DeadMG

Actually, the problem was the complete inverse of what I (and apparently everybody else) expected. There were too many control surfaces, not too few- since the rocket is so tall, a little goes a long way. Once I finally thought to replace the Winglet/Tail Fin with something less effective (the delta-deluxe winglet), it flew fairly easily. Just for notes, the reason why I have that stuff on the interplanetary stage is because it's actually probe-controlled (there's a probe core hidden under the docking port) so that i can control it whilst the lander is detached.

There's a bug in the game which can cause rockets to veer even when built symmetrically- I've had a few craft suffer from this. Recommend rebuilding the craft from scratch- this usually alleviates the problem.

Here's my rocket. I've created this rocket in my career for exploration of Duna- specifically, I want to land on both Duna and Ike and return. I've got four main stages- liftoff, orbit, interplanetary, and lander. Liftoff is just a huge mass of SRBs. Orbit is a Mainsail with some fuel (hidden by all the SRBs). After that, the interplanetary stage is an atomic engine (the top fuel tank above the orange one also contains oxidiser for refueling the lander stage). Then underneath the fairing I have a 3-man lander powered by LV909 with a lot of parachutes, etc. I haven't really played KSP much since about 1.1 or so, and it seems like some of the aerodynamics have changed and my rockets are no longer stable. I tried asking MechJeb to fly them instead but even Mr. Jeb can't fly my rockets stably. I've tried a few variants on this like adding a bunch more wings in various places, but it doesn't seem to have made much difference. Any suggestions on how I can improve the aerodynamics of this rocket?

Here I'm thinking about the general development contracts, e.g. Orbit Kerbin, Orbit Mun, etc. I often found that the contracts don't keep up with what I want to do. I orbited Kerbin but I didn't get Orbit Mun until after I flew by the Mun. Now I want to land on Minmus but the contract doesn't appear- the guy wants to send me back to Mun instead. I think that the game should much more aggressively provide contracts for new bodies, so the player has a choice as to where to go and if they can get a contract for it. It's always a bit excrements to do something new but not have a contract for it so you don't get a big reward. Maybe instead all of the "Orbit Mun" style contracts should just be eliminated and folded completely into the World Firsts thing, so you never lose out by going further than you have a contract for.

Clearly, we should just add gravity generators to the game.

These things are ridiculous. It seems to me like the game always has one of these up. All you have to do is put a satellite in orbit around each body and you can instafulfill the contract by transmitting 0 science data of gravity scan. Then it just respawns over and over again. I'm on the fence about this. On the one hand, it was a nice motivation to actually use satellites, probes and such, which I previously had absolutely no use for whatsoever. It's actually been quite motivating to launch quite a few satellites. On the other hand, it totally blows away every other contract. Especially those SUPER annoying "Test X part in Y place" contracts. As long as you have a satellite in orbit around all of the bodies you've "explored" so far, you have basically infinite resources. I think that contracts really need re-working. They should directly reward different ways of playing the game- like "Land probe on X body" vs "Land a manned craft and return it safely", "Put satellite into orbit", "Put super-heavy object on the ground safely", "Land on a particular point on a body", etc. The current contracts divide into three groups- "Stuff I'd do anyway", like explore another celestial body, "Stuff I'm never gonna spend my time doing no matter how many imaginary points you give me", like "Test explosive decoupler on escape trajectory from Mun" or "Plant the fourth flag on Mun", and "Stuff that gives every reward possible whilst being fun to play", which is exclusively "..... transmit science from orbit". Those still need a little nerf. I'm going to say that contracts should really be mostly in the 3rd category.

Sure, but I don't need extra torque or a 2.5m stack. I only need the mk1 can but can hold two Kerbals.

The MK2 lander can needs a lot of love. It's a convenient part for holding enough crew to operate the experiment cleaner, but it weighs so goddamn much, it's way cheaper to just use two mk1 lander cans. I think the mk2 can should have a weight of maybe 1-1.2 tons at most.

Ultimately, what it really comes down to is this. Yes, I'm lazy. Yes, I'm not thinking about this. That's what I bought a computer for. To think about mundane things for me so I could think about interesting things. If I didn't want to do that, I'd play KSP by simulating every facet of the mission by hand on paper. But I don't and it's because it's way cheaper and easier to just buy a machine to do it for me. So yes. The computer should do what I want, when I want it to, and it should act in a way that maximally benefits me, always. That's what I paid for. When I play KSP, I want to be thinking about how I can increase my rocket efficiency, or what is the most efficient maneuver to make, or other gameplay details, and nothing else. The core problem with the current save system is that it only makes sense when applied to a totally different kind of game than KSP. A game where it's a roughly linear progression, as long as you didn't lose then you're fine to continue, and it's clear when you've "lost" and to resave. KSP is not that kind of game. It is the kind of game where the player creates complex decision trees with no clear failure condition and he needs to reload to arbitrary points from arbitrary points.

Or, option 3, the game could just NOT OVERWRITE THE SAVE, so you don't have to confirm anything because there's nothing to confirm, and people keep all their saves! Gosh, it's almost like it's the best of both worlds and clearly the best option, or something.

Good. Just like they should do. The state of a KSP game is tiny and saving hundreds of them would barely be a tiny tiny scratch on hard drive space. Really? I don't remember being asked if I wanted to permanently lose my current game state when quickloading, or permanently overwrite my old quicksave when quicksaving. They all fall into one of two buckets- either it's SUPER UNBELIEVABLY ANNOYING (the bucket KSP is in), or, the game has clear failure conditions on which it automatically reloads the previous autosave, for example, death in Dishonoured. Even if you die in KSP or run out of fuel, it's not a clear failure condition.

I agree that the way KSP handles saves is abysmal. The game should NEVER overwrite a save if you don't explicitly tell it to, and you should be able to manually load from autosave- any of them.

You can just EVA and collect the data before ditching the transfer stage.Also, if you read the OP, I explicitly stated that this is the setup I used. And it still yields more science than going to Duna/Ike. Each biome averages out at north of 200, even though most of them you don't use a goo/lab, and I didn't transmit crew reports (100% efficiency). You can get 2500 for the whole trip, compared to just 2000 for landing on both Duna and Ike, which is a far more complex and tech-requiring mission. You don't need more than 3 goos/mats for a Minmus mission to be ridiculously rewarding.

Every biome on Minmus, including upper and lower orbit, one trip, low tech. The first time I used fuel ducts so you have a little more oomph to spare if things go wrong and a couple niceties like ladders that I cut to account for lack of ducts, doing them all here is fairly tight, but as you can see, it's quite low tech. The rocket isn't a complex or large design, it's extremely simple. It's the 101 of how to build a bigger rocket with radial decouplers. As for going to Duna and Ike, I doubt I could do it without fuel ducts. You can't get BACK from Duna and Ike on this tech level. Getting off Duna is way too expensive, by the time you've done it you've got nothing left. The rockets don't scale high enough without fuel ducts and atomic engines. But that's not the point, because going to Duna/Ike takes forever compared to Minmus, yields less science, and would require considerably more skill.

The ion engines have a nice efficiency, even if low TWR. I suggest that they should have a radial version, which would allow you to add them as "support" almost to other stages. This way they can be useful with the long burn times, because you'll be using them whilst burning most of your other stages.