Admiral Regulus

If something goes wrong during a mission, I send out another to bring back the crew, and whatever science they have with them. I rarely bring vessels back to Kerbin unless it's really early game, or it's part of a mission objective. In mid-game or later, I do not design crafts to survive atmospheric entry on Kerbin. Instead, I use a cheap, reusable, dedicated space shuttle to retrieve crew from LKO.

When I first started playing KSP, I didn't care much for asthhetics. I had trouble building anything that could land on the Mun and come back, much less a craft that looked nicr while doing it. Back then, many of the spaceplane parts didn't look so nice, and there was a huge lack of niche parts that could be used to add coolness or tidy things up. Now, many years later, I've discovered ways to complete just about any task in KSP (barring Eve ascent). I find now that the challenge in playing KSP comes from designing vessels that both look nice and have all thr functionality necessary to complete the mission. Sure, I can make something thaty can land on Duna and come back back home. That's easy. The challenge is building something to do that, while having it look nice as well.

I build the payload first, whatever it is. This could be a vessel designed to land/return from Duna, a refinery/refueling ship, a science probe, or any number of things, really. Depending on the mass of the payload, I adjust the lower booster stage to LKO, so that I only use rockets as big as necessary. If I'm going somewhere like Eeloo or Moho with high delta-V requirements, I add a middle interplanetary stage with LV-N's, too. Since 1.0, I no longer return any vessels directly to Kerbin. When I want my Kebals to come back home, I return their vessel to LKO and launch a (mostly reusable) space shuttle to transfer everything back to the surface. I don't think this is the most efficient way to do things, honestly, but I think it's more fun. It was one heck of a challenge to get a space shuttle to work, but that's one of my favorite things to use now.

No. A lot of the cuts to realism we see in KSP are implemented for a very good reason. Smaller solar systems don't require so much memory. That's why everything is scaled down so much in the stock game. Larger, more realistic solar systems would also require a thirty minute ascent to orbit, which seems awfully boring to me. I imagine that would be especially so considering a more realistic model would add (and necessitate) autopilot features, too. In addition to introducing even greater performance issues, the larger values for numbers would also accumulate (relatively) greater round-off errors during calculations unless some dramatic changes were made. Similarly, truly accurate aerodynamic and thermodynamic simulations require too much computational power. The same could be said for simulating orbits using an n-body simulation. What was added in the 1.0 patch already kills performance, and that's pretty far short of what's used in the real world for engineering applications. And if it's only the engines and other parts that are made more realistic, then all parts would be incredibly overpowered compared to the smaller solar system currently implemented. So the scale of everything else would need to be increased as well, and for reasons already explained I'm not so fond of that idea. This isn't even to mention that if parts were more realistic, our choices of propulsion would be severely limited. Most applications would work best with ion engines with their millinewtons of thrust or absurdly overpowered NERVAs five times the thrust and an additional 50 seconds of isp. That would seriously outperform everything, making all the tough design decisions in the stock game trivial. What (sarcastically expressed) fun.

It's because the LV-N is too massive to see any benefit from the better efficiency. The Terrier has an Isp of 345 and mass of 0.5, whereas the Nerv is 800 and 3.0. So, the respective delta V's are: 345 * 9.81 * ln([m,full]/[m,empty]) for the 909 and 800 * 9.81 * ln([m,full]/[m,empty]) for the LV-N. This is a pretty complex equation if you're trying to solve for the masses, but if you look at it in terms of the mass ratios, it becomes a little simpler. The LV-N is only more efficient if the following holds true: [m,full,909]/[m,empty,909] < ([m,full,LV-N]/[m,empty,LV-N])^2.3188 So, if, with the 909, you have a mass of 2 t initially and 1 t after the burn, you'll have more delta V than you would if you add the nuclear engine. If, however, you have a mass of 3 t at start and 2 t at the end of your burn, you'll be better off with the LV-N. In general, the more fuel you're burning, the more beneficial it is to use the LV-N. You'll get more use out of your fuel, so with longer burns you're saving more. However, the LV-N weighs a lot. So, if you have a small amount of fuel on your vessel, a lighter engine will help you more than a higher efficiency engine. A heavy engine behind a heavy fuel tank doesn't change the mass significantly, but a heavy engine behind a light fuel tank makes a world of a difference.