game.

It should only require two parts:

Radial Snack Bin: For extra snacks; it could be put in the service bay part for better packaging, and should have a maximum capacity suitable for most long range missions for a kerbal.

LS ISRU: it needn't be heavy or large, but it would generate a fair amount of heat, and require large amounts of electricity. I imagine it would have a logo on it or something.

By keeping it simple it should mean that long duration missions don't create laggy ships of computer death because we need another hundred parts just to get them somewhere, and should have margins large enough that new players don't waste hours of their time on a mission only to lose control of it because they miscalculated.

[CodedThreat]

I am pretty sure the reason KSP is so great is because of the mods. Every time someone brings up something they wish was in the game, the general response to it is "There is a mod for that!" I also believe the reason Squad left the game so open to modding is cause they knew they would not be able to fill every niche and every want of every player.

Some people like to use RSS and Remote tech and the USI packs and EVE and on and on until they are basically making their own experience and challenges, cause for them, if it was just stock with no mods, they would have lost interest after a few weeks of playing. On the other hand, you have people who really like the challenge of the stock game and play with no mods and try to build amazing things under the tight restrictions that the stock parts offer and make the challenge in the game play for themselves that way.

What I am trying to say is that, every mod that has been made, was produced to fill a need a person felt should be in the game and they are all correct. We should have this that and the other thing, but not everyone wants it. There are new players coming to the game that have a hard enough time getting to orbit or building a proper rocket. So, on top of needing to learn the already difficult base game, you want to throw another thing into the mix and yet another thing to keep track of, then you would not be helping build the community but would actually be pushing people away from it.

Also, a lot of kids play this game, I play this game with my 10 year old niece when she comes to visit, she loves the little Kerbals. So if Squad introduce life support, and she starts playing her career and suddenly her Kerbals die on Mun or Minmus cause they ran out of life support, I am pretty sure she would be heartbroken (she already does everything she can to make sure they always get back home). And it might take away from the fun of the game and cause her to not want to touch it again, it was hard enough when she was 8 teaching her how to build a rocket and fly it to orbit, now I would have to teach her how to plan her missions so she doesn't run out of snacks?!?

I love TAC LS and play with it all the time, cause I can do the planning and I like the challenge it gives me. And I have tried other life support mods that are just as good and offer the same challenge, but it does not belong in the base game. It is something for people to discover that they can add to their game when they do reach a point where the challenge of just launching another rocket to Mun is just a mundane event!

See what I did there, Mundane, Mun.... oh, I crack me up.

TL:DR It has it's place as a mod, not in the base game!