[WIP][1.8.x] SSTULabs - Low Part Count Solutions (Orbiters, Landers, Lifters) - Dev Thread [11-18-18]

[Sudragon]

Just now, Jimbodiah said:

Did you modify any cfgs or add any patches? It could be a techtree issue in sstu, but I have not noticed it in a recent career playthrough (but I rarely use SRBs to be honest).

only the mechjeb.cfg. took out the techtree limiters.
I have SSTU Nova pack and SSTU Expansion, but -Expansion has been cut down to just including the 'no docking port' LC-2, -3 and -5 patches.

[Shadowmage]

2 hours ago, Sudragon said:

only the mechjeb.cfg. took out the techtree limiters.
I have SSTU Nova pack and SSTU Expansion, but -Expansion has been cut down to just including the 'no docking port' LC-2, -3 and -5 patches.

Its the SSTU Nova / SSTU expansion pack.

They screw up the tech-tree upgrade stuff because they haven't been updated in a long time.

Remove them and your problem will go away.

[Sudragon]

That fixed it, but what am I going to do for an array of 72 F-1's now? 

[TachyonGMZ]

I use the SMURFF mod to make my rocket have more realistic performance. I'm not sure if this is normal or it is not working properly with this mod. I made two rockets, one with the stock 2.5m pod, a stock fuel tank and a skipper and one with the apollo pod, a 1.50x sstu fuel tank and a skipper. The stock one had about 5600 dv while sstu had about 4400 dv. Exact same engine exact same fuel tank size and fuel type. Pod was different as mentioned. Any idea why this is happening?

[Pappystein]

4 minutes ago, TachyonGMZ said:

I use the SMURFF mod to make my rocket have more realistic performance. I'm not sure if this is normal or it is not working properly with this mod. I made two rockets, one with the stock 2.5m pod, a stock fuel tank and a skipper and one with the apollo pod, a 1.50x sstu fuel tank and a skipper. The stock one had about 5600 dv while sstu had about 4400 dv. Exact same engine exact same fuel tank size and fuel type. Pod was different as mentioned. Any idea why this is happening?

I would try the same test setup WITHOUT SMURF.  If the values end up being the same then the issue is with support for SSTU by SMURF and should be addressed in that forum. 

ALSO Please make certain your Masses are the same in the test builds as that directly affects dV

[T-10a]

SMURFF doesn't recognise the SSTU tanks as tanks, as the SSTU tanks do not use RESOURCE tags.

Try this patch:

@PART[SSTU-SC-TANK-MFT-*|SSTU-SC-TANK-MUS]:NEEDS[SMURFF]
{
	mass /= 4
}

This'll cut down the mass of the tanks by a good bit, enough to be inline with SMURFF.

Edited January 2, 2018 by T-10a

[TachyonGMZ]

What do I do with the patch and would sstu work with realfuels?

[T-10a]

Paste it into a new config file (name it SSTU-SMURFF.cfg), and put it in your GameData folder.

And SSTU does have a RealFuels config.

[TachyonGMZ]

can I just install realfuels through ckan or do I need to put the realfuels config somewhere. Will the realfuels config icrease delta v from sstu?

Edited January 2, 2018 by TachyonGMZ

[T-10a]

Yes, RealFuels is in ckan (I don't use it, as ISRU is not usable w/ it.). Delta-V increase is subjective. Due to the changes to use real densities and propellants, certain parts may either gain or lose (!) deltaV.

Edited January 2, 2018 by T-10a

[Shadowmage]

General development update:

Today's (and recent days') work is consisting of figuring out the best way to define what models can be used in any given modular part -- how they are defined in the part config.  One of the goals of this is to make it easier to write the configs, and as such, the amount of information needed to set up the parts is being reduced quite a bit -- mostly in regards to variant model setup and support (e.g. the different variants in the MFT-A tank).

To this effect, each model definition now tracks what 'profiles' are compatible with its top and bottom attach nodes.  These are purely a config-level delineation, and not defined through code anywhere -- so you can add new profile types simply by using them in the model definition configs (similar to docking node compatibility).  What this is all for is to allow for only the valid adapter options to be available for selection depending on what current models are selected in other slots in the part.  For example the 'CORE' model will determine what the valid options are for upper/lower slot depending on its current selection, and the upper/lower slots will determine valid options for the nose/mount slots.  Any change to the core (or upper/lower) models might result in a change to subsequent models further down the hierarchy.  Instead of defining per-variant adapter selections in the part-config, you simply define all valid adapter options, and which ones are currently valid for any particular setup are determined by the profile types.  Each model can have multiple compatible profiles, and top/bottom profiles are tracked independently so a model might have a 'round' top profile, but a 'square' bottom profile.  'none' is also valid, and specifies that the model has no compatible profiles (e.g. becomes the end of the chain; adapter options would be disabled).

And what this is all building up to -- a major change in how fuel tanks are 'built' in the editor.  Rather than having a specific pre-built model for each 'tank length' for each variant (e.g. A-0-5, B-0-5), the standard MFT tank will allow for selection of the intertank variant as the 'core' model slot (lower pipes only, double pipes, no-pipes cryo, framed cryo), while the upper/lower slots will be populated with a selection of models of various tank-lengths.  So you will still have a standard tank with  a lower, intertank, and upper portion, but with explicit ability to configure each of their lengths.  The 'nose' and 'mount' slots in the model will still be populated with the standard selection of adapter models.

One other interesting capability that is being built in, and I've discussed it briefly in the past, is 'model-scale-chaining' between the model slots in a modular-part.  Basically it works like this:

• Each model definition tracks 'diameter', 'upper diameter', and 'lower diameter'
• User 'diameter' UI control directly influences the size of the 'CORE' model slot.
• UPPER/LOWER model slots query the 'upper' and 'lower' diameter of the CORE slot to determine how they should scale their model to ensure it lines up with the upper/lower attach-point sizes of the CORE.
• NOSE/MOUNT model slots query the upper/lower diameter of the UPPER/LOWER slots to determine their scaling.
• Basically allows use of multiple adapters in a row while all maintaining proper ratio.  Or perhaps conical tanks.  Or something....  (N1 tanks with adapters?)

Not sure that I'll be using the feature in any of the existing/reworked parts, but I've got some ideas on where it can/will be used in the future.

 

Getting about to the point where I can start working on updating a few model definitions and loading them into the part-config, and maybe start testing the code a bit later this week.  I'm sure its going to be all sorts of broken for at least a few more weeks... but as I haven't heard much regarding the ETA on KSP1.4, I'm in no huge rush.  Would rather take the time to do it all right and get all the desired features implemented properly rather than rushed and hacked/tacked on afterwards.

[Jimbodiah]

1.4 releases tomorrow

 

 

 

 

 

jk jk

[TachyonGMZ]

The SSTU SMURFF patch did not work what else can I try. I decided against realfuels for the isru

[Jimbodiah]

Try my RSS patch

[T-10a]

It doesn't work because I goofed it up.

Try this instead:

@PART[SSTU-SC-TANK-MFT-*|SSTU-SC-TANK-MUS]:NEEDS[SMURFF]
{
	@mass /= 4
}

Either that, or use Jimbodiah's patch (It instead buffs engine ISP to give RSS delta V with stock part thrust & mass.)

Edited January 3, 2018 by T-10a

[Nightside]

@TachyonGMZ, did @T-10a's patch work for you?

I suspect it won't work for SSTU's modular fuel tanks, since the tankage mass is set in the editor. This is what I use to give SSTU tanks a realistic tankage mass of about 5% re and volume for RSS in my Transmogrifier patches. It ignores SRBs. 

//Module Manager patch to change tankage mass and volume in line with SMURFF's values
@PART[*]:HAS[@MODULE[SSTUVolumeContainer]|!MODULE[SSTUModularBooster]]:NEEDS[SSTU]
{
	@MODULE[SSTUVolumeContainer]
	{
		//name = SSTUVolumeContainer
		//default placeholder volume; set by MFT module below
		//volume = 100000
		//enableContainerEdit = true
		//enableFuelTypeChange = true
		//baseContainerIndex = 0
		@CONTAINER,*
		{
			//name = Main Tank
			//percent = 100
			@tankageVolume = 0.17 //0.15
			@tankageMass = 0.05 //0.15

		}
	}	
}	

 

[T-10a]

2 hours ago, Nightside said:

@TachyonGMZ, did @T-10a's patch work for you?

I suspect it won't work for SSTU's modular fuel tanks, since the tankage mass is set in the editor. This is what I use to give SSTU tanks a realistic tankage mass of about 5% re and volume for RSS in my Transmogrifier patches. It ignores SRBs. 

//Module Manager patch to change tankage mass and volume in line with SMURFF's values
@PART[*]:HAS[@MODULE[SSTUVolumeContainer]|!MODULE[SSTUModularBooster]]:NEEDS[SSTU]
{
	@MODULE[SSTUVolumeContainer]
	{
		//name = SSTUVolumeContainer
		//default placeholder volume; set by MFT module below
		//volume = 100000
		//enableContainerEdit = true
		//enableFuelTypeChange = true
		//baseContainerIndex = 0
		@CONTAINER,*
		{
			//name = Main Tank
			//percent = 100
			@tankageVolume = 0.17 //0.15
			@tankageMass = 0.05 //0.15

		}
	}	
}	

 

TIL the mass is set via SSTUVolumeContainer. That'll be why it doesn't work.

[Nightside]

30 minutes ago, T-10a said:

TIL the mass is set via SSTUVolumeContainer. That'll be why it doesn't work.

TIL what TIL stands for, so thanks,  I guess we're even.

[RedParadize]

Quick question for anyone that can answer it (almost certain that someone had done this) 

I added Mount-Skeletal-S as a Nose on the Mus. I was wondering how to RCS to the top as well. Is it even possible?

[tater]

23 minutes ago, RedParadize said:

Quick question for anyone that can answer it (almost certain that someone had done this) 

I added Mount-Skeletal-S as a Nose on the Mus. I was wondering how to RCS to the top as well. Is it even possible?

You can either add separate RCS parts, or you could put another MUS on top, flipped, I guess.

[Shadowmage]

10 hours ago, RedParadize said:

I was wondering how to RCS to the top as well. Is it even possible?

Built-in?  Not possible with current code.

(the upcoming rewrite will have multi-rcs-model-slot functionality though; lower + upper, with UI selections for each)

[TachyonGMZ]

@Jimbodiah T-10a's patch did not work. Does your RSS patch work on all engines or only certain engines?

[Jimbodiah]

It is a global patch for ALL engines; stock and from mods. It looks for RealSolarSystem and any part with the ModuleEngine* and then changes the ISP and Thrust so that a realistic looking rocket will do what the real life version does.

It's my personal patch for playing RSS without needing complete rebalancing like RO or SMURFF.  I tend to test it on a realistic looking Delta IV Heavy built with SSTU which should just get an Orion CSM + HUS into LEO. As the patch is so simple, you can fine tune the isp/thrust to your own liking.

 

See the link in my signature for the download, or just make an RSS.cfg file that contains the following:

 

@PART[*]:HAS[@MODULE[ModuleEngines*]]:NEEDS[RealSolarSystem&!SSRSS]:FINAL
{
    @MODULE[ModuleEngines*]
    {
        @maxThrust *= 1.7

        @atmosphereCurve
        {
            @key[1, ] *= 1.7
            @key,1[1, ] *= 1.7
            @key,2[1, ] *= 1.7
        }
    }
}

 

Edited January 4, 2018 by Jimbodiah

[Shadowmage]

Today's bit of development fun:

Just finished writing up an external tool to help with writing ModelDefinitions and ModularPart configs.  Or rather, to output all of the valid model combinations for the model definitions specified inside of parts that use the SSTUModularPart module.  It loads the config nodes from the GameData folder, so works with the actual live part configs (does not include ModuleManager patches though; so the values must be actually defined in the config and not patched in afterwards).  It does not check solar or RCS options, only the models specified for the NOSE / UPPER / CORE / LOWER / MOUNT slots.  It loops through every possible combination and writes to an output file each valid combination that is found; a valid combination being anything that has a model for each slot, where all of the models' profiles are compatible with each-other.

The 'profiles' system I described earlier in the week works, and works very well so far, but trying to check every possible combination by hand/in your head would be.... difficult.  Hence this new tool.  No need to boot up KSP just to check if the adapters/mounts/etc are usable like they should be -- simply run the tool and check the output file that lists the combinations.  Will likely adapt it to also include an exception report -- a list of what models are not valid combinations, and a short explanation on -why- they are not seen as valid -- will be invaluable when doing more complex setups.

And, as a bonus, I'll finally be able to get a count on the total possible model/mesh permutations for any specific part   (it was a huge number before; going to be even bigger with 5 main model slots in the parts; likely 100k+ on the MFT-A alone)

 

@TachyonGMZ  What exactly are you trying to accomplish?  (what is the goal, what other mods are in play)

It should be relatively simple to write a patch that adjusts the SSTU fuel tanks to have more 'realistic' mass fractions, but there are a few other considerations when other mods are also used.

(I'm not familiar with SMURFF or what it does... so trying to use that as an example means absolutely nothing to me)

On 1/1/2018 at 4:44 PM, TachyonGMZ said:

Exact same engine exact same fuel tank size and fuel type. Pod was different as mentioned.

You can't compare rocket (stage) DV unless you use exactly the same parts in the upper stages.  A very small change in the upper-stages' mass makes a massive change in the lower stages' dV numbers.

Try the comparison with all other things being equal except for the fuel tank.

[Nightside]

23 minutes ago, TachyonGMZ said:

@Jimbodiah T-10a's patch did not work. Does your RSS patch work on all engines or only certain engines?

@TachyonGMZ, read back a few posts. T-10’s patch Only works for stock style parts. Try the patch I posted in response. That is the method for changing tank/fuel  mass ratio for SSTU tanks.