Life Support system?

[tseitsei89]

1 minute ago, XLjedi said:

@Brikoleur  What are your thoughts on a relatively simplistic "supply line" type radius implementation?

I would like to see something where a base-to-base supply line network might need to be established and yet not have to deal with repetitive resupply missions.  I'm guessing the coding would be very similar in many respects to how the current comnet works?  Maybe a refinery on one planet or moon can simply be connected to another planet or moon base with drone resupply ship connector.  When the orbits are within acceptable windows, supply levels are tested and resupply drones are launched if needed.  I don't think we really need to literally model the drone launching in much detail.  Maybe only an active supply line on the big map, and that's it?

The modpack I just talked about is a crude version of exactly that

[Guest]

27 minutes ago, tseitsei89 said:

Sorry my bad. You just said that all existing LS implementation are tedious or trivial. I showed you to a mod pack that has LS that is neither. And you just chose to ignore that

I shouldn't have made that comment on mods. I withdraw that remark. I don't have much experience with LS mods (as I said, I've tried a few but did not like them), so it was ill-advised.

However, I'm attempting to ignore your tangent on mods because it is completely irrelevant to my objection: that implementing LS well enough that it's interesting but not tedious or trivial is likely too much work for a small team on a limited time and budget, who already have ambitious new goals for their game. There are all kinds of other things that will come up.

For example, consider the UI. Suppose, say, EVA suits have 6 hours of LS. That's more than enough for most EVAs, but you really don't want to forget a kerbal outdoors. So you'll want an alarm clock that warns you when LS is about to run out. How does this work in the UI? What if you have very large numbers of kerbals and craft out there, how will the UI scale to them? What if an automated route you've set up runs out of resources and risks stopping? You really don't want this to happen without you noticing, that needs a different type of alarm. Should some resources (e.g. funds) be allowed to go "into the red?" Which ones? What are the thresholds for these alarms? Should they be hard-coded or configurable? 

Then the UI for the automation. How exactly do you set up the routes? Once you've got a system set up for them, you'll need a UI that gives you an overview of them with the status of each route. Where is that UI? What does it look like? What elements does it have? How do you display the status?

And this is just one subsystem in the set of systems that would be needed to make LS meaningful but not a chore. It would certainly be doable, but it is a big set of systems. Entire games have been built only around these kinds of systems, and it would be just one small aspect of KSP2.

All this is a daunting prospect -- and yes I still feel that you lot fail to acknowledge just how daunting. It seems to me that you're hand-waving away the problems, or like you're doing, dismissing them with "oh, easy peasy, even modders have managed this."

11 minutes ago, XLjedi said:

@Brikoleur  What are your thoughts on a relatively simplistic "supply line" type radius implementation?

Depends on how simplistic. If it's just a supply line radius -- i.e., a surface base automatically supplies any installations within its SoI with the resources it's capable of producing -- then it's IMO too trivial to be worth doing at all. You could certainly flesh it out with more meaningful gameplay, but then you start very quickly running into the kinds of complexities I listed above.

Basically: to be meaningful, there has to be a possibility of failure. If there is a possibility of failure, you need the tools and systems to avoid failure, and these tools and systems are a significant effort to implement.

Edited September 12, 2019 by Guest

[Guest]

1 minute ago, Brikoleur said:

I have said several times ITT that this is not what I believe. 

If you can't be bothered to even read what I'm saying, then there is absolutely no point in my talking with you at all.

If you think that resupply are tedious, designing challenges trivial and logistic systems not viable you will never like any LS, Mining, Colonies or Shipyard gameplay at all, but this is fine, anybody has his way of playing the game.

 

There can be 3 kind of solution to any consumable problem:

1. You can abstract it away by not having the system at all, no LS, colonies are just fancy Kerbal containers and Shipyards just work with money like the VAB does.
2. You could require setup missions to make the thing work and so making it a design challenge, in this category you have all the options with automation or abstracting that require a setup first (like various logistic systems in colony mods)
3. You could just make the player drive every truck, dig every mine and fly every resupply making it a micromanagement hell.

 

They confirmed colonies needing resources and advanced fuels needing to be manifactured so I would exclude the first option, they also said that they don't want to turn this game into "space simcity" so i would remove option 3. This leaves us with the second option, a system that needs some setup/design but then works on its own. Any system of that kind could solve every LS problem.

[XLjedi]

5 minutes ago, Brikoleur said:

Depends on how simplistic. If it's just a supply line radius -- i.e., a surface base automatically supplies any installations within its SoI with the resources it's capable of producing -- then it's IMO too trivial to be worth doing at all. You could certainly flesh it out with more meaningful gameplay, but then you start very quickly running into the kinds of complexities I listed in the post just above.

Based on that logic, one might suggest they shouldn't bother to reimplement the comnet system then?

[Guest]

2 minutes ago, XLjedi said:

Based on that logic, one might suggest they shouldn't bother to reimplement the comnet system then?

No, CommNet is actually interesting because it's not simply "a relay of strength X will always connect to KSP with station level Y." You have to deal with occlusion, signal strength, distance to Kerbin at different orbital phases, and deal with temporary loss of signal -- and the worst that will happen in a failure is usually just losing a probe. That's a good balance of challenge and complexity.

Edited September 12, 2019 by Guest

[XLjedi]

2 minutes ago, Brikoleur said:

No, CommNet is actually interesting because it's not simply "a relay of strength X will always connect to KSP with station level Y." You have to deal with occlusion, signal strength, distance to Kerbin at different orbital phases, and deal with temporary loss of signal -- and the worst that will happen in a failure is usually just losing a probe. That's a good balance of challenge and complexity.

M'kay... same level of challenge/complexity could be implemented to work with connected supply lines.  Worse that could happen is kerbals get put in hibernation pods until supply lines restored.

Edited September 12, 2019 by XLjedi

[Guest]

Just now, XLjedi said:

M'kay... same level of challenge/complexity could be implemented to work with connected supply lines.

Yes it certainly could, but it would require designing and implementing systems with that complexity, which, as I seem to have to keep repeating, would likely be too ambitious given the the time and resources available for KSP2.

CommNet is a pretty good analogy actually. A meaningful LS system would have to be approximately as complex. Now CommNet is based on two fundamentally simple ideas: signal strength and occlusion. Yet when you start thinking about everything that's gone into it, you'll realise just how complex the system is and just how much work went into implementing it.

• The UI: first hop, vessel links, network. In map view in vessel, and in the tracking station.
• What happens when a probe core loses the connection? What about a command pod with a non-pilot in control? What about a combination of the two?
• What about sending messages (Science) back home?
• What parts are needed for it? Where should they be in the tech tree? What should their models look like?
• We have three types of antennas: built-in, relay, and regular. What are their characteristics and how should they be balanced against each other?

Add on top of this kind of thing the stuff you would really really want in a LS UI (like the alarms I mentioned above), and you should see pretty quickly that it gets pretty big pretty fast.

CommNet would've been pointless if had just been about slapping on an antenna. It needs all that stuff to be worthwhile. LS is the same.

[XLjedi]

@Brikoleur  LOL, OK...   I don't presuppose what may or may not be too difficult for developers to implement.  They seem to be pretty sharp folks!

For the sake of "wishlist-type" discussions, I'm pretty OK assuming they're clever enough to tackle the challenge.  ...then they can tell us later that they saw people wanted it, but it was just too hard or time constrained, or whatever and they just didn't have the time to implement it.

I think they could implement supply-lines and piggy-back off the comnet structure.

[Jimmidii]

7 minutes ago, Brikoleur said:

Yes it certainly could, but it would require designing and implementing systems with that complexity, which, as I seem to have to keep repeating, would likely be too ambitious given the the time and resources available for KSP2.

CommNet is a pretty good analogy actually. A meaningful LS system would have to be approximately as complex. Now CommNet is based on two fundamentally simple ideas: signal strength and occlusion. Yet when you start thinking about everything that's gone into it, you'll realise just how complex the system is and just how much work went into implementing it.

• The UI: first hop, vessel links, network. In map view in vessel, and in the tracking station.
• What happens when a probe core loses the connection? What about a command pod with a non-pilot in control? What about a combination of the two?
• What about sending messages (Science) back home?
• What parts are needed for it? Where should they be in the tech tree? What should their models look like?
• We have three types of antennas: built-in, relay, and regular. What are their characteristics and how should they be balanced against each other?

Add on top of this kind of thing the stuff you would really really want in a LS UI (like the alarms I mentioned above), and you should see pretty quickly that it gets pretty big pretty fast.

CommNet would've been pointless if had just been about slapping on an antenna. It needs all that stuff to be worthwhile. LS is the same.

What's convinced you that they haven't already been working on an enjoyable LS implementation from the start?

They've obviously gone into development of this with their eyes open, with new multiplayer, interstellar, and colony systems already confirmed.

If LS is implemented, it's almost certainly not because they've suddenly decided to tack it on now.

[Guest]

1 minute ago, XLjedi said:

too difficult

Not this again.

I don't think it's too difficult. I think they won't have the time to do it right, what with everything else on their plate, and I'd rather see LS cut than half-baked, or implemented at the cost of base gameplay or the colony system.

1 minute ago, Jimmidii said:

What's convinced you that they haven't already been working on an enjoyable LS implementation from the start?

Nothing. What's convinced you that I'm convinced about that?

[Jimmidii]

Just now, Brikoleur said:

Not this again.

I don't think it's too difficult. I think they won't have the time to do it right, what with everything else on their plate, and I'd rather see LS cut than half-baked, or implemented at the cost of base gameplay or the colony system.

Nothing. What's convinced you that I'm convinced about that?

Your main argument against LS now, that you're constantly repeating, is that it's too ambitious for the devs to add it with the time and resources they have left.

If you're not convinced of that, what do you have against LS being in the game?

[XLjedi]

Just now, Brikoleur said:

I think they won't have the time to do it right, what with everything else on their plate, and I'd rather see LS cut than half-baked, or implemented at the cost of base gameplay or the colony system.

OK then, to further clarify...   I'm perfectly OK assuming the opposite of your quoted statement. 

Particularly as it relates to semi-hypothetical wishlist discussions.

[Guest]

1 minute ago, Jimmidii said:

Your main argument against LS now, that you're constantly repeating, is that it's too ambitious for the devs to add it with the time and resources they have left.

If you're not convinced of that, what do you have against LS being in the game?

"Too difficult" is not the same thing as "too ambitious given the time and budget." The former is an assumption about ability, the latter about resources.

Thank you for correctly paraphrasing my argument, I'm happy someone did! 

[Guest]

Again, they already have to solve the consumables problems for the planetary VAB, the Shipyard, the BAE and the advanced fuels manifacturing even if colonies just work like fancy crew containers with no gameplay attached to it (which we know it won't be the case), LS is not different.

All of this, again, presuming that LS can only be of the "Kerbal Fuel" kind.

Edited September 12, 2019 by Guest

[XLjedi]

6 minutes ago, Brikoleur said:

"Too difficult" is not the same thing as "too ambitious given the time and budget." The former is an assumption about ability, the latter about resources.

Thank you for correctly paraphrasing my argument, I'm happy someone did! 

OK, so...  provided it's not "too ambitious given the time and budget" then you're OK with LS being implemented?

[Guest]

5 minutes ago, XLjedi said:

OK, so...  provided it's not "too ambitious given the time and budget" then you're OK with LS being implemented?

Absolutely, I would love it!

Edited September 12, 2019 by Guest

[XLjedi]

Just now, Brikoleur said:

Absolutely!

Excellent... noted.   I think we can safely move on now to what features folks might like to see in an implementation of a supply-line or LS system.

[Yakuzi]

On 9/11/2019 at 2:23 PM, Lu K. said:

oh where was that? i'd like to see, i must have missed a lot of info released during this latest convention thingo.

 

According to creative director Nate Simpson, there is going to be some form of life support:

Interviewer: Will there be such a thing as life support systems so you have to ensure oxygen supply, water supply and food supply for the Kerbals?
Nate: I can say so much that the need to keep Kerbals alive is a feature we're going to introduce now. But I can not say more about that at this point, but I can say so much. In case you tried mods on life support systems; it will not be so detailed, but as I said I can not say too much because there are a few secrets.

source (in Kerman). Complete transcript on the forums here, courtesy of @nikokespprfan

 

On 9/12/2019 at 12:15 PM, Brikoleur said:

The difference is that you can just not build orbital bases, or only build them for very specific and very limited uses, and use bases on low-gravity moons for construction instead, whereas you can't just ignore LS. I.e. if they screw up LS, there's no way around the grind, whereas if they screw up supply for orbital bases, that's just one and largely ignorable gameplay feature that's kind of broken.

(Of course it's possible they'll also screw up supply for surface bases but I doubt that; ISRU is a thing and it would be very weird if they don't build surface base gameplay around that.)

I understand your concerns about LS being implemented properly to avoid grind. On the other hand, LS is such an integrated part of space exploration, KSP2 would feel incomplete to me if it'd ship without it (as did KSP1). The current career mode is the perfect example of a dumbed down system that is leading to a grindfest, inhibits immersion and stifles creativity. I know that Star*Theory are overhauling the career/science system into something called "progression", and am cautiously hopeful that LS was part of the design process of this overhaul. If they can turn career into something fun and sensible, I'm sure they can do the same with LS. Also since multiplayer will be thing, implementing actual automated supply missions would not be an impossibility.

In the end, we can discuss this and get worked up about it until we turn blue green, but the best thing to do is wait what the devs will come up with and take it from there. I, for one, am very excited to find out what they've come up with!

Edited September 14, 2019 by Yakuzi

[Guest]

3 minutes ago, Yakuzi said:

The current career mode is the perfect example of a dumbed down system that is leading to a grindfest, inhibits immersion and stifles creativity.

My experience of it is kind of the opposite and I mostly play career.

TBF I have adjusted the difficulty settings to avoid the grind -- specifically, I turn up Funds rewards and turn down Science rewards, as most of the grind comes from lack of Funds when you really need to upgrade a building and I don't enjoy that.

Career does have its issues (in particular, most contracts are just pointless; fortunately there are so many that it's possible to pick only the ones you're interested in) and it is clearly just barely one step up from placeholder, but it is one step up from that and for me at least gets the job done -- I still set my own large-scale objectives, but it sets small-scale ones that can be worthwhile, and most importantly it sets constraints that make otherwise trivial design challenges interesting again -- stimulating creativity.

[Bartybum]

Moving aside all discussion of available developer resources and whatnot

(Assuming LS is optional of course, when there are those who don't want to have to deal with it)

Regarding LS on the ship scale, I'm really in favour of a simple LS system ala the Snacks! mod, where you have food, waste/fertilizer, growers and recyclers. To me it's exactly the same as needing to carry solar panels to generate electricity.

The complexity really comes along when you start thinking on the scale of installations. I really like the idea of having to personally fly a resource supply line with a capable ship, then being able to automate it to fly at optimal dV windows (think porkchop plots). By endgame, having tens of installations is going to be extremely tedious if even if you only have to fly a single resupply mission per station per year, so automated colony management is key imo.

[XLjedi]

@Bartybum

I would love to be able to describe my own mission parameters.  On a relatively simple check-the-box scale...  something like just set the base you want to goto with whatever rocket/cargo.  Once you demonstrate that a mission can be done, the result can then be automated to just rerun the same mission cost/benefit when a valid launch window is available.  Maybe a dropdown list of your own resupply missions with red/green indicators.  ...and I'd be fine with allowing those to just be auto-completed with a % complete bar to properly account for the time it would take to complete the mission given the current window.

On the one hand, I don't want resupply to be tedious at all... but on the other, I really want to be able to demonstrate that at my current tech-level I could successfully run the mission to provide whatever supplies are needed.  I'd like to setup orbital refueling stations with a system like this.  Then as my tech improves maybe I can redo the supply mission to improve on the cost-per-ton of delivery and delete the older less efficient missions.  I'd like to be able to manage my own system of supply-line missions at the strategic level, but establish each mission as viable at the tactical level.

I acknowledge that I do kinda group together life support supply requirements with all types of supplies in general...  and see the overall logistical challenge as less of an individual ship thing, and more of a proximity to a supplied base thing.

 

Edited September 13, 2019 by XLjedi

[Drakenred65]

On 8/31/2019 at 4:10 AM, tseitsei89 said:

I dont want to be sending constant resupply flights to all 25+ bases BUT that is why I would make the colonies self sustaining by collecting resource on the planet. But achieving that self sustainability should (IMO) require more than just slapping down a hitchiker can with a kerbal in it.

Option to turn it off is of course great and much needed to make the game fun for as many people as possible

Who said anything about just slap down a hitchhiker can adding a drop of water and boom instant Industrial  Civilization that can rival China in production output.

they clearly said you will have to supply resources and equipment to do this.

the only real question is will it be like taking Rover dudes mods and deleting the automated spread sheets and just dropping x number of modules a b c d e and f to fill up timer bars on those while dropping cargo shuttles to speed things up, or will it require more detail than the real World economy to simulate.

[Yakuzi]

On 9/12/2019 at 5:26 PM, Brikoleur said:

My experience of it is kind of the opposite and I mostly play career.

TBF I have adjusted the difficulty settings to avoid the grind -- specifically, I turn up Funds rewards and turn down Science rewards, as most of the grind comes from lack of Funds when you really need to upgrade a building and I don't enjoy that.

Career does have its issues (in particular, most contracts are just pointless; fortunately there are so many that it's possible to pick only the ones you're interested in) and it is clearly just barely one step up from placeholder, but it is one step up from that and for me at least gets the job done -- I still set my own large-scale objectives, but it sets small-scale ones that can be worthwhile, and most importantly it sets constraints that make otherwise trivial design challenges interesting again -- stimulating creativity.

We all play KSP in our own way, and that's completely fine. In this case it seems we're different sides of the same coin. While I've failed to learn how to enjoy career mode, you apparently have and, granted, that takes creativity. But what it really boils down to, is that 'career' mode in KSP1 is indeed barely one step up from a placeholder.

Since Star*Theory have the chance to build KSP2 from the ground up, wouldn't you prefer progression mode that is fun because the challenges you have to overcome are actually immersive, rewarding and make sense? Which would include integrated space and planetary exploration, scientific discovery, investments in infrastructure development and maintenance (including LS), logistics to supply/maintain the infrastructure, and automated missions to avoid the supply grind. To me, this beats gimping a glorified career placeholder so it's not super grindy any day.

I fully realise this would take quite a bit of thought and time to implement, but Star*Theory have been working on KSP for a while (plus they had an experienced dev team from the get-go and are asking a decent amount of money for their product). Do I expect this to be implemented? No. But I'm not ruling it out either, particularly since the devs haven't even given us any detailed info yet.

In the meantime, I'm keeping my fingers crossed!

Edited September 14, 2019 by Yakuzi

[Arugela]

I'll put this simply. A video game does not require one answer. You can have all answers. It's called customization. IE settings. Just make settings for everything so people can choose. Then you get more game out of it. 8\

[Dr. Jet]

Do not oversimplify Life Support, please! 

COMPLEX life support will call for more complex ISRU sysytems when building a self-sufficient base. More building. More transportation. More management. More fun.