viila

My super cheap minimalistic satellite to fulfill a Minmus satellite contract. 12k funds, rocket barely lifted off the pad, but through 1 part good planning and 2 parts luck, the satellite made it with 2 out of 60 units of monoprop left! Good for maybe 50 m/s... (It used that one spherical mono tank to do Kerbin circularization (the booster left it little bit short to not create debris), Minmus transfer burn, course corrections and Minmus capture into the final orbit)

RCS is pretty bad in KSP... I often have horrible oscillation when using it with SAS or MechJeb. Plus, it's really hard to design a good RCS system onto a spacecraft because you can't designate nozzles to do only one thing or adjust their relative powers. Easiest way to get RCS working well with translation is to put linear ports inline with the CoM, which makes them useless for rotations (except for wasting fuel), but if you add more ports with longer moment arms for rotation, that screws up your linear setup because you can't usually get them exactly balanced around the CoM. So in almost all cases you have unwanted translations when rotating and unwanted rotations when translating when you try to have a RCS system that can do both. And in all cases you have nozzles that are wasting RCS fuel since they are all firing if they think they can apply even a little momentum in even vaguely the direction you're turning/translating towards. Plus, if you have really tiny probes, even the weakest RCS nozzles in precision mode will still spin it at a billion RPM at the slightest touch. (This makes the reaction wheel-less probe cores almost unusable, because anything you'd want to use them in to save weight, you can't, because RCS would be too twitchy to actually be able to control them, so you need to add reaction wheels, and then you may as well use the next heavier core with proportionally sized wheels built-in!)

The problem with introducing an intermediate engine is the dreaded R-word. Because realistically you can't have a liquid bipropellant engine with higher Isp than 450 seconds. That is the absolute theoretical maximum for LH+LOX chemistry. Our liquid fuel has similar density to RP-1 (refined kerosine) which puts the absolute cap at 360 seconds. A few exotic tripropellants could get up to around 500 seconds, but those are insanely nasty stuff like elemental fluorine (with exhausts containing the ever wonderful hydrogen fluoride), or combinations that no sane person would attempt (cryogenic oxygen/fluorine + cryogenic hydrogen + molten beryllium or lithium!). I real life there just is a huge jump from chemical propulsion to nuclear propulsion. And Nerva style NTR is relatively tame nuclear engine to begin with. (Because it's limited by hopefully keeping the nuclear core solid. There are even more crazy concepts like liquid or gas core NTRs that would leak radioactivity like crazy, but would have an order of magnitude higher Isp) ------ As to the engine improvements, I like them as well, except that we no longer have a small 3m lifter engine, nor an inline lifter engine. The Mammoth limits options since it's a cluster, so it has to be a terminal part. (Nor can we substitute the largest 2m part either, since that's a cluster as well.) We can't have anyting bigger than Mainsail inline if we need atmospheric performance which is sometimes necessary on superheavy launches since the first stage might not get you up more than few kilometers.

What's up with all the second floor floodlights inside the fully upgraded VAB constantly pulsing on and off? The exact same floodlights in the upper floors do not pulsate! It's highly distracting since they're constantly in your peripheral vision which is really sensitive to motion. Windows, 32 bit, both 1.0.0 and 1.0.2

You'd have to add control surfaces then to control it. But landing works just the same with the smallest possible horizontal one (to minimize any lift gained; vertical one is split to get roll control without needing two horizontal ones): https://dl.dropboxusercontent.com/u/17789257/lolglider/lolglider1.jpg https://dl.dropboxusercontent.com/u/17789257/lolglider/lolglider2.jpg https://dl.dropboxusercontent.com/u/17789257/lolglider/lolglider3.jpg https://dl.dropboxusercontent.com/u/17789257/lolglider/lolglider4.jpg https://dl.dropboxusercontent.com/u/17789257/lolglider/lolglider5.jpg Touchdown! https://dl.dropboxusercontent.com/u/17789257/lolglider/lolglider6.jpg Full battery - no torque used.

No, they were, uh, completely confused by units. For example, the Rolls-Royce Pegasus engine (used in Harrier Jump Jet) that someone linked earlier masses 1.8 tonnes, which is exactly the same mass as the J-X4 Turbo Ramjet we have now. The mass is just distributed oddly, because ours is concentrated entirely on the outlet. The P&W J58 which powers the SR-71 which is more in-line to what our advanced jet can do masses almost twice as much at 2.7 tonnes.

With all these threads about the new aero and when you should do your gravity turn, I'm curious if anyone has tried a realistic single continuous burn (apart from staging, if applicable) to orbit launch? Can it be done in KSP? In real life rockets have extremely limited restart capabilities; main stages usually can't be restarted at all and even most upper stages have very limited number of starts they can do.

My Munbase Alpha in the night. I had just landed those floodlight carts to help with night operations during the long Mun nights. (You can see the wreckage of the lander further back between the rocks. It was meant to drop the carts from low altitude and then boost away and do a soft landing, but a snafu with forgetting to set the cart brakes meant that I had to choose between saving the carts before they accelerated too fast to stay upright or saving the lander...)