Jump to content

Space colonies


Recommended Posts

1 hour ago, shdwlrd said:

You can do that now in KSP. You just need enough boosters, proper strutting, and proper staging.

Not with Xbox limitations. and If I tried serious part clipping, it would summon a new kraken just for that purpose.

Stratenblitz attempted a 1 kiloton craft and he said he had himself a slideshow of it

Link to comment
Share on other sites

23 minutes ago, Mikenike said:

Not with Xbox limitations. and If I tried serious part clipping, it would summon a new kraken just for that purpose.

Stratenblitz attempted a 1 kiloton craft and he said he had himself a slideshow of it

I've never said it wouldn't be a slide show. :D Even on a PC something that large would end up running at single digit frame rates. 

That's something that KSP2 is trying to fix. Also if you look at the ground colony parts, they are huge. If that is going to be the scale for some of the craft parts. You should be able to build something that heavy with much fewer parts; thus not turning the game into Kerbal Slide show Projector.

Link to comment
Share on other sites

46 minutes ago, Pthigrivi said:

It stands to reason that if they've got a scheme for doing repeat transfers of resources from colony to colony this same process could work for mining outposts.

Well the system was brought up as a "automation for resupply flights and other milk runs" and never as a specific "colony to colony" thing, I wouldn't be surprised if automating the refilling of a LKO fuel depot is the tutorial mission / example to introduce the new players to the feature.

On the other questions I would bet that the transfers are going to be "magic" but the crafts will be reusable at least in the KSP1 sense fo the term (refunding the whole cost when recovering).

 

46 minutes ago, Pthigrivi said:

On the science front I'm mainly worried about some of the insane grind KSP1 induces. One thing I would like to see automated in some way are the experiments themselves. Nate has said there will be lots of new ways to collect science and that sounds great unless we're manually clicking through dozens of parts every time we enter an area with new science.

I would be surprised if KSP2 works anything like KSP1 on this front and if it does it's 100% "keeping the KSP tradition". 

 

46 minutes ago, Pthigrivi said:

So maybe Im just talking about autonomous kerbals and not autonomous vehicles? That seems much more feasible. The flying and landing part of KSP is crazy fun. All the manual clicking and micromanagement is less so, and especially if we're working in hundreds of kerbals and new colony resource management mechanics there seems very little room for it. 

I bet that astronauts and colonist are 2 different kind of Kerbals at the very least or even a completely abstract "population/workforce" counter and there I expect the same thing as the rest, you control your astronauts directly like in KSP1 and the colonists in a more abstract way.

Obviously flying and landing in KSP, especially crafts you designed is were the most entertainment comes from, that's why I'm starting to think that KSP2 is going to be a lot like OTTD for me, a gameplay excuse to make more interesting and challenging messing with trains rockets. Basically de-abstracting the contract system / economy of KSP1 career mode (or your head-canon if you play sandbox).

Edited by Master39
sent 3 times due to a connection error.
Link to comment
Share on other sites

42 minutes ago, Master39 said:

On the other questions I would bet that the transfers are going to be "magic" but the crafts will be reusable at least in the KSP1 sense fo the term (refunding the whole cost when recovering).

I don't think it will be magic. I think it will behave in a similar fashion to the Davon supply mod. That way it will force you to expand out and create your own resources instead of relying on Kerbin for everything you need.

That's my take on Nate's vision for logistics anyway. 

Link to comment
Share on other sites

2 hours ago, Master39 said:

I bet that astronauts and colonist are 2 different kind of Kerbals at the very least or even a completely abstract "population/workforce" counter and there I expect the same thing as the rest, you control your astronauts directly like in KSP1 and the colonists in a more abstract way.

I guess what Im saying is KSP2 seems like a solid opportunity to remove us from manually controlling kerbals themselves in a 3rd person kind of way, and allow them to wander about on their own or be given commands to execute different tasks without us having to micromanage. You could quickly after landing let them disembark, stretch out their legs, and tell them "take sample here", "plant flag here", "examine this surface feature", or "place seismometer here" and watch them all go about their little tasks. It would speed things up, improve the user experience, and make them feel more alive, and because their physics and behavior is all known it doesn't seem like an insurmountable programming challenge. 

And like I said, from a purely cosmetic standpoint you're going to have these potentially huge colonies that could end up looking somewhat lifeless. At the very least having a certain number of kerbals wandering about on their own and goofing off or carrying out designated tasks would make the whole thing feel more real and active.

 

2 hours ago, Master39 said:

On the other questions I would bet that the transfers are going to be "magic" but the crafts will be reusable at least in the KSP1 sense fo the term (refunding the whole cost when recovering).

I do agree that this is probably the way it will work because its so much easier and less likely to result in bugs and problems. Still, MechJeb does a reasonably okay job auto-piloting custom craft, so it doesn't seem completely out of the question that mining rovers and supply vessels could be physically modeled and controlled. It's probably a lot to hope for but certainly has its gameplay advantages.

Edited by Pthigrivi
Link to comment
Share on other sites

7 hours ago, Pthigrivi said:

KSP2 seems like a solid opportunity to remove us from manually controlling kerbals themselves in a 3rd person kind of way, and allow them to wander about on their own or be given commands to execute different tasks without us having to micromanage. You could quickly after landing let them disembark, stretch out their legs, and tell them "take sample here", "plant flag here", "examine this surface feature", or "place seismometer here" and watch them all go about their little tasks. It would speed things up, improve the user experience, and make them feel more alive

If they improve on the possible EVA activities and make them more important I would love that, maybe while keeping the ability of direct control but not as the only option.

I would also love to see the crew assembling the modules or building new base parts when you use the BAE.

Same for crafts, it would make the colonies feel more alive, as you said, the only problem I can see is the performance one.

 

9 hours ago, shdwlrd said:

I don't think it will be magic. I think it will behave in a similar fashion to the Davon supply mod. That way it will force you to expand out and create your own resources instead of relying on Kerbin for everything you need.

That's my take on Nate's vision for logistics anyway. 

For "magic" I mean "you do the mission and then the system uses the data from that mission (costs and gains) to simulate other similar missions without actually spawning any CPU-controlled craft" as opposed to the same but with the crafts really being spawned and doing their thing.

Link to comment
Share on other sites

9 hours ago, Master39 said:

For "magic" I mean "you do the mission and then the system uses the data from that mission (costs and gains) to simulate other similar missions without actually spawning any CPU-controlled craft" as opposed to the same but with the crafts really being spawned and doing their thing.

Ok, I see what you mean. That makes me wonder how early game transfers will work. Surface to orbit and vice versa is easy enough. Planet to planet transfers I can see as a sticking point until you get better engines. 

Link to comment
Share on other sites

13 minutes ago, shdwlrd said:

Ok, I see what you mean. That makes me wonder how early game transfers will work. Surface to orbit and vice versa is easy enough. Planet to planet transfers I can see as a sticking point until you get better engines. 

I think it was mentioned we will get a panel the shows a porkchop selection or relevant data.

Using that its would be easy to calculate transfer times and dV costs. You could demonstrate a ships capabilities by reaching orbit with the max load a vessel can handle and saving the vessel weight and dV left from at the moment the vessel achieve orbit.

Essentially with something like this you should be able to list the mass of supplies desired and you could be given a slider that ranges from high cost (dV heavy) low transport time to low cost high transport time. Perhaps the supplies would even have to fit in the same fairing or to go even further we would have to build subassemblies to fit in the same fairing going on top of the same rocket if you wanted to simulate a physical ship or have overly simulated ship that never actually enters the game but the transfer still happens like a wire transfer...

Link to comment
Share on other sites

Nate mentioned periodicity in milk runs. That indicates that it'll look for transfer windows equal to or less expensive than your initial mission, which determines how frequently the milk run can be performed -- or something similar but cleverer, e.g. comparing somehow to the dV available in your craft.

Link to comment
Share on other sites

I wonder how this will work with re-usable/recoverable craft. I think its been said that we won't have "funds", but it does seem that we will need resources to build craft.

Will it be able to tell that a SSTO was recovered and adjust resource costs accordingly? What about 2 stage recoverable vehicles, will it be able to adjust costs accordingly? This last point may be quite important for supplying an orbital colony above Eve, because SSTOs even with BG are really really really impractical. I've had much more success with 2 stage recoverables, and an even haul up cargo that way in a mid size mk3 cargobay.

I'm really hoping that multistage reusables will work with their automated supply system, because they can bemuch more resource efficient (vessel size, propellant used) than single stage reusables.

Link to comment
Share on other sites

12 minutes ago, KerikBalm said:

I think its been said that we won't have "funds", but it does seem that we will need resources to build craft.

I believe this was assumed by some members on the forum when adventure mode was announced but was never explicitly stated (can't even remember if it was actually hinted)

 

EDIT: It was first asserted here I believe

 

Specifically:

On 5/29/2020 at 8:28 PM, Acid_Burn9 said:

   So today i stumbled upon recent PC Gamer article about KSP 2. I recommend to watch the videos, reviewing this article, before reading this post(or you can check out transcript somewhere on this forum). Videos can be found on ShadowZone and Matt Lowne YT channels. https://www.pcgamer.com/space-odyssey-our-first-big-look-at-kerbal-space-program-2/ I'm not gonna talk about the whole article, but rather about the part, that states, that KSP 2 will have what is called "Adventure Mode". From my understanding of this article, there is planned to be no funds/contract system in the game, because dev team considers, that it was too grindy in the original game. I disagree with this position. In fact the Career Mode is my favorite game mode for a very long time now. The only time i'm not playing it, is when i feel an urge to tinker and have fun with airplanes in sandbox, which is one of my hobbies (BTW this is why i really expect KSP 2 to improve in this area as well, but that's the story for another time), but when it comes to playing a campaign of my own Space Program - Career Mode is my only choice. I've never been an active user on this forum, or on any forum tbh, but the threat of not having this, EXTREMELY important feature, in the sequel, made me sit and write this post anyway. I really hope, that sharing my thoughts here, will make players give it a seconds chance and KSP 2 devs reconsider their position about adding it to the game, because it really deserves it.

It goes on at length...

looking into the article that is an assumption to which I find no basis. This is the paragraph of the article being referenced:

Quote

The goal of adventure mode is to provide a far more ambitious campaign for players to embark on, building ships powerful enough to leave the solar system. The structure will include specific missions, but creative director Nate Simpson says these will feel more “compelling” than some of the first game’s missions, which would direct you to fly to a specific latitude/ longitude and trigger a part on your ship. “Those felt grindy. We’re going out of our way to make the mission goals for adventure mode feel meaningful: real firsts that feel unique relative to every other goal in the game.”

To me that sounds like they just want more fulfilling missions and less of the ones we all dislike. See here for community opinion:

 

Edited by mcwaffles2003
Link to comment
Share on other sites

1 hour ago, mcwaffles2003 said:

I think it was mentioned we will get a panel the shows a porkchop selection or relevant data.

Using that its would be easy to calculate transfer times and dV costs. You could demonstrate a ships capabilities by reaching orbit with the max load a vessel can handle and saving the vessel weight and dV left from at the moment the vessel achieve orbit.

Essentially with something like this you should be able to list the mass of supplies desired and you could be given a slider that ranges from high cost (dV heavy) low transport time to low cost high transport time. Perhaps the supplies would even have to fit in the same fairing or to go even further we would have to build subassemblies to fit in the same fairing going on top of the same rocket if you wanted to simulate a physical ship or have overly simulated ship that never actually enters the game but the transfer still happens like a wire transfer...

Ok, I understand the surface to orbit part. What about orbit to orbit. Over the years playing with MKS and Pathfinder I've created a small collection of stock and modded freighters that are never ment to leave space. How would that affect the system? 

1 hour ago, Brikoleur said:

Nate mentioned periodicity in milk runs. That indicates that it'll look for transfer windows equal to or less expensive than your initial mission, which determines how frequently the milk run can be performed -- or something similar but cleverer, e.g. comparing somehow to the dV available in your craft.

So by that theory, I should do my initial transfer at the worst possible time to create a profile that I can launch a transfer anytime that is needed.

PS. I know this is getting more into the logistics side of colonies, but its still apart of colony operations, right?

Edited by shdwlrd
Link to comment
Share on other sites

1 minute ago, shdwlrd said:

Ok, I understand the surface to orbit part. What about orbit to orbit. Over the years playing with MKS and Pathfinder I've created a small collection of stock and modded freighters that are never ment to leave space. How would that affect the system?

Once a vehicle is in space it would be easy to limit the transfers it's capable of. The whole set of available transfers could be found in a 2d array like in mechjeb's porkchop plot or in transfer window planner:

Spoiler

TWPMainWindow.png

The crafts total dV in orbit would allow a selection only within the contour of [your ships dV  - dV required for capture at destination]. The plot can then be simplified then to a 1D array of soonest arrival per dV presented to the user as a slider. This, from which dV is given a maximum based on the tonnage you've required of the ship, can then just be shown as time per $ as the dV directly correlates to funds based on cost of fuel making the matter simple to the user.

Link to comment
Share on other sites

1 hour ago, shdwlrd said:

So by that theory, I should do my initial transfer at the worst possible time to create a profile that I can launch a transfer anytime that is needed.

PS. I know this is getting more into the logistics side of colonies, but its still apart of colony operations, right?

At the worst possible time that's still achievable with the craft of your choice, yes. Assuming it works like that.

I don't know if it's strictly part of colony operations or its own subsystem so that you could e.g. automate fuel runs from KSC to LKO and such.

Link to comment
Share on other sites

On 10/28/2020 at 12:08 PM, mcwaffles2003 said:

I believe this was assumed by some members on the forum when adventure mode was announced but was never explicitly stated (can't even remember if it was actually hinted)

I think it was also by inference from them describing adventure mode as being more like science mode than current career mode. Who knows if that was correct or if it still holds if it was. They may have meant structurally similar and not specifically similar in that one doesn't have funds. Im personally of the opinion that they could and maybe should remove funds from the game because ideally you'd be transitioning to colonial living-off-the-land gameplay somewhat early in the progression after which money means less and less. You're harvesting most and eventually all of your own resources after which there is no "payee".

At the same time maybe there are reasons you'd want an early-game resource that eventually phases out, or maybe the way funds remain relevant is through this auto-delivery system. If the transfers are "magic" they could be abstracted in the form of funds payments. That gets to some of the murkier complications with it though... 

On 10/28/2020 at 4:11 PM, shdwlrd said:

@mcwaffles2003 @Brikoleur So, both of you are saying the same thing. If I have a freighter that can brute force a planetary transfer, then it should be acceptable to automatically send a transfer anytime I need it to.

If the cost for the delivery was set to the total converted base cost of the transfer vehicle at its point of departure you'd have the option to either brute force the transfer and have it take less time or make it more efficient and take longer. That would kind of assume non-reusable craft though, so it would be nice if after demonstrating that the transfer vehicle returned to point A its recovery cost could be subtracted from future deliveries. This would help with 100% reusable, space-based freighters that do things like run fuel from Minmus to a station in LKO. You demonstrate a there-and-back run, after which you can order repeat deliveries equal to the net delivered fuel at LKO

I think it would get more complicated if you were trying to manage a Joolean local economy. Say for the sake of argument you can harvest Xenon on Laythe, Nuclear fuel on Tylo, and H2 and Ore for new rocket parts on Vall. You use a high-TWR H2 burning lifter to deliver uranium to orbit above Tylo. That fuel is then loaded onto an ion drive transfer vehicle which loads up Ore and H2 above Vall, and then drops off the Uranium and Ore at a primary dry-dock station above Laythe. The Ion-transfer vehicle then tops off Xenon and delivers the H2 to Tylo to power the Uranium lifter. How do you repeat that run? At each step there is fuel being consumed, different resources being onloaded and offloaded, transfer windows that may or may not synchronize, etc. I guess you'd need to break it into its constituent legs and time it that way? And then there's the payee question... does any of this cost funds? Because at this point every part of the process is being produced in the field by the player. 

Granted this latter example is exactly the kind of thing that excites me about the prospect of KSP2, but how all that is packaged and interfaced with is no small feat. 

Edited by Pthigrivi
Link to comment
Share on other sites

53 minutes ago, Pthigrivi said:

Granted this latter example is exactly the kind of thing that excites me about the prospect of KSP2, but how all that is packaged and interfaced with is no small feat. 

I was totally visualizing your example as a Factorio train.

Link to comment
Share on other sites

@Pthigrivi without quoting your whole post, yes you are correct, it would be complicated. But instead of managing the Joolian system, I'm thinking of the Kerbol system as a whole. Instead of dodging Jool when the moons are on the wrong side of their orbits, imagine the cost of dodging the sun and the possible situation where the planets are at the opposite sides of their solar orbits.

Now put the freighter in context, the stock one is designed for 300 tonnes of cargo with a dv of 3800 m/s, non nuclear. By reducing the cargo tonnage, disabling engines, changing the thrust values, you can significantly change the dv without physically altering the vehicle itself.

Here is where everything is hazy. How would a system handle a non-recoverable, extremely variable dv craft? What types of systems could handle that? I get the idea of distribution networks since I work within the transportation industry, but a lot of those variables are set. Certain equipment is used for certain jobs, the distances are set and don't change. How would you plan for the infinite types of equipment that could be used and the huge variations for the distances?

Link to comment
Share on other sites

15 hours ago, shdwlrd said:

 

Here is where everything is hazy. How would a system handle a non-recoverable, extremely variable dv craft? What types of systems could handle that? I get the idea of distribution networks since I work within the transportation industry, but a lot of those variables are set. Certain equipment is used for certain jobs, the distances are set and don't change. How would you plan for the infinite types of equipment that could be used and the huge variations for the distances?

This seems like the hardest part to me even if they are doing magic transfers. A lot of the calculus could happen in the back of house but it still has to be robust enough that players can't easily abuse the system and intuitive enough that they aren't getting wildly unexpected results. Without getting too speculative we know at least that colonies will have multiple resources and that automated milkruns will be part of how resources are transfered. This will mean designating some craft as colonies/stations, others as transfer vessels, and somehow doing 'proof' runs and saving them for future reuse. That really means we're going to need at least a couple UI screens to manage resource transfers: one that lets you see at a glance how much of each resource you have stockpiled at each of your colonies and stations, and another that lets you pick which saved transfers are active, how often they should repeat, and what their status is (active/inactive, enroute, awaiting window, etc.) I like what you guys are saying about dV and periodicity, that if you've done a proof run you could re-run it at the next soonest window with a similar dV, factoring flight time and return if its a round trip. 

So take my intentionally complicated Jool problem. You'd probably need transfer stations in Tylo and Vall orbit with some storage capacity. Your 'proof' run starts at Laythe, boosts up to Tylo, drops off H2 and picks up U. It then drops down to Vall, drops off some U, loads H2 and Ore. From there it drops back down to Laythe orbit and offloads all Ore and remaining U. Thats a complete run that can be saved and repeated any time similar dV windows are available. The trips from the surface of Tylo and Vall to the orbital transfer stations are separate legs that you could deliberately time to coincide, or at least balance depending on stockpile capacity. 

That's about as simple as it can be I think. You could of course get deeper with conditional transfers (wait till H2=X, then transfer) and @Master39 and I would love it, but it should be an optional optimization and not necessary for players to worry about early on. Like the kind of thing you start dabbling in and getting better at as you're building your first interstellar mothership.

Edited by Pthigrivi
Link to comment
Share on other sites

1 hour ago, Pthigrivi said:

So take my intentionally complicated Jool problem. You'd probably need transfer stations in Tylo and Vall orbit with some storage capacity. Your 'proof' run starts at Laythe, boosts up to Tylo, drops off H2 and picks up U. It then drops down to Vall, drops off some U, loads H2 and Ore. From there it drops back down to Laythe orbit and offloads all Ore and remaining U. Thats a complete run that can be saved and repeated any time similar dV windows are available. The trips from the surface of Tylo and Vall to the orbital transfer stations are separate legs that you could deliberately time to coincide, or at least balance depending on stockpile capacity. 

Why would you have to time them? You could just run both of those errands separately with 2 specialized craft instead of using a single generalized vessel to run 1 long complicated route.

Link to comment
Share on other sites

14 minutes ago, mcwaffles2003 said:

Why would you have to time them? You could just run both of those errands separately with 2 specialized craft instead of using a single generalized vessel to run 1 long complicated route.

Of course you could, but this is a thought experiment designed to imagine and test what different resource transfer schemes could and couldn’t handle. If a route returns to point A I don’t see any reason it couldn’t have multiple stops along the way, so long as the same qty of each resource was onloaded and offloaded at each stop. A similar thing might happen with a Minmus  fuel tanker that drops off some fuel at Munar orbit before descending to LKO. Even if the transfer isn’t physically modeled the timings of different resource transfers really should be constrained by flight durations and dV windows.   

Edited by Pthigrivi
Link to comment
Share on other sites

@Pthigrivi again without quoting your huge wall o' text. I do have to agree, it's very complicated thing to figure out the right balance of vehicles, distances and times.

I've done your Jool example on a smaller scale within the Kerbin system using MKS. Moving resources around and from the surfaces of the Mun and Minmus was easy. Moving resources from orbital stations between Kerbin, the Mun, and Minmus was kind of a pain, but doable. Where everything started to fall apart was when I setup a mining outpost on Duna. Trying to ferry resources from Duna to Kerbin to process them to usable stuff was nasty, but trying to setup a routes to ferry supplies to Duna so the life support doesn't run out was ridiculous. (The primary reason I made some freighters with huge transport capacities. Huge capacities, less transfers, reduced number of crafts, less overall costs.) I still had to do a few transfers a year, even when the transfer window wasn't optimal.

So please bare with me if I'm skeptical about any logistics system until the devs actually detail how it works. Because if you want your colonies running efficiently, you need to bring in supplies from another planet until they large enough to supply all the operations within that planets SOI.

Edited by shdwlrd
Stupid phone
Link to comment
Share on other sites

On 10/27/2020 at 7:34 PM, Mikenike said:

Not with Xbox limitations. and If I tried serious part clipping, it would summon a new kraken just for that purpose.

Stratenblitz attempted a 1 kiloton craft and he said he had himself a slideshow of it

1 kiloton is easy--- although we have to differentiate between launch mass and mass in orbit. 1 Kiloton launchmass is easy to manage, using the larges parts.  Just 4 Kerbodyne S4-512 Fuel Tank will get you over 1 kiloton.

This isn't a particularly large rocket, and its already nearly 700 tons (and recoverable)

Spoiler

6XKXPwE.png

 

 

My SSTOs for 3x rescale were routinely around 1 kiloton on the runway before I switched to 2 stage reusable designs and shaved off 200 tons while not needing to use nukes. I made them so large because payload fraction at 3x is much lower, and I wanted to maintain the ability to launch >100 ton payloads with recoverable/reusable systems.

When I tested my craft on 1x kerbin, I could do a single stage to jool intercept with a >100 ton payload, 986 ton starting mass.

Spoiler

Rcf6ODr.png

And similar chonky bois:

Spoiler

6vRlSxh.png

z7wgpqw.png

If KSP2 runs smoother, then I see no reason that 1 kiloton (starting) craft won't make it to orbit (where they can refuel, and reach 1 kt again).

I really really hope the supply runs to space colonies will be able to factor in supply runs with this mission architecture (done on 6.4x IIRC):

Spoiler

Airbreathing climb:

R7bUPq5.png

Seperation, orbiter pushes to orbit:

prw9mEC.png

kyQOQXQ.png

VCRwp8l.png

^note that the 1st stage is still quite high, and the 6.4x scaling does not apply to the atmosphere, which I only gave 1.25x scaling

2kNSXrq.png

Reentry:

klXz9GD.png

After turning around the carrier craft, high speed cruise back:

7HZixYg.png

 

Cia9FN4.png

Landing:

5oKvI5e.png

*taxid back to the runway from the flats, most times I land the thing on the runway

iWdAnTe.png

Then I can even have relatively efficient supply runs on much larger scales.

2 stage reusable is not really worth it on 1x kerbin... is better than SSTO at 3x scale (but doesn't blow them out of the water in terms of payload fraction), and is much much better at 6.4x scale (though I mostly play at 3x scale, I say others doing 6.4x and wanted to see if my 3-4x 2 stage designs were up to the challenge, and they were).

If such reusable multistage resupplies to orbital stations aren't modeled, I will be very disappointed.

Link to comment
Share on other sites

1 hour ago, KerikBalm said:

1 kiloton is easy--- although we have to differentiate between launch mass and mass in orbit. 1 Kiloton launchmass is easy to manage, using the larges parts.  Just 4 Kerbodyne S4-512 Fuel Tank will get you over 1 kiloton.

I meant to say one megaton......... 1 kiloton has been done by xbox a few times. In fact it was done with my SAT 5 replica, cuz part clipping. 

My sat 5 replica ended up being like 1.35 kt

Link to comment
Share on other sites

×
×
  • Create New...