Eric S

The crash tolerance of the ore containers is 7 m/s, so the speed of the impact can't be ruled out. If the ore containers are the first part to hit, it's almost definitely the problem, in fact. I haven't tried this yet, but I just got a contract to do so, so I'll know more later.

The part does have a higher crash tolerance, though that really doesn't matter if it's being pushed by an LV-N, which is the case for most people that are complaining about the situation. - - - Updated - - - Understand that while I'm a programmer, I've never dealt with Unity 4 or 5, so this is just general impression. Some programs were able to switch over quite easily. KSP will not be one of them (or at least I belive so), because it rather heavily uses the physics simulation functionality in Unity, which is the one major area that is not backwards compatible between U4 and U5. Besiege, which is more KSP-like than most games based on Unity 4, has yet to release a version based on Unity 5, despite at least two weeks of effort, as last I heard. Squad has already mentioned that they have an in-house version of KSP running on Unity 5, though they also mention that there are issues.

Personally, I'm using a lot more SRBs in early career mode than ever before, and more in mid-range career mode. They don't feel overly nerfed to me. I'm using them for things similar to their real life uses. As a primary first (and occasional second) stage for light payload, to help low TWR craft get moving faster off the launch pad, or to get liquid fuel stages up high enough that their efficiency requires. If the craft might be able to use SRBs, I always try and compare prices, and they usually come up cheaper than all-liquid designs for those kinds of uses.

To go into a little more detail: 1) SRBs have a lower ISP than liquid lifter engines. This means that it takes more fuel mass to produce the same amount of thrust. Because of this, SRBs generally should be used as early as possible so that you can get rid of the less effective mass ASAP. In early career mode, I might use SRBs to get a rocket up to a high apoapsis, then circularize using liquid fuel engines. As I transition into having more variety of liquid fuel engines, I tend to use SRBs less often outside of the first stage, and then mostly to get the rocket out of the lower atmosphere so that the liquid engines have a higher ISP. Also, as rockets get heavier, I find SRBs to become less and less useful. 2) it used to be that you wanted to follow terminal velocity for efficiency, but that's not the case so much with the reduced atmospheric drag in 1.0 and later. I still find myself not exceeding terminal velocity below 25-30km altitude, but this is now because of rocket stability. Going transsonic alters the way drag affects the craft, and quite often crafts that are stable below terminal velocity may not be stable above it. 3) First, understand that the atmosphere drops off rapidly, enough so that other factors become more important than aerodynamic drag. The two major factors are gravity losses and steering losses. For gravity losses, you're looking at two factors. First, if you thrust straight up with a 2.0 TWR, 1 G of gravity cuts your acceleration to 1G of thrust upward. If you're at 30 degrees above the horizon, however, your thrust equals 1.0 upward and 1.73 toward the horizon. 1G of gravity negates the upward thrust, but now you're accelerating towards the horizon considerably faster than you were accelerating upward when thrusting straight up. Second, as your horizontal velocity increases, the more your forward momentum cancels out gravity (the phantom centrifugal force) meaning you spend even less of your thrust fighting gravity. For steering losses, if we ignore gravity for a moment, if you burn in one direction for 1000 m/s of acceleration then turn 90 degrees and burn another 1000 m/s of acceleration, you've only given yourself 1414 m/s of acceleration. If you had turned 45 degrees in the same direction then burned for 1414 m/s, you'd have the same velocity, though not necessarily the same position. So basically, a gravity turn is all about balancing all of these. Turn too aggressively, and you're fighting too much aerodynamic drag, don't turn aggressively enough, and your gravity and steering losses increase enough that they increase than the aerodynamic drag decreases. 4) I haven't tried doing spaceplanes in 1.0 or later, but from what I've heard, turbojets aren't as good for spaceplanes as they were previously. Many players have switched over to either strictly using RAPIER engines or a mix of RAPIER and turbojet engines. In either case, in 1.0, spaceplanes are more about reusability than low fuel cost. Prior to 1.0, I had a no-cargo spaceplane that would come so close to orbital velocity that it only took 50-80 m/s of delta-v to circularize. That's not happening anymore. 5) Different experiments respect biomes at different altitudes. The negative gravioli detector, for example, is the only science part that pays attention to biomes at the "high above" altitude range, all other science experiments will get you one shot at a full-value report. Most, if not all, science experiments respect biomes at ground level. EDIT: SRBs SHOULD be used ASAP, not shouldn't.

The only thing I use it for was very short hops to biomes near KSC.

What ascent profile are you trying to do? If it's the old "10km up then pitch 45 degrees" then that's not likely to work. What ascent profile works is really a per-craft thing depending on the aerodynamic efficiency and TWR of the craft. Craft with a higher TWR have to turn more aggressively and also have greater aerodynamic forces applied, so a high TWR craft that isn't aerodynamic is actually rather hard to get into orbit efficiently because the only time you can safely turn more than a few degrees away from prograde is before you pick up too much speed. For most of my craft that keep the TWR between 1.4 and 1.8, I find that tipping one degree when the craft reaches 35 m/s, a second degree at about 65 m/s, and a third at about 85 m/s. past 110m/s, I just follow the prograde marker. If I got the early part right, I tend to hit 10 degrees pitch from vertical at about 3500-4000m altitude, and 35 degrees pitch at about 9500-12000m. This is the ascent profile that tends to work best with my craft that like to flip. If your craft can safely turn farther off of prograde, you can wait longer and pitch over farther, but I still wouldn't expect to be able to pitch over more than 10 degrees off of prograde.

I'm still of that opinion. The KSC facility is built on a large flat area of the ground, except that it's really flat, not flat relative to the center of gravity of Kerbin, so anything far enough from the center winds up having a slight tilt towards the center. The reason I think you don't notice this without fins is because the initial tilt is so slight that it isn't noticeable without being magnified by a gravity turn starting at ground level and largely controller by aerodynamic forces. Without the aerodynamic forces, craft won't follow the prograde but rather maintain the same heading.

1.0.2 needs a few memory leaks fixed and it needs some tweaking so that direct-to-landing interplanetary transfer velocity aerobraking actually needs a heatshield. Those are the major issues to me. Yeah, they could tweak the aero parameters more so that spaceplanes are a bit more useful.

It comes down to that being a poor substitute, for the most part. In particular, if someone accidentally enters the SoI with a periapsis close to the center, rounding errors will tend to slingshot the craft out of the SoI at Kerbin's escape velocity or higher. You can already fake L4/L5 in patched conics, which are the two you can use without station keeping, and the points that need stationkeeping (stable on two axis, unstable on the third) don't behave close enough to a mini-SoI for that to be an approximation. Basically, the people that actually care about Lagrange points as something other than "an easy place to park satellites and space stations" wouldn't find anything short of full N-body physics as close enough to be acceptable. There is someone working on an N-body mod for KSP, though last I saw it was still in early alpha.

I've seen other people mentioning this, seems like a fairly common problem. It's never bitten me, and I'm not sure what I'm doing differently.

There's a post here that goes into the forces that cause rocket flipping. Staging can be a particularly vulnerable moment in an ascent because you're dropping a mostly-empty stage. This means that it doesn't have a lot of mass, so your center of mass isn't going to change much, but it's got all of its surface area still, which means your center of pressure is going to move more significantly.

I left KSP running over night, sitting in the VAB with a low part count ship up, and when I got up in the morning, there was a popup saying that I needed to close KSP because it was using too much memory. Not the normal CTD, but a hint that something was wrong none the less.

I'll chip in with not having a problem with parachutes in 1.0. A single starter parachute on the starter capsule resulted in a 100% recovery rate of the capsule provided nothing overheated. Then again, except for sub-orbital flights, I always went for a fairly shallow reentry (30km periapsis or so) which gave the atmosphere plenty of time to slow down before it came anywhere near the ground. I'd usually tweak the pressure trigger so that I got a partial deployment somewhere around 10km and full deployment at 350m, which would result in a long, slow descent. What kind of reentry profile were you using when you had these problems?

For that matter, the constant altitude method of landing is actually more fuel efficient than a suicide burn in most cases, but not by a large amount unless your lander has a low TWR. As your TWR increases, the two methods converge. The suicide burn is the most efficient manner of landing in the case where you're already on a not-easily-fixed suborbital trajectory.

I agree as well. 1.0 had a problem where deployed parachutes wouldn't burn up and heatshields affected the center of pressure but not the center of mass, causing flipping. And there might have been a few minor balance issues. Minor things considering the fact that the memory leaks seemed gone and we had a more realistic aerodynamics system, etc. Then in 1.0.2 (I didn't get the chance to install 1.0.1 before 1.0.2 came out, so I missed out on that one), the parachutes and heatshields are fixed, but now the balance is wonky enough that heatshields aren't necessary. The worst part about it is that we got a memory leak mixed in there. So mostly minor bugs mixed in with balancing. I'm having fun, haven't gotten frustrated by any of the bugs except the memory leak.

Yes, and according to the devs of that game, they had to upgrade to Unity 5 in order to ship a decent product. It wasn't as critical to KSP, so the devs didn't switch over to U5 as soon as it came out because they were in mid-dev-cycle for 1.0.

More specifically, if the game's not using a fairly comparable set of Unity features, it woudn't be a fair comparison. I check out the Unity dev logs from time to time, and some of the bugs that have already been fixed are incredibly specific to certain features. Random crashing in the raycasting code in the Win x64 client. Yeah, there was no way KSP could use that client before it got fixed because KSP uses a lot of raycasting. FWIW, that bug was fixed before the first round of community x64 stuff, but after Squad had said that Win x64 KSP was crashing every few minutes. Yes, they converted to U5 in beta. In fact the devs went to Unity 5 because of their 64bit U4 issues, if I remember right.

Two things: In career mode, fuel transfer isn't available until you upgrade your R&D building. Second, if you want to keep your rocket from flipping, you want to move the fuel up, not down. You want a high center of mass, not a low one. This post does a good job of explaining why.

I think they did a good job giving each engine a nitch. I think I've used every non-radial engine and SRB except the 48-7S and LV-1 multiple times. And yes, that includes the flea, though it gets the least use. Once I've got enough science parts to make it worth it, I wind up using the RT-5 for a few near-KSP biomes on a suborbital biome hopper. The RT-10 tends to get used as a second stage on my suborbital and lighter orbital craft. The rest of the SRBs get used during the first stage, either alone or to help a liquid main engine get up to a better altitude, or occasionally just to get it up to speed faster.

Have you changed the difficulty settings? I'm usually getting more than that for tourists that don't even need to land on the moon, just get into a suborbital path there. It takes less than 30 m/s delta-v to go from a 20km circular orbit to a suborbital path then back to the 20km circular orbit. I took one tourist on my first Munar orbiter and got over 700 science, XP for Jeb and Bob, and almost 100K funds profit, and that was with a seriously overbuilt craft that cost 60K because I used 2.5m parts instead of stacking 1.25m parts. And I've got another 7 tourists that want the same or similar flights plus two more that I haven't accepted yet. At this rate, as long as I don't start getting tourists that want to go somewhere else, I could do a one-tourist-to-one-biome plan and suck the Mun dry in all biomes and make a whole lot of funds. :-)

I'd be good with that, as long as turning it off actually gets it completely out of the way. I'm just thinking something like the aerodynamics, where there's a way for a mod to say "Yeah, I'll be handling this, don't waste any computation on it". Really, anyone that feels the need for something more realistic in this is going to have different opinions about what matters, so they might wind up modding it regardless of what type of life support you put in. Just look at the number of different life support mod options that already exist.

I think I'll be using a link to this page for people that don't get why their rocket is flipping or even the ones that argue that a high CoM leads to tipping. As is, I'm linking the simpler explanations several times a day (mostly on reddit).

Pretty sure it means that they won't get the benefit, not that they won't be able to run the game.

Ah, THAT's what you're basing this on. That's taking separate issues and tangling them up together. The resource system that they were talking about was originally planned for 0.19, they didn't like the way it worked, so they canned it. There was no "Hey, good idea, let's sell this separately!" talk. As for the fuss, the fuss was about them mentioning DLC and people going orbital because they bought the whole game and never wanted to pay for anything else. As to why they'd mention DLC, 0.23.5 was originally intended to be free DLC, but it updated too much of the game so they had to integrate it into the game. HarvestR was talking of a hypothetical situation, not any plans they had. The only things these two things had in common was that they were both announced at a big online event (three days of streaming) and both created a big stir in the community. They've already talked about the fact that they're planning what's going into 1.1, and yes, Unity 5 is on the tentative list. I can't definitively say that there will be a 1.2, but I'd be quite surprised if there's not.

It looks like this is getting resolved, but I'll chip in with ten degrees pitch below 1000m being way to much in my experience unless you've got a high TWR. The rockets I'm launching right now, I do a one degree pitch maneuver when they hit 35m/s, let it go long enough that the prograde moves off of vertical, then just follow the prograde up. Usually hits ten degrees pitch at about 3500m and 35 degrees at 10km altitude. These rockets maintain a TWR between 1.4 and 1.8 until about 30km altitude.