Thoughts on opening cargo doors?

[Cepheus]

I am aware that Unity doesn\'t allow convex collision meshes, but is it still possible to have 'opening' cargo doors?

[drewmacrae]

Yes, they can certainly be made with multiple meshes, perhaps they require multiple parts though. The trouble is, if used for capture, you\'ll have to do something to attach the part to your ship or remove it before you\'ll be able to warp.

[Cepheus]

At this point, I\'m not thinking of capture (this\'ll be easier when we get docking). I\'m just thinking about the possibility of transporting cargo to, and from the surface.

[Tosh]

There\'s already an example of a part having multiple moving colliders.

http://kerbalspaceprogram.com/forum/index.php?topic=8104.msg129886#msg129886

[C7Studios]

I plan on making a universal 'door' module. I just need some free time heh. I also need to finish of the gear. So much to do. If not, I\'m sure someone will beat me to the punch. Sarkun is making some really good progress on his module.

[Quinton]

There\'s already an example of a part having multiple moving colliders.

http://kerbalspaceprogram.com/forum/index.php?topic=8104.msg129886#msg129886

Hopefully I\'ll have some time to update this further this weekend and address a number of useful pieces of feedback I\'ve gotten from folks here.

One thing my collaborator (who did the modeling) was interested in was building openable doors that the panels would be folded up behind during launch. We\'re not sure if we\'re going to try to do this as multiple pieces (might be more in the spirit of reusable KSP components) or a fancier all-in-one part.

Things I learned about working with multiple colliders:

1. The game only converts one mesh (named 'node_collider') into an actual collider. The game renames this mesh to 'collider'. This happens sometime before your onLoad() method is invoked.

2. It seems best to not name other meshes 'node_collider...' as we ran into some odd issues with some nodes going missing (though it may have been pipeline issues with getting our DAE out of Lightwave). We ended up naming our sub-collider meshes 'boundbox...' so we could recognize them by name.

3. if you\'re going to animate things, it\'s convenient to have the colliders be children of the meshes you\'re animating, so they

4. the fixupColliders(Transform t) utility function here (http://code.google.com/p/f13aerospace/source/browse/trunk/F13Utils.cs) walks a Transform tree and any node with a name starting with 'boundbox' gets converted from a renderable mesh into a collider mesh.

5. In onPartUpdate() you can apply local transforms to adjust the position of the children based on input (this is hacky, hardcoded, and always active -- going to clean this up soon) - http://code.google.com/p/f13aerospace/source/browse/trunk/F13SolarPanel.cs

Anyway, hope this is helpful.

I really need to figure out how to set up the drag model right and adjust it based on the panels being deployed or not -- my goal is to ensure if the panels are deployed in atmosphere they get torn off in a horrible and hilarious manner.

[sarkun]

I really need to figure out how to set up the drag model right and adjust it based on the panels being deployed or not -- my goal is to ensure if the panels are deployed in atmosphere they get torn off in a horrible and hilarious manner.

Look at the Parachutes class - it adjusts the maximumDrag value depending on the state of the Canopy - the interesting thing is that it seems to adjust the value gradually - possibly to avoid ripping the chute apart - if you have the part tracking it\'s opennes state anyway - you can easily use that to keep updating the maximumDrag value and the gradation will happen automagically.

I haven\'t tested it yet, but this is the approach I\'ll be taking in my MultiJoint Part plugin. EDIT2: Tested and it works.

EDIT:

Oh, and remember that the current Drag Model is bonkers anyway, and the actual drag is Mass * MaximumDrag.

[Feragorn]

I always thought large clamps would be more fun than working cargo bays.

[BioSehnsucht]

I always thought large clamps would be more fun than working cargo bays.

Mike! You killed the Kubble!

[Feragorn]

No, man, they just built it wrong.

[sarkun]

Well I thought about it for a day, and decided that since I\'m still waiting on my client... I might as well do something for my Kerbals right?

Sorry it\'s so dark - crank up the brightness of your screen!

Hope Shuttle CargoBay Test

[Emilio]

Well I thought about it for a day, and decided that since I\'m still waiting on my client... I might as well do something for my Kerbals right?

Sorry it\'s so dark - crank up the brightness of your screen!

Hope Shuttle CargoBay Test

:o :o

How do you made the automatic panel deploy ?

I suppose the cargo bay is a lander leg ?

[je_day]

Well I thought about it for a day, and decided that since I\'m still waiting on my client... I might as well do something for my Kerbals right?

Sorry it\'s so dark - crank up the brightness of your screen!

{AWESOME VIDEO}

I\'d like to post Mother of God picture but i am too lazy to find it

[sarkun]

:o :o

How do you made the automatic panel deploy ?

I suppose the cargo bay is a lander leg ?

Lander leg? Pffft! I take you are not familiar with my MultiJoint Framework?

[Emilio]

Oh, yes, of course... I have it.

There is only a little thing : since they produce more drag when deployed, you can use them as parachutes...

[AaronLee]

Oh, yes, of course... I have it.

There is only a little thing : since they produce more drag when deployed, you can use them as parachutes...

Only if their crash tolerance is fairly high, I\'d think. You\'d need to balance the door joints to resist impacts from normal landings but be torn off at high speed in the deployed state. So I guess it\'s a matter of balancing drag and impact tolerance.

Has anyone thought about doing a bulkhead with attachment points for the wobbly rockets revamp farings? You can already build a decent detachable, two-sided nose faring of arbitrary length. If the bulkhead part just rotated its attachment nodes then you;d have an instant door.

[sarkun]

Only if their crash tolerance is fairly high, I\'d think. You\'d need to balance the door joints to resist impacts from normal landings but be torn off at high speed in the deployed state. So I guess it\'s a matter of balancing drag and impact tolerance.

Has anyone thought about doing a bulkhead with attachment points for the wobbly rockets revamp farings? You can already build a decent detachable, two-sided nose faring of arbitrary length. If the bulkhead part just rotated its attachment nodes then you;d have an instant door.

Working cargo bay:

http://youtu.be/ADn_LfEZQRg

Jump to 0:28 if you have ADD ;-)

You can download it with the rest of my pack.

[Tiberion]

Yeah Cargobays are definitely 'here' now, and pretty much fully functional - I replicated sarkun\'s bay using an old c7 fuselage piece pretty easily and it works great, even capturing satellites works (with the same \'no time warp\' limitations we have for other such opertions)

Pretty much the biggest downside is the attachment logic and trying to build/insert cargo, but that is a drawback of using the already tempermental collider objects in odd ways - future official features should help this work better.

[Emilio]

Yeah Cargobays are definitely 'here' now, and pretty much fully functional - I replicated sarkun\'s bay using an old c7 fuselage piece pretty easily and it works great, even capturing satellites works (with the same \'no time warp\' limitations we have for other such opertions)

[snip]

Pretty much the biggest downside is the attachment logic and trying to build/insert cargo, but that is a drawback of using the already tempermental collider objects in odd ways - future official features should help this work better.

Will it be in the SBS pack ? Say yes ! Say yes !

[Tiberion]

Released version 2.0 last night, its included