[1.1.3] Procedural Parts - Parts the way you want 'em - v1.2.5 July 3

[OliverPA]

Same here - procedural tanks are empty to half empty upon loading in VAB even though their are full in the craft file.
Simply replacing the 1.2.3 ProceduralParts.dll with 1.2.2 was enough for this terribly annoying issue to disappear.

Looking at the source code there are a few changes where the part mass used to be saved in 1.2.2 and now in 1.2.3 just an event is triggered...

[Liquid5n0w]

34 minutes ago, OliverPA said:

Same here - procedural tanks are empty to half empty upon loading in VAB even though their are full in the craft file.
Simply replacing the 1.2.3 ProceduralParts.dll with 1.2.2 was enough for this terribly annoying issue to disappear.

Looking at the source code there are a few changes where the part mass used to be saved in 1.2.2 and now in 1.2.3 just an event is triggered...

I still got the issue with 1.2.2 and also 1.2.1

[Razorfang]

Same issue as @TMS, also believe this is causing NRE spam for me.

Spoiler

(Filename:  Line: -1)

NullReferenceException: Object reference not set to an instance of an object
  at Vessel.<GetUnloadedVesselMass>m__7E6 (.ProtoPartResourceSnapshot x) [0x00000] in <filename unknown>:0

  at UniLinq.Enumerable.Sum[ProtoPartResourceSnapshot] (IEnumerable`1 source, System.Func`2 selector) [0x00000] in <filename unknown>:0

  at Vessel.GetUnloadedVesselMass () [0x00000] in <filename unknown>:0

  at Vessel.GetTotalMass () [0x00000] in <filename unknown>:0

  at Vessel.CalculatePhysicsStats () [0x00000] in <filename unknown>:0

  at Vessel.FixedUpdate () [0x00000] in <filename unknown>:0

Never mind that. NRE spam is caused by something else. Trying to trace it now.

Edited May 15, 2016 by Razorfang

[Kendricks]

Does this mod have procedural hollow interstages?

If not, will they be added some day?

 

Hollow interstages, preferably procedural, aber the only thing I really miss when building stuff.

[CodeFantastic]

On 5/16/2016 at 6:12 PM, Kendricks said:

Does this mod have procedural hollow interstages?

If not, will they be added some day?

 

Hollow interstages, preferably procedural, aber the only thing I really miss when building stuff.

 This is a procedural fairings thing, go there instead.

[Gordon Dry]

I there kinda boiloff for LqdHydrogen?
By default RealFuels doesn't "provide" it for LqdHydrogen, but perhaps I chose the wrong tank type?

A LqdHydrogen-only tank is set to cryogenic, I got a huge probe with 30,000 m/s (it was for 4.5 days at 100% thrust), while I was away I let it run for 22 maneuvers to get to 900,000 km circular (Earth), each of them ~300 m/s, now at the beginning of maneuver 18 the LqdHydrogen is gone - so about 20% of the LqdHydrogen was used, the rest was wasted.

I don't find a clue, so I ask.

[Gordon Dry]

Does this "Child v smaller than first point" found in the log, mentioned in
https://github.com/Swamp-Ig/ProceduralParts/blob/master/Source/ProceduralAbstractSoRShape.cs
have something to do with the "torn apart and gaps between" issue?

Now after I updated B9PartWwitch to 1.3.0 it became worse, now every procedural part has a gap to the next one ...
But:

Check:
https://www.dropbox.com/s/zcwaqcxeq7m52yf/2016-05-27-1%20B9PartSwitch%20update%20broke%20procedural%20parts%20craft.7z?dl=0

[Phineas Freak]

@Gordon Drywhat kind of tank type did you use? Tanks other than the "Cryogenic" cannot hold cryogenic propellants for a very long time (and especially liquid hydrogen).

[Gordon Dry]

I just set all tanks that hold LqdOxygen to cryogenic.

btw what kind of tank type is cryogenic balloon? Just balloons inside instead of a huge single tank volume?

[Phineas Freak]

No, the "BalloonCryo" type means tanks like the SM - 65 Atlas and the Centaur upper stages, where their shape is preserved by continuous pressurization. They have the least dry mass of all other tanks but can hold the same amount of propellants.

[Gordon Dry]

Yes, now I clearly recognized it that the cryogenic tanks holding LqdHydrogen boil off, on time warp the decrease was clearly visible...

It was 100x and the value was 5.7 or so, that means 0.57/s in realtime...

 

Edit:

there is definetely something wrong, I added this patch
 

Spoiler

@TANK_DEFINITION[*]:FINAL
{
	@TANK[*],*
	{
		@loss_rate = 0
	}
}

GameData\zFinal\zNoBoiloff.cfg

and inside the ModuleManager.ConfigCache every instance of "loss_rate =" ends with a 0 - so why does this happen anyway?

Edited May 27, 2016 by Gordon Dry

[Gordon Dry]

About the boiloff:

 

[OliverPA]

Alright, so line 481 in TankContentSwitcher.cs is the problem - it references part.mass instead of mass. I changed that line, recompiled the dll and the issue is gone.
I thought about putting the change into Github but not sure how. Pull request? Just an issue report?

Anyways, here's the change:

TankContentSwitcher.cs, line 481

currently:
	return Math.Round((res.unitsConst + tankVolume * res.unitsPerKL + part.mass * res.unitsPerT) * shapeMultiplier, 2);

fixed:
	return Math.Round((res.unitsConst + tankVolume * res.unitsPerKL + mass * res.unitsPerT) * shapeMultiplier, 2);

 

On 5/15/2016 at 9:11 PM, OliverPA said:

Same here - procedural tanks are empty to half empty upon loading in VAB even though their are full in the craft file.
Simply replacing the 1.2.3 ProceduralParts.dll with 1.2.2 was enough for this terribly annoying issue to disappear.

Looking at the source code there are a few changes where the part mass used to be saved in 1.2.2 and now in 1.2.3 just an event is triggered...

 

Edited May 28, 2016 by OliverPA

[Phineas Freak]

@OliverPAopening a new issue is good but it is even better to create a PR for these since it can be easily reviewed and merged.

Edited May 28, 2016 by Phineas Freak

[OliverPA]

2 hours ago, Phineas Freak said:

@OliverPAopening a new issue is good but it is even better to create a PR for these since it can be easily reviewed and merged.

Thanks @Phineas Freak. I opened an issue... The pull requests are weird

[speedwaystar]

in PP 1.2.3, a 2m long tank masses half as much (dry and wet) as a 1m long tank, and contains the same amount of fuel. *scratches head* perhaps it is made of Unobtainium? ;-)

, 

Edited May 28, 2016 by speedwaystar

[NathanKell]

* Fix an issue with non-RF tank masses. Thanks @OliverPA77

https://github.com/Swamp-Ig/ProceduralParts/releases/download/v1.2.4/ProceduralParts-1.2.4.zip

[larkvi]

Can I make a procedural KIS container, or do procedural parts only contain resources?

[speedwaystar]

11 hours ago, NathanKell said:

* Fix an issue with non-RF tank masses. Thanks @OliverPA77

https://github.com/Swamp-Ig/ProceduralParts/releases/download/v1.2.4/ProceduralParts-1.2.4.zip

masses are looking good.

fuel sliders don't update until you dismiss the properties window and recreate it.

if you drag the sliders around before dismissing and recreating the properties window you get:

[OliverPA]

12 minutes ago, speedwaystar said:

fuel sliders don't update until you dismiss the properties window and recreate it.

Indeed they don't - it was like that before the last update as well. Fortunately it's just a visual issue.

[Gordon Dry]

NullReferenceException
  at (wrapper managed-to-native) UnityEngine.Component:get_transform ()

  at ProceduralParts.ProceduralPart.OnEditorPartEvent (ConstructionEventType type, .Part part) [0x00000] in <filename unknown>:0 

  at EventData`2[ConstructionEventType,Part].Fire (ConstructionEventType data0, .Part data1) [0x00000] in <filename unknown>:0 

  at EditorLogic.<SetupFSM>m__10F () [0x00000] in <filename unknown>:0 

  at KerbalFSM.RunEvent (.KFSMEvent evt) [0x00000] in <filename unknown>:0 

  at EditorLogic.OnNewRootSelect (.Part newRoot) [0x00000] in <filename unknown>:0 

  at PartSelector.LateUpdate () [0x00000] in <filename unknown>:0 

I wrecked my last vessel, saved the probe alone and made it subassembly by moving it to the folder and built a new rocket.

VAB -> build -> launchpad -> go

okay so far, but not enough dV overhead, so

-> revert to VAB -> change some tank lengths -> place a second thrust plate below another thrust plate, but smaller to have another circle of engines inline (!?) -> go to next stage to repeat that, detach it, it got no node and is un-re-attachable -> vessel broken -> detach probe, save it, make it subassembly

A new rocket is needed.

So now? What is the culprit?

Log:

https://www.dropbox.com/s/iytp76xqd0uuz6n/2016-05-29-3%20PP-PF%20the%20gap%20stuff%20again%20this%20time%20with%20NRE.7z?dl=0

btw the dead ghost nodes that float around are still there after I deleted the vessel to only have the probe left.

When clicking "New vessel" and then afterwards loading the probe, these ghost nodes are still hovering at the same spots.

[Guest]

Suggestion: add an "offset slider" to adjust the position of the point of a cone so that you can a shape like this

into this

[NathanKell]

Suggestion: Someone to do that.

[Van Disaster]

I looked at that a while back & thought "I hope someone else does that" along with segments so you can have half a cone, or a U shape.

Thanks for the fix, I was running out of teeth to grind...

[etheoma]

Oh sorry I see someone already suggest making an offset feature not but 2 posts ago my bad, Totally agree that would be a great thing to add.

personally I'm trying to make a Skylon alike with RO and RSS but the length of the craft requires an offset tail so it doesn't hit the ground but I can't find a parts pack with a long enough tail in the right shape. I'm currently using Procedural Parts but as said the tail hit's the ground more often than not and the speed required for take off is about 280 - 300 m/s because I can only pitch the craft to about 4 degrees on the runway. If I could only get a degree or 2 more I would be able to take off at 220m/s - 250m/s which is much more reasonable.

Also it's the same design as the proposed skylon so it would be more realistic as well as more functional.

Edited May 31, 2016 by etheoma