GregroxMun

I've released 0.6.0

Such cruelty.

Jebediah, having used his fuel to jetpack down to the surface of Jool's moon Bop, waving to nothing in particular.

RSS/RO does not do what I want, which is balance the stock parts to have realistic part masses and unrealistically low specific impulse. SMURFF does the first part to stock parts, but that's neither RSS nor RO. The I don't care enough about living to continue

It would sound like that if you didn't actually pay attention to anything I was saying, wouldn't it?

KSP rockets are currently balanced against the 1/10th size solar system incorrectly. The problem is this. Real rockets, with realistic performance, would get insane performance in the stock system. The mass ratio from payload to fuel would be huge. This is all well and good if you're a space program on a budget, but it doesn't work so well if you're trying to make a video game with any kind of challenge. So KSP has done two things to reduce the effectiveness of rockets and result in payload/fuel ratios that are much more reasonable (though still much better than real world mass ratios). This is accomplished by increasing the mass of engines, the dry mass of fuel tanks, and the mass of some structural and crew parts. But the specific impulse of the rocket engines themselves are more or less within reason for real world kerosene/liquid-oxygen or some hypergolic rocket engines (a reasonable fit for the density of KSP's fuels and engine throttleability, in fact.) This method isn't bad, but it's not physically correct either. The Tsiolkovsky Rocket Equation: Δv = Ln[m0 / m1] * Ve m0 = Initial mass, m1 = dry mass, Ve = Exhaust Velocity = 9.81 * Specific Impulse, Ln = Natural Log function, Δv = Delta V. This equation, which many of you will have seen or even used before, is the embodiment of the rocket in equation form! And it can give us a hint as to how to scale rockets, we just need context for the Δv requirements. Another equation, the Rescaled System equation, will help us. It turns out that the delta-v to pull off one manuever is multiplied by the square root of the rescale factor. The rescale factor going from the Real Solar System to the Kerbol System is about 0.1. (Technically it's closer to 0.09091 but let's not be too accurate, these are estimations after all) Δvrescaled = (Δvdefault)^(1/2) It turns out that rockets need 0.31623 = 10^0.5 times less delta-v to get around the stock system as they do to get around the real solar system. This is fairly accurate, as 3200 m/s is a fairly standard launch into Kerbin orbit, and that corresponds to a real world delta-v of 10,000 m/s, which is about true of real rockets. It also works on orbital periods too, but that's not important. Anyway, how do we get all rockets to have exactly 0.31623 times less delta-v? The variables are the mass ratio (which is what KSP changes) and the exhaust velocity. But the mass ratio is not a linear change. The correct way to scale all rockets to have 0.31622 times less delta-v is to reduce the exhaust velocity (or the specific impulse). So my proposal, and I say this not having tested this out in game to see how it feels--hence "conjectural", is to rebalance parts so that they have realistic masses, and instead reduce the Isp of rocket engines to 0.5-0.32 or so. I can forsee some problems with this plan, in addition to the specifics of totally rebalancing a game that has been released for years now!

But the game is *more* balanced if you do it the other way around. The Skiff can actually push its fuel tanks with TWR > 1, and you don't get unrealistically huge amounts of Delta-V in the CSM. That, specifically, is the problem here. And if it was a specific balance choice and not a mistake, then why would the Isp numbers be so close to each other's real counterparts instead of something refined by balance? Also, 400+ seconds Isp is unbalanced. KSP rockets are already much more forgiving than real rockets, and 400 seconds of specific impulse with a dense kerosene-like fuel gives you a lot more Delta-v for the same fuel tank volume than a hydrogen rocket of the same size would. (This also happens to be a problem with the Nerv NTR) Anyway I want to make it clear that I'm not asking for a balance pass, not yet anyway. For now at least, the Skiff can keep its intended 412 seconds of specific impulse as long as it's on the Skiff, and not the Wolfhound. Heck, to make it easy for ships that are already launched, you could switch the model but not the part name and title--so people's ships can still work as they built it, just using the correct appearance. Until this bug is fixed (or if I'm not playing with the patch I made that fixes it), I can not bring myself to use these two engines. I know that might sound silly, but to be fair we've made do with the stock engines for a good long while. You're right, you're right. But it is incredibly frustrating to see people defend something which was clearly a mistake. This is something that occurs way too often, people will actually defend bugs, let alone poor design choices and mistakes. Yeah, I did it about March 15 days too late though :v

That is unacceptable because these have nothing to do with the game balance. From a gameplay standpoint it doesn't matter which engine model goes to which stats, and I'm sure Scott didn't realize just how switched they were when he said that. The idea that anyone can defend this simple mistake is exactly what's wrong with this community. This isn't a balance problem, it's a very basic art problem, and a very important authenticity problem.

* Fix reentry VFX showing from hidden side of vessel (ie. showing through the vessel). FINALLY * Removed Red Shell. Nice. But I still don't see any fix for the swapped engine models of the Skiff and the Wolfhound.

Well that was fun.

I am not ready to update, but I am ready to show off my progress. Full album here

Here's a photograph recovered from the archives of Echobal Space Center, of the first picture of the whole disk of Echo taken from space. Echobals use orange light most of the time, I had to find a dark closet with a white LED flashlight (my smartphone camera's flash gave it an ugly reflection) to get this lit well. The Echobals were very nice about me having stolen a copy of this precious photo, they only kept me in jail for one year.

Use 1.3.1 until it is rewritten. There's no major changes between the two versions.

I am very frustrated to tell you the answer is yes, because a week ago if someone asked me if I was going to update them to look like the new artist's illustrations I'd tell them that's absurd, and mine never really resembled the artist's renditions to begin with... except, they really rather did in large part. My original plan was to keep them their own thing, they'd be a "kerbalized" version of the system, and Kerbalized can mean different. But then I got talking with @eggrobin et al about TRAPPIST-1 and Principia, and realism really spoke out to me and we discovered "water tongues." And it turned out that with the new radii and masses known for the planets, many of them have to have different characteristics than they initially had. It's exciting, and I'll be ready to show it off soon enough. Here's a work in progress preview:

This is also a problem caused by PostSpawnOrbit and Kopernicus' automatic assigning of FlightGlobalsIndex. See, the game thinks Kerbin is in orbit around the Sun, so it will try to go Kerbin (because it's the default), Moho (because it's next out in distance from Sun), Eve (ditto), Jool (ditto), Minmus (because it's a moon), Laythe (next out), Mun (next out), and Kerbin (because it's next around Jool.) But then the next after that is Moho! The only fix would be to add FlightGlobalsIndex and Thomas has made it clear in the past that we should not be specifying that on our own. I'll bring the problem to his attention though.