Jump to content

[1.3.0] Kerbalism v1.2.9


Recommended Posts

So I found that food and water consumption were too high for a RSS game experience. 

And I think a patch for KSPI is really needed.

Those large antenna in KSPI with light-years range do not work in kerbalism. A small patch for that would be nice!!!

And the MTBF of parts , they are just not too low for RSS, maybe you can adjust that

 

 

 

anyway Thx for this fantastic mod

Edited by Iso-Polaris
Link to post
Share on other sites

TAC Life Support GUI shows an accurate amount of time left before I run out of food, water, and oxygen. The Kerbalism GUI shows a much shorter time, but moves at a really slow rate. Is there a way to fix this so they sync? I get an alarm for low water when I actually have plenty left according to TAC LS GUI. Thoughts?

Link to post
Share on other sites

On a different subject, would it be possible to to add airlock functionality to Kerbalism?

Historically, only a few space vessels have had dedicated airlocks. To EVA, it was necessary to vent the entire capsule. 

What I'd suggest is taking the volume of the part the kerbonaut is exiting from, and then dumping oxygen and nitrogen to match.

Airlocks don't have a lot of volume, so you don't lose a lot of atmosphere doing this. As far as I know, real-life spacecraft airlocks vent their atmosphere overboard to depressurize. This, of course, is done because an air pump is a lot heavier than the air you're likely to dump for a good long while. 

Or is there a way to automate this with the automation module? 

Link to post
Share on other sites

@John Nowak Originally the pressurization was meant to include that. Entering/exiting would have required the pressure to be equal, so that you would had to move to an 'airlock habitat', isolate it from the rest of the vessel, equalize its atmosphere with the outside, then eva out. Likewise to eva in, the opposite process. From the exterior it would also been possible to 'vent' the whole vessel, to enter in uncontrolled ones (a similar valve is present on the ISS). On simple vessels (like, a moon lander) where there was a single habitat part, the only solution would have been to vent it before going on EVA, then re-pressurize it from scratch after the EVA.

It wasn't implemented due to: time contrains (I was developing a lot of other things), unclear how to manage 'isolation' between habitat parts, and general lack of airlock parts.

Edited by ShotgunNinja
Link to post
Share on other sites
1 hour ago, smckamey19 said:

I get an alarm for low water when I actually have plenty left according to TAC LS GUI. Thoughts?

Mmm, I was thinking more about it. Probably, given that TAC does its own 'post-facto' simulation, it is showing to you the Water level including all its own things. These things, Kerbalism can't be aware of, nor they result among the vessel resources when it is unloaded, because TAC simulation is fundamentally different from my own. So, what this mean at the end of the day? That if Kerbalism and TAC share some resources (like Water in this example), you are going to have problems bigger than wrong estimates.

Link to post
Share on other sites

@ShotgunNinja Are you by chance interested in a small suite of Mk2 parts dedicated to the life support experience? For Kerbalism by itself I think I have everything figured out (what each of my parts will hold or do) but Kerbalism + TAC complicates things. I've grouped Nitrogen and Oxygen under Essentials; Carbon Dioxide and Ammonia under Wastes; and just happen to have a neutral place to put Hydrogen since I don't know which group it goes into. I'm also intent on providing some functionality apparently not already provided by Kerbalism.

Link to post
Share on other sites

@JadeOfMaar Absolutely interested. What parts do you have in mind, and how can I help you?

11 minutes ago, JadeOfMaar said:

Kerbalism + TAC complicates things

I would suggest not to worry about it, and make them for the default profile first (I just found out that non-trivial interaction between Kerbalism + TAC are fundamentally broken beyond hope of fixing).

20 minutes ago, JadeOfMaar said:

I'm also intent on providing some functionality apparently not already provided by Kerbalism.

You mean code-wise, or something that could be made with new processes?

I'm adding a couple new ones in next version: a WasteCompressor that produce Shielding, and a MonopropFuelCell. They are already on github here. I've also received a WasteIncinerator process as a contribution, that maybe will end up in the ISRU chemical plants.

 

Link to post
Share on other sites

 @ShotgunNinja The presumed new feature was an atmospheric harvester as I'm exploring your mod right now and haven't seen any part that hints at being it. I can't do much in terms of coding (I can't do DLL magic) so I might need you to program the features yourself if you prefer not to have my parts use Firespitter for resource switching. So far I can do well with MM patches and have made my mod, Airline Kuisine, compatible with the other major life support systems.

Airline Kuisine provides 9 parts in Mk2 body shape:

  • 2x long tanks for holding goods and wastes
  • 3x short tanks for holding goods and wastes (one of which is switchable between everything)
  • 3x processors (short tank size) for injecting converter modules
  • a modified crew cabin that holds only 2 kerbals explicitly so there's more room for intangible things like comforts/ hab enhancement

I've reserved two procssors already. One for all air functions and one for standard convert-o-tron functions but the third is blank and can become the Monoprop fuel cell.

Edited by JadeOfMaar
browser being weird. couldn't do line breaks and complete my post
Link to post
Share on other sites

@JadeOfMaar That is nice :) 

The Harvester module can do atmospheric harvesting, when 'type' is set to 2 (the valid values are the same as in the stock harvester). You can have multiple switchable harvesters (each one for a specific resource) in a single part by using the Configure module (that look messy to use at first, but it grows on you). You can see an example of how this is used here, where a configurable atmospheric harvester is added to the stock atmospheric sensor part.

 

Edited by ShotgunNinja
oops, type = 2 for atmo harvesting
Link to post
Share on other sites
On 2/13/2017 at 8:10 PM, podbaydoor said:

Yes, in 1.1.9 the capacity of processes (scrubbers, waste processors, pressurizers) scales with the crew capacity of the part that it's set up on. So for every crew slot on a part, the part's processors cover 5/3 of a Kerbal. The old values were fixed so that a processor always covered 4 kerbals, but people who had big part mods were running into trouble. I ran into the exact same problem as you did and I assumed it was because the capacity of the scrubbers is hardcoded in the save file, and the old 1x capacity scrubbers you have on your vessels now only clears enough COfor one Kerbal. Just as I was testing some fixes, it turned out restarting the game fixed it for me. If that voodoo solution fails, you can edit the "amount" of the _Scrubber pseudo-resources in your save.

I was playing a bit today and discovered this issue is more insidious than it appeared at first. Launching new vessels, I was still having less scrubber capacity than I should. It turns out the scrubber capacity is specified in the .craft file so not only do you have to change the amount of the _scrubber pseudoresource to 1.67 * [part crew capacity] in your save file, you need to do this to your .craft files as well. Or just remove the part in the VAB and replace it with a new version of itself.

Link to post
Share on other sites

@ShotgunNinja I've got a modest grasp on that example code for configuring parts. Just looking at parts and modules in the VAB I can see why people loves this so much. :D 

I'd personally add atmosphere and ocean harvesting to intake parts and replace their default tankage with a converter-- namely anything that calls itself a precooler as they already serve the purpose of drawing air at superb rates, compressing that air and providing tankage for something. I'm not expecting all this in a tiny science part.

When I get to coding I'll bring other questions like: Does the efficiency or processing rate scale with the volume of the part? The syntax seems so simplified that this seems to be the case. Also, do I have to delete the profile cfgs I'm not using? I have them all in and I guess that's why I can't actually configure anything in flight and can't configure the sensor part at all.

Edited by JadeOfMaar
Link to post
Share on other sites
15 minutes ago, JadeOfMaar said:

Does the efficiency or processing rate scale with the volume of the part

There is no scaling of the rates in code, rather they are specified in the configs. In there you can do some scaling using MM-fu. For example I scale the ECLSS rates using crew capacity.

18 minutes ago, JadeOfMaar said:

Also, do I have to delete the profile cfgs I'm not using?

Multiple profiles can coexists, but only the one selected in Settings.cfg is used.

 

@podbaydoor Yeah, sorry about that :( You could also switch setup in the configure window, that should re-setup the right amount of resource in theory.

Link to post
Share on other sites
55 minutes ago, ShotgunNinja said:

@eberkain I'm using NASA estimates for a human.

Does the rate scale with the switch from 6 hour to 24 hour day?  

I found this article on NASA.gov that says an astronaut on the ISS uses about 2.5kg or food and 11 liters of water per 24 hours.  In the Default profile config I see a note that says 1.77 Kg per-day for food, I'm not following what the rate/interval/degeneration numbers mean.   A Mk1-2 Command pod carries 94.5 food internally, about 27kg.  That should be enough food for the three passengers to eat on for 3.6 days (86 hours) according to the numbers in the article I cited, or 5 days (120 hours) at the 1.77 rate from the config.  In-game the Kerbalism planner says the food supply will last 30 hours, which is the same as 5 six-hour days.  Looking at that it seems that the rate actually is much higher than it should be when you go by hours passed.  Is there something important I'm missing?  

Link to post
Share on other sites
2 hours ago, eberkain said:

Does the rate scale with the switch from 6 hour to 24 hour day

It doesn't, that's probably the issue.

In default profile each kerbal eat 1.77 Kg of food every 6h. Irregardless of your KSP day-length setting.

Also everywhere the UI is telling you 'days', it mean 6h or 24h depending on your KSP day-length setting.

So in your example:

  • crew of 1, consume 1.77 Kg every 6h
  • pod has 27 Kg
  • with 6h day-length setting, the estimate say 15.25 'days' left (it mean: 91.52 hours)
  • with 24h day-length setting the estimate say 3.81 'days' left (it mean: 91.52 hours)
Link to post
Share on other sites
2 hours ago, ShotgunNinja said:

It doesn't, that's probably the issue.

In default profile each kerbal eat 1.77 Kg of food every 6h. Irregardless of your KSP day-length setting.

Also everywhere the UI is telling you 'days', it mean 6h or 24h depending on your KSP day-length setting.

So in your example:

  • crew of 1, consume 1.77 Kg every 6h
  • pod has 27 Kg
  • with 6h day-length setting, the estimate say 15.25 'days' left (it mean: 91.52 hours)
  • with 24h day-length setting the estimate say 3.81 'days' left (it mean: 91.52 hours)

So to relate that to the real world.  1.77 Kg every 6 hours is 7.08 Kg per 24 hour period.  If you divide that by the typical three meals we eat on earth, you get 2.36 Kg per meal.   That is more than 5 pounds.  That would be like eating three of these a day, plus some side dishes.  I suspect that the NASA figures for consumption were listed per day, and that was applied to the Kerbin days (which would be a 400% increase when looking at the actual passage of time).  I would argue that it should be balanced around the hours instead of the days since the hour is the largest measure of time that is the same in both the Kerbin time scale and Earth time scale.   But I see it both ways, at the stock scale with the current default consumption rates you get usage that makes sense in the context, but scaling up to 10x for RSS and using the default consumption rates I can see why it would seem like they are too high when you start looking at how much gets consumed per orbit, etc..  It would either take another 'realscale' profile, or a way to just flat scale all consumption to like 25%. 

Edited by eberkain
Link to post
Share on other sites
6 minutes ago, eberkain said:

But I see it both ways, at the stock scale with the current default consumption rates you get usage that makes sense in the context,

That was my reasoning. Essentially I went for 1 kerbin day = 1 earth day, and kerbal = human.

9 minutes ago, eberkain said:

but scaling up to 10x for RSS and using the default consumption rates I can see why it would seem like they are too high when you start looking at how much gets consumed per orbit, etc..  It would either take another 'realscale' profile, or a way to just flat scale all consumption to like 25%. 

I could rescale rules automatically from day-lenght. However, all process/module would have to be rescaled as well. There will be subtle changes in balance too, for example EC consumption reduced to 25% against unchanged solar panels, etc. Is probably best to write a new profile for RSS.

17 minutes ago, Calvin_Maclure said:

Hi,

fantastic mod! I was wondering if there is a list of all compatible/incompatible mods for Kerbalism? If something about this has already been posted on this forum, please point me towards it. 

Cheers!

There is a small list in the OP, but most mods should work without issues. Some aren't simulated in backround (like KSP-IE), but otherwise work.

Link to post
Share on other sites
3 hours ago, Calvin_Maclure said:

fantastic mod! I was wondering if there is a list of all compatible/incompatible mods for Kerbalism? If something about this has already been posted on this forum, please point me towards it.

3 hours ago, ShotgunNinja said:

There is a small list in the OP, but most mods should work without issues. Some aren't simulated in backround (like KSP-IE), but otherwise work.

There's an issue with the recently-released UpgradesUIextensions that I've reported on GitHub.

 

Link to post
Share on other sites

 

23 hours ago, ShotgunNinja said:

@John Nowak Originally the pressurization was meant to include that. Entering/exiting would have required the pressure to be equal, so that you would had to move to an 'airlock habitat', isolate it from the rest of the vessel, equalize its atmosphere with the outside, then eva out. Likewise to eva in, the opposite process. ...

It wasn't implemented due to: time contrains (I was developing a lot of other things), unclear how to manage 'isolation' between habitat parts, and general lack of airlock parts.

 

I can certainly understand priorities, but I suspect that all you really need is to dump the air in the part the kerbal is EVA-ing from. The "Enable habitat" controls seem to work in a similar manner, requiring the user to enable or disable habitats on a part-by-part level, as though linked parts were split by pressure bulkheads. Which seems like a reasonable assumption to me. 

Suppose I were to move kerbals into (say) an airlock, disable the habitat, re-enable it when they returned, and just pretend that I can't move them from an enabled habitat to a disabled one. Would I lose more or less the same amount of nitrogen and oxygen?

 

 

Link to post
Share on other sites

@John Nowak

This is rather complex, and I just woke up, so please bear with me.

Suppose it was possible to split a vessel into multiple habitats, even if only an 'airlock' and a 'main' one. Now all habitat properties have to be valid and defined for each one of them. This is not possible with the stock resource system, as you would effectively have multiple 'sub-vessels'.

The next logical step would be to throw away the pseudo-resources in habitat, and implement the same in a more complex way instead. That would be possible, but it would also be a less flexible system. How much less flexible? I'll make an example: you can create a 'waste compressor' that produce shielding, using a stock module resource converter or equivalent, right now. This kind of thing would not be possible anymore if I go this route.

Another solution would be to implement 'sub-vessel resources'. The resource cache system could be extended in such direction. However, what about anything that doesn't use it, like for example non-kerbalism modules in loaded vessels? Those will be unaware of the sub-vessels.

That's why I didn't dwelve futher in this matter at the time the pressurization was introduced. Instead, I went around these issues with some minor tweaks:

  • the habitat part 'lock/unlock' was renamed to 'enable/disable'
  • there is only a single habitat in the vessel at all time, composed of all enabled habitat parts
  • transfering crew inside a disabled habitat will automatically enable it
  • you can't disable an habitat where some crew is inside it
Edited by ShotgunNinja
Link to post
Share on other sites

Kerbals are 64% scale, so the human number should be too high. That makes a kerbal volume on the order of ~1/3 or less of a human. Food and air use could be set to anything, I suppose, but the combination of a crazy short day, and the tiny size of kerbals means that the mass of food/water, and even O2, should be substantially lower if they scale accordingly. A kerbal might mass 20kg? If a typical human is 60-70, that would put kerbal food needs closer to 500 g per day, and maybe 3+ l of water.

Edited by tater
Link to post
Share on other sites

I found the issue with control loss and flags.

Flags are vessel, and apparently they have vesselID = 0. The vessel cache is a FIFO one, the oldest entry is removed and recomputed. The code that checked the oldest entry had a special meaning for vesselID = 0. So as soon as a flag became the oldest entry in the cache, the whole cache stopped recomputing.

This was evident after EVAing a vessel without antenna. Before the EVA, the vessel cache entry was controllable (because manned). When it went outside, the vessel cache was updated to uncontrollable (because unmanned and no antenna). After planting the flag, the cache stopped being update shortly afterward. Then on re-entering the vessel, it still figured as uncontrollable according to the cache, leading to control loss.

Special thanks to @blakemw for providing me with a test case, and to Bob Kerman for landing on the mun a hundred times.

 

Link to post
Share on other sites
Guest
This topic is now closed to further replies.
×
×
  • Create New...