krakken232

I plunked this little guy on Duna as the first step of my Duna base that never came to fruition... probably still up there, somewhere...

Nothing too special -- just a stock, non-mechjeb little lander I made. I think it's cool just because of the inconspicuous placement of the ASAS and RCS tank which is sometimes hard to pull off with the oversized crew compartments these days. Everybody's getting ready for exploration!

Alright -- so I moved the attach node up and things were way more stable. However, I then realized that the engine wobbles on the bottom of the tank. No matter how many decimal places out I took the \'y\' of the bottom attach node, I could not hit the sweet spot between not connecting and the engine being wobbly. I did what I should have done in the first place -- gone back to the drawing board with some rudimentary objects for experimentation. I\'m using Blender, again, and I am wondering if there are any setup options on the interface or the DAE exporter I need to tweak? Here\'s the deal -- if I make a cylinder that is 4 meters tall exactly in Blender, and set my node positions to 20 and -20 (cause of the .1 scale), perfect match. If I make a cylinder that is 4.5 meters tall, and set my node positions to 22.5 and -22.5, the engine is wobbly. Even if I move the node down to like 22.4 or 22.3, it still wobbles. So I can achieve a perfectly solid connection if I keep everything in nice, even, meter increments apparently, but if I mix in decimals I can\'t? I have no way to test at the moment but I am also curious if I move the node like two or three tenths inside the collision mesh and set allow_collisions to 1 - will it result in a solid non-wobbly connection?

Ok-- yeah, so the part itself is long and skinny (just like three stacked fuel tanks would be). So that makes sense, the node attach Y dimension is zero, the middle of the part, so if I were to set the node attach Y to be up near the top of the part, and use struts to secure the bottoms (again, just like you would three stacked fuel tanks) theoretically it should be a lot more stable. I will have to try it out when I get home!

So I thought I would fire up blender and try to make some parts for personal use-- mainly trying to create some parts that mimic my common assemblies (i.e. three stacked fuel tanks, triple stacked triple fuel tanks) so I can get some bigger ships airborne with less stress on my processor. Well I made some awesome looking parts and put them in my game. But here\'s where I run into trouble. Take my first part: it\'s one part that is the same exact size and shape as three liquid fuel tanks stacked (I know because I imported the DAE\'s of the fuel tanks into blender to base my model off of). It looks great, and works great in the VAB (all the connection points line up, etc). I modified the CFG file to change the mass, fuel capacity, and dry mass to exactly 3 times a regular fuel tank. All in all, it should be pretty much exactly the same as having three fuel tanks. I can snap them on and launch just fine. Now, when I try to create even a slightly complicated ship, it just falls apart. For instance, a lander on top of my part, with four more of my parts attached symmetrically will barely stand up on the pad, whereas if I stack three regular fuel tanks on top of each other, and duplicate that four times, it\'s perfectly stable. Likewise, if I try one stage bigger, it wobbles around briefly and falls apart and explodes, whereas if I build the same ship out of multiple regular fuel tanks it stands tall and proud (and laggy). So how come a single part with a single collision mesh with the same overall stats as three regular fuel tanks is so much more wobbly and unstable than three regular fuel tanks put together? Are there other stats I need to modify? Is it a problem with the collision mesh vs the attach node values? Help?

So how would you account for this or offset it using the game\'s ship controls? Would it be possible to maintain a nice, low, circular orbit of the Mun?

Having slightly more than a basic knowledge of gravitational principles and n-body physics (n-body simulators being a hallmark of Computer Science programs at the university I attended) I knew there are huge differences between single-body physics universes and n-body ones. I know N-body physics might never be implemented in KSP, which is fine, but I thought it would be interesting to get some perspectives on what the implications of N-body physics on the Kerbal universe might be. What possibilities would this open up for Kerbonauts? What, for instance, would be different about lining up a Mun shot, transferring orbits, or even orbiting Kerbin itself?