totm may 2024 [1.12.x] - Modular Kolonization System (MKS)

[RoverDude]

2 hours ago, Murdabenne said:

How will this interact with something time oriented like KCT?

No idea.  Certainly there is no consideration for it, if that's the question.

[DoktorKrogg]

3 hours ago, Murdabenne said:

How will this interact with something time oriented like KCT?

By KCT I assume you mean Kerbal Construction Time. If so, then we have no plans to officially support that mod. I see that as another example of a mod that aims to make KSP more of a simulation than a game and that just isn't really the focus in USI. Like I said in regard to transport routes in WOLF, I'm happy to provide hooks in Konstruction for other mods to inject their own behaviors (like introducing a delay between queuing up a vessel to be built and actually spawning it in game). It would fall on the maintainer of Kerbal Construction Time to actually make that mod "Konstruction-aware" though and do all the work of managing the build queue, providing a UI for it, etc.

[Plomie]

The mod works know! Thanks amazing mod best mod ksp probably has ever known

[CateredCarrot]

On 1/8/2021 at 6:08 PM, DoktorKrogg said:

Solving this problem is non-trivial. It would mean adding a bunch more code to check for the types of fuel used on the vessel, tracking fuel burn, creating multiple types of TransportCredits and determining which ones were used to complete the route (i.e. what if multiple fuels were used along the route?), what to do about stages that are jettisoned along the way (which type of TransportCredit does that cost?), etc. Ultimately this boils down to "is KSP a simulator or a game?" and everyone has their own opinion on that. @RoverDude and I tend to fall on the "it's a game" side of that argument and gravitate more toward approximation vs. ultra-realism because it makes our lives easier and lets us get more stuff out the door in our limited free time. 

If someone wants to make a more realistic transportation system that ties into WOLF, I'd be happy to answer specific implementation questions (by "specific" I mean don't ask me "how do I make a mod?" ). It's already fairly extensible and we're open to accepting PRs like @TheSuMa mentioned to make private methods protected, etc. to improve the extensibility.

I'm not categorically opposed to this idea but I'm also not convinced that it is an essential feature. Ultimately it sounds like you're suggesting an "oops" button but getting your crew capsule back from the Mun full of science only to realize you forgot to put a heat shield on it for re-entry is 2/3 of the "charm" of KSP. Planning and executing a "perfect" mission in KSP after 1,000 poorly planned/executed missions is one of the greatest sources of satisfaction (for me anyway). On the other hand, I'm not a fan of tedium. If I get to a point in my own saves where transport routes are a source of continual boredom, then I'll probably change my tune. I haven't hit that point yet though.

As mentioned previously, we are working on a new way to handle the payload portion of transport routes (i.e. having WOLF cargo crates with a fixed mass that will be used for the payload calculation instead of using the vessel mass at the end of the route). This should make the payload a lot easier to plan and not a guessing game like it is now.

About the new wolf containers, it made me remember the KSTS mod. It allows you to dump mass (eg fuel or ore) and count it as payload.

I am not a programmer and I had never took a look on its code, but it sounds like easier and more straightforward than having a specific part and functionality. This intends to be more a tip than a request.

 

On 1/7/2021 at 8:19 PM, DoktorKrogg said:

We are indeed working on a new system for building vessels in-situ and it will live in the Konstruction mod. It will use MatKits and SpecParts along with some new resources (and corresponding converters for them in MKS and hoppers for them in WOLF). The new resources were necessary to account for things like gravioli detectors whose mass is very low but cost is very high. Otherwise you could just dump a crate of MatKits on the runway, pump out a bunch of gravioli detectors and recover the vessel for a substantial profit. A big difference between our system and systems like GC is that vessels will instantly spawn nearby. Our reasoning is that the "build" time is baked into the assembly of the resources themselves. We don't see the simulation of a build process as being "fun" since you can (and most times would) just time warp through that part of the process anyway. It's still in development and all of this is subject to change at any time but that's where things currently stand. 

Thank you! I will just start from scratch the time comes them... lol 

I understand what you mean and I think it would be ok to increase complexity, but IMHO it would not be necessary to avoid “cost cheating” as you mention. If someone really wants to go this road than it just a matter of mining exotics from KSC. What I usually do for money is to setup exotics mining from the mun, so at least is a space business...
And I agree that it is not necessary to wait for building, although decreasing building time on GC also brings a bit more of a challenge to the game. 
And many thanks for supporting this mod and the constellation as well!

 

[Kerbals_of_Steel]

On 1/8/2021 at 2:12 AM, SpacedInvader said:

I'm returning to the game after a couple of years away and have decided As a result, I figured I'd better ask if 1.4.1 is a better option.

I used 1.4.1 with KSP 1.10 for a few days with no Ill effects.

 

KoS

[Murdabenne]

21 hours ago, DoktorKrogg said:

By KCT I assume you mean Kerbal Construction Time. If so, then we have no plans to officially support that mod. I see that as another example of a mod that aims to make KSP more of a simulation than a game and that just isn't really the focus in USI. Like I said in regard to transport routes in WOLF, I'm happy to provide hooks in Konstruction for other mods to inject their own behaviors (like introducing a delay between queuing up a vessel to be built and actually spawning it in game). It would fall on the maintainer of Kerbal Construction Time to actually make that mod "Konstruction-aware" though and do all the work of managing the build queue, providing a UI for it, etc.

Fair enough.  I disagree it making it less of a game and more simulator, but as a way of adding a fun level of complexity by adding that delay and thereby rewarding good strategy and advanced planning. Having an additional "oops" factor when you plan poorly at a strategic level is very Kerbal - example using USI-LS, if I strand a Kerbal did I plan well enough and have a in-process build  to rescue them, or fly the next mission if not needed as a rescue?  Or do I have to build the rescue from scratch, and if so do I have the parts and tech to make a design that can be built quickly?  Also this adds to your research challenges as well, because the delay in launch introduces more fun decision making.  For example - because of the commitment for a delayed launch, do I build what I have to get it sooner or do I wait until I have the next tech level for a "better" set of parts? Do I build a cheap and fast to construct rocket using "old" parts that are heavier and less capable but much faster to build, or do I take the longer build time with a more expensive cutting edge but more capable rocket? Should I use my optimal design  which will make me wait until another launch window, or would a less capable but quicker build enable me to get to the first available window? I find it fun to see just how much I can tweak my build and still fit it into the next launch window.

Adding a nice level of manageable complexity adds a lot to game-play -- IMHO that is why USI-LS and similar mods add to gameplay by adding the flavor of simulation, without the bulk of a true simulation.  That's unlike the more complex LS mods which become a game in and of themselves, or the overwhelming work like Kerbalism and/or Realism Overhaul and others impose (making it no fun for players like me who do not want a full simulation). This is one of the reasons I use mods like USI-LS and MKS/OKS - and KCT ; they are balanced to add complexity for better game-play and challenge, but do not destroy the "game" aspect of it in the process. 

I'll post the questions I have over to KCT to see maybe what kind of hooks and interface they might need to take on the build process that Konstruction is changing to, and incorporate it.  Things like EPL might be affected as well.

But opinions differ; I just hate having to choose between enjoyable mods. 
 

Edited January 10, 2021 by Murdabenne

[RoverDude]

@Murdabenne There's nothing wrong with agreeing to disagree   At the end of the day, modders make stuff to enhance their own gameplay experience.  And both @DoktorKrogg and I are in the camp of not liking the long construction delays and twiddling with that specific aspect - other folks are.  So while Konstruction will be bundled (and a first class citizen) in MKS and the USI Constellation as a whole, you can still use it side by side with other mods if that's more your flavor.   Just that you'll need to use those other mod's support channels vs USI, and it would be up to either that mod (or the community) to patch in any shims needed.

[Murdabenne]

I have a feeling that if it comes to it, I'll end up losing KCT, because the USI Constellation brings so much more to the table - Ive had to part ways with build systems before, EPL, GC, and a number of one that simply died off (I go back to beta, so Its been a while).  I like Nertea's stuff, and Kopernicus+OPM, USI, and a few others (mostly from linuxgurugamer's ever expanding orphanage for abandoned mods).  I consider the game not worth the bother unless I can run those; I get bored in a hurry without them.  Maybe I can find another complexity mod to take its place  

Just be assured your work for the mod as well as your KSP contributions are very appreciated - as a former lead software engineer for an aerospace company who started out as an online multiplayer game coder at a company long gone (Kesmai),  I have an idea just how much this takes in terms of brain cycles and time. Glad to see you still love your work.

[Terwin]

12 minutes ago, Murdabenne said:

I have a feeling that if it comes to it, I'll end up losing KCT, because the USI Constellation brings so much more to the table

What about KCT is not compatible with MKS?

I do not see any reason why it might not work 'out of the box' for kerbin based launches.

I would assume that KCT would have defaults for unfamiliar mod parts based on cost and weight, but even if they do not, I would be surprised if it takes more than a MM patch to add the required KCT metadata for MKS parts.(such a patch might even already be available for previously available MKS parts)

[Cruesoe]

2 hours ago, Murdabenne said:

<snip>  I like Nertea's stuff, and Kopernicus+OPM, USI, and a few others (mostly from linuxgurugamer's ever expanding orphanage for abandoned mods). 

Does Nertea's suite of mods work nicely with USI Constellation?

[Poodmund]

6 minutes ago, Cruesoe said:

Does Nertea's suite of mods work nicely with USI Constellation?

They have community made patches bundled with them to integrate functionality with USI to an extent. Obviously, this does not extend to WOLF at the time of posting.

Edited January 10, 2021 by Poodmund

[Cruesoe]

1 minute ago, Poodmund said:

They have community made patches bundled with them to integrate functionality with USI to an extent. Obviously, this does not extend to WOLF at the time of posting.

Thanks. I'm about to start my first MKS play through, I'm a Kerbalism fan usually. Hardest part of setting up a new mod collection is checking all the potential conflicts!

[TheDog]

1 minute ago, Cruesoe said:

Thanks. I'm about to start my first MKS play through, I'm a Kerbalism fan usually. Hardest part of setting up a new mod collection is checking all the potential conflicts!

Indeed.

I am running most of the USI stuff, most of Nertea's things, and many wildblue mods.

Nertea's mods, especially station parts, always fully supported USI-LS and integrate very well. With his latest things (farfuture) you have to carefully look if you want to switch over to the systemheat approach (heating topic) and the space dust (resource distribution topic). That could potentially conflict with usi, or you have 2 things running side by side.

Wildblue is a different story, but i have my own solution for that.

 

[Murdabenne]

4 hours ago, Terwin said:

What about KCT is not compatible with MKS?

I do not see any reason why it might not work 'out of the box' for kerbin based launches.

I would assume that KCT would have defaults for unfamiliar mod parts based on cost and weight, but even if they do not, I would be surprised if it takes more than a MM patch to add the required KCT metadata for MKS parts.(such a patch might even already be available for previously available MKS parts)

I was speaking to the revamp of the Konstruction system. From what I gather, it instantly builds (like stock VAB) on orbit , and its being revised to work into WOLF eventually.  I can see the reason for that - the time harvest/ship the raw materiel, turn them to kits/specialized-parts, etc, and move those to the orbital build facility can be seen as a "construction delay" sufficiently to have the same effect as KCT.  That mod basically takes the build system and applies their own delay and build queue, and If I recall correctly, some of the old orbital build systems did the construction over time as well.    We are going  a bit off topic since @DoktorKroggand @RoverDudepretty much answered my question and explained their design choice well, so I think any remaining questions can be discussed over on the KCT thread as a compatibility issue, and likely some smart person will make a MM script if it is possible.

For me, I'm content to just wait and see what new Konstruction brings to the table, and at that time see how other mods interact (or not) with the changes.

Edited January 10, 2021 by Murdabenne
clarification

[RoverDude]

2 hours ago, Cruesoe said:

Thanks. I'm about to start my first MKS play through, I'm a Kerbalism fan usually. Hardest part of setting up a new mod collection is checking all the potential conflicts!

Indeed.  My philosophy is that if you (a) use Stock as the yardstick and leverage it's systems, and (b) make it a point to not do wholesale module manager replacements, you're doing right by the community with your mod.

[Kielm]

Konstuction Pre-Release Bug:

Unresearched items are available for construction (e.g. the microwave power transceiver is available before unlocking it in CTT). 

Attempting to build these items costs resources but produces nothing.

Actually, I can't build anything. The logs says "[ERR 00:22:11.528] [ModuleInventoryPart]: Unable to create a snapshot part at index 0. Unable part name=battery-rad-125". 

I have a craft landed on Minmus with a Konfabricator included. MaterialKits are being produced in a Ranger Workshop and stored in an ILM. These MaterialKits are being deducted so I don't think that's the problem. The only cargo storage available is a Duna Logistics module and the Konfabricator. itself. There are multiple other craft within 150m if that makes a difference. Craft is powered and functional, two 2-star engineers on board (one in the Konfabricator itself). KIS is currently installed, would that conflict?

 

 

Edited January 11, 2021 by Kielm
context and clarity

[RoverDude]

hmm... if you launch a super simple vessel (KonFab, MatKit storage, crew, and something with inventory) on the launchpad does it work?

 

[SpacedInvader]

9 hours ago, Murdabenne said:

I have a feeling that if it comes to it, I'll end up losing KCT, because the USI Constellation brings so much more to the table - Ive had to part ways with build systems before, EPL, GC, and a number of one that simply died off (I go back to beta, so Its been a while).  I like Nertea's stuff, and Kopernicus+OPM, USI, and a few others (mostly from linuxgurugamer's ever expanding orphanage for abandoned mods).  I consider the game not worth the bother unless I can run those; I get bored in a hurry without them.  Maybe I can find another complexity mod to take its place  

Just be assured your work for the mod as well as your KSP contributions are very appreciated - as a former lead software engineer for an aerospace company who started out as an online multiplayer game coder at a company long gone (Kesmai),  I have an idea just how much this takes in terms of brain cycles and time. Glad to see you still love your work.

One thing you can do is "simulate" the KCT delay by using the KCT timeframe for the craft you want to build to set a KAC alarm for the correct date, then when the alarm goes off, construct the vessel at that time. IMO KCT brings too much to the table to completely remove it as you'll likely still use primarily the KSC to construct the majority of your vessels and only use Konstruction to build a relatively small number of craft offworld.

[Jasseji]

I just wanted to ask quick, is it me, or do you need a heckload of Transport Credits for any designs for rescaled solar systems ?

I tried to get TundraExp Falcon to Orbit, the display showed me 2200 transport credits required for orbit.

Is this normal or is it a question of adapting the other mods to work with WOLF ?

I am using TundraExp and JNSQ Solar System

Edited January 11, 2021 by Jasseji

[Kielm]

8 hours ago, RoverDude said:

hmm... if you launch a super simple vessel (KonFab, MatKit storage, crew, and something with inventory) on the launchpad does it work?

 

Yep, using a super simple craft on the launchpad does work. 

[TheSuMa]

32 minutes ago, Jasseji said:

I just wanted to ask quick, is it me, or do you need a heckload of Transport Credits for any designs for rescaled solar systems ?

I tried to get TundraExp Falcon to Orbit, the display showed me 2200 transport credits required for orbit.

Is this normal or is it a question of adapting the other mods to work with WOLF ?

I am using TundraExp and JNSQ Solar System

• Currently the Transport Credits required are only based on vessel mass lost between connecting to source depot and connecting to destination depot.
• Payload is currently also calculated by vessel mass (mass while connecting to destination depot)
• There is no logic present that accounts for solar system rescaling, calculation is static and hardcoded
• Transport routes are not cheap, even in stock they are expensive, i think this is by design

That said, yes, transport routes are more expensive with rescaled solar systems as you need more Dv to achieve orbit. But 2200 seems to be a lot. Probably your craft is simply too big to kickstart your WOLF routes. A vessel with fewer mass would probably fit better. Another thing to consider could be an SSTO spaceplane. At least for me they often win the transport credit / payload battle.

As the whole WOLF system is very new and the devs are open for suggestions and pull requests i would expect that there will be some changes in the future to account for such cases. Either a configurable cost-multiplier that other mods can change or automatically account for rescaled solar systems  in any way. If i find the time i can take a look at it once the currently open pull-requests are merged.

[Jasseji]

51 minutes ago, TheSuMa said:

• Currently the Transport Credits required are only based on vessel mass lost between connecting to source depot and connecting to destination depot.
• Payload is currently also calculated by vessel mass (mass while connecting to destination depot)
• There is no logic present that accounts for solar system rescaling, calculation is static and hardcoded
• Transport routes are not cheap, even in stock they are expensive, i think this is by design

That said, yes, transport routes are more expensive with rescaled solar systems as you need more Dv to achieve orbit. But 2200 seems to be a lot. Probably your craft is simply too big to kickstart your WOLF routes. A vessel with fewer mass would probably fit better. Another thing to consider could be an SSTO spaceplane. At least for me they often win the transport credit / payload battle.

As the whole WOLF system is very new and the devs are open for suggestions and pull requests i would expect that there will be some changes in the future to account for such cases. Either a configurable cost-multiplier that other mods can change or automatically account for rescaled solar systems  in any way. If i find the time i can take a look at it once the currently open pull-requests are merged.

Yes, i get the higher DV Requirement in Rescaled Systems, only i wasnt sure why the cost was so much higher compared to the Tutorial Rockets from the wiki but now i know and i have found out the reason - the Actual cost for a 100km Orbit in a x2.5 System is 159TC (for the Tundra Falcon 9/Ghidorah 9), the additional 2000 came from the Launch Tower... The dropped booster didnt raise the cost as far as i saw (although theoretically it should lower the cost but this would propably require integration into StageRecovery which i assume wont be in scope of this)

 

So it all makes sense now (back to actually trying to figure out how to produce all those Transport Credits :D)

Edited January 11, 2021 by Jasseji

[RoverDude]

@TheSuMa - sorry, should have reminded you to put any PRs against the DEVELOP branch vs MASTER as Dev is where we work on the next release.  I'll sort this first batch and reverse integrate for you since I think we're in a (relatively) safe place to do this, but putting it out there for any other PRs

 

[CateredCarrot]

Considering that Transport Credits is an abstraction of capability to move things around by building ships and filling them up with fuel, shouldn’t be possible to actually transport Transport Credits?

for instance, if I want move something from Kerbin surface to the Mun surface, it would be better suited and more efficient to build specialized ships for each leg, and I could only shift my payload to each ship. But if I cannot transport Transport Credits, I would need to build a ship to do the hole trip or building Transport Modules on Kerbin and/or Mun orbits. 

[RoverDude]

1 minute ago, CateredCarrot said:

Considering that Transport Credits is an abstraction of capability to move things around by building ships and filling them up with fuel, shouldn’t be possible to actually transport Transport Credits?

for instance, if I want move something from Kerbin surface to the Mun surface, it would be better suited and more efficient to build specialized ships for each leg, and I could only shift my payload to each ship. But if I cannot transport Transport Credits, I would need to build a ship to do the hole trip or building Transport Modules on Kerbin and/or Mun orbits. 

Transport credits are an abstraction of the infrastructure required to handle the logistics of transportation, which is why they are paid by the source.  And you already do make different legs for Kerbin to the Mun.  Do a route from Kerbin to Kerbin Orbit.  Then the leg from Kerbin Orbit to Mun Orbit.  And then your last leg to the surface.   This is already how it works.