Rejected Spawn

I've been in strong favor of relaxing the shackles of extreme realism for a long time but everything has a limit. You basically took my limit, drag raced over it with a forklift leaving a skid mark reeking of burned rubber and flew off into the sunset on a gas giant at the edge of some other galaxy. KSP1 "gamified" a lot of stuff but was always about "what we already proved works". KSP2 seems to be about "what the majority of the scientific community agrees SHOULD work" with some minor wriggle room. Faster Than Light travel and wormholes are so far from realistic I'm not even gonna get into that can of worms, however one thing is certain; no sane and serious scientist would call it "possible". You'd be throwing the entire concept of realism out your transfer window by letting it anywhere near the game, letting it slip in as any kind of official content in any form would make my skin crawl. Want to mod your game and turn a simulator into a nonsensical playground full of magic? Go ahead and mod that in, I sure won't be getting in your way - but don't you dare make me lose every microgram of respect for the devs for all eternity by having them defile their own creation right out of the virtual box, DLC included. Regarding intergalactic travel I'd say the probability of that is impossible to calculate until we know more about what's going on with the engine, if you can specify an objects position down to a Kerbals neck hair you end up with pretty huge coordinate numbers to specify the objects position in relation to even a single solar system. Sure there could be some trickery involved across interstellar distances that gets around the problem but I don't think "intergalactic" will be doable because there is no need to develop such a thing. Galaxies are pretty big you know, not sure why you'd even want to go see another one when yours already has a hundred billion stars to investigate...

Arguably the most obvious thing that both KSPs would benefit greatly from would be to just include some actual significant variations in sizes. Sometimes we really need a single large wing, not 46 wobbly pieces that barely differ in size. Sometimes we really need a long 1.25m tank that only holds liquid fuel, not 10 stumpy little ones stacked like a noodly house of cards. Sometimes we need a motorized wheel that is really large but isn't just a self destructing joke with suspension range shorter than a Kerbals neck hair. Sometimes it would be nice to visit Eeloo with an RTG powered mining ship that has a part count lower than 3 digits. More directly to the point on the topic though: I'm expecting zero improvement in general parts variation over KSP1 when KSP2 hits the virtual shelves, however I am blatantly expecting that the post launch refinement of the game over time will at some point also include at least some missing pieces of the interplanetary puzzle. There seems to be some disagreement in this thread about including huge rigid parts because it would make the game easier... to that I say: The challenge in a simulation game should be in dealing with the relevant physics, NOT in dealing with how the game fails to accurately represent the relevant physics. I'm not demanding a completely lifelike simulation here, the word "Kerbal" is in the name of the game for a reason, I'm just saying the game itself shouldn't be what's getting in the players way; if you do the math on the fuel and thrust you should be going places instead of your rocket randomly going several other places all at once because the part count just got too high and the noodlestruts went full krakenightmare.

I'm just going to chip in and say that this AI thing is really shady and makes wild claims, on top of it being completely unsuitable for systems with too many types of objects all at once. That video claimed a simple simulation of a scrap of cloth was running at TWO frames per second by simply properly calculating it, I find that very difficult to believe if the physics engine used had been even remotely well programmed. The claim that the AI ran it at thousands of frames per second is less difficult to believe - because that simulation is extremely simple and the AI is likely just regurgitating the closest known approximation with terrible accuracy. Not once did the AI presented actually have to work on more than an abysmal number of types of objects at the same time, this is a huge red flag if you want to "sell" an AI that's competent and can be applied to complex systems. The claim that it "only" uses 9MB (or whatever it was) is for a pathetic number of types of objects in play. I'm now going to use math so if you can't stomach that, it's time to stop reading: To properly calculate the outcome of an impact between two objects you need to know their exact physical shape, weight, speed, rotational speed, angle of impact and point of impact. When you let the same two objects collide thousands of times you can get a number of good examples as well as edge cases - then all you need to do is match the closest one to what's about to happen next time and you have an ALMOST GOOD ENOUGH really fast prediction. Now let's add just one more object, Object3, and run the simulation for a few thousand times between that and Object1 and then again a few thousand times for Object3 colliding with Object2, your database now grew to three times its first size by simply adding A SINGLE OBJECT, this results in slowdowns since the AI has to dig through the much larger database for a match. Now let's ATTACH Object1 to Object2 and collide that with Object3, this introduces yet another factor for the AI to consider: how much force it takes to separate Object1 from Object2. The database is growing. Now let's attach Object1 to Object2 in a slightly differrent spot and we have yet another scenario that causes database growth. Now let's attach Object3 to Object2 and collide with Object1 that is attached to a second copy of Object1. At this point the AI is already using a huge database that is nothing like what that shady video would have you believe, and we're still on THREE object types. How many pieces do you build your rockets out of? I certainly never build a rocket out of just two pieces. I'm not sure at what point the database hits the terrabyte mark but for something like KSP2 I'm pretty sure the database size would make you pale in the face - and it still wouldn't be anywhere close to accurate... not to mention its selling point is completely lost because it would no longer be even remotely fast. I could go on but anyone who still doesn't understand is not going to listen to reason and logic, I have better things to do with my life. Final nail in the coffin for this topic: KSP2 is - as far as I understand - already developed to the point where it can handle big constructs accurately and it will use some kind of "grouping" of objects that greatly reduces the number of calculations needed in most cases and that still permits simply dividing the group if circumstances demand more finely detailed calculations. This system is close to perfectly accurate, runs fast even on low spec hardware and is already close to done. Anyone who wants to throw out this system 6 months before the game hits the market has no business giving advice to game developers.

Voted as close as I could to "Currently trying to get a passport and figure out how to get inside the studio at night to play in secret and leave mysterious feedback and crates full of just barely legal strength energy drinks scattered to speed up the devs because every single second until KSP2 goes full release is complete hell on earth that can only be somewhat alleviated by the KSP 1.8 update that is already taking so long it physically hurts my entire existence."

Thanks so much for the replies, sometimes this forum is just the best. @Incarnation of Chaos That version was certainly well hidden, I caught wind of something like it existing quite some time ago but the way M$ has since redirected me at every turn when I went on a hunt for it I started wondering if I'd gone crazy and just imagined the whole thing... or if I'd fallen for an urban legend or something. Been churning through this mess in my head for a few hours after a refreshing nap, I'm unaffiliated with any corporation so getting to stuff they don't want to sell to us mad scientists in early retirement might be tricky. There is actually a way around it by starting my own business, in Sweden it's possible to have a personal business that doesn't actually *do* anything, however it would require a deterring amount of paperwork to be handled yearly to keep it going without running afoul of any laws or regulations. Gonna keep thinking about this for some time, with all the food for thought you served me I must say thanks for the meal! @linuxgurugamer Are you absolutely certain that it doesn't force a reboot if you leave it running for a year though? According to specs and user experiences it doesn't force reboots on users other than for specific updates that are typically far between (if I understood things correctly) so it can easily behave for a long time but then suddenly throw a fit. All the work I do is so far unpaid (and even worse: much of it costs me a lot of time and money) and sometimes I just can't muster the motivation to keep at it for a while so at any given point in time I might just not touch the relevant PC for a week or month, during that time it might receive one of those priority updates and think it's being a good little OS giving me several days of warning to save the data... Typically my critical PCs are left running until they either start acting funny or the power goes out in my part of town, had one purring like a kitten for 500 days recently but then a power outage hit me... @mcwaffles2003 This idea did cross my mind long ago but I shelved it due to knowing my own unreliability; I'll totally end up not disconnecting the storage because I either forget or think I need to save another file in a couple of minutes but then get distracted and forget, or because I get tired of having to juggle it all the time and leave it connected for days on end. Though given that some of the stuff I'm gunning for patents on is of a much more dangerous nature than design work I might end up going with a removable storage for those things at some point, or attempt to airgap... One of the more complicated projects is a propulsion system that should work with astonishing fuel efficiency and generate abysmal waste heat, while the TWR is terrible the top speed should be remarkable and it would be super well suited for extremely high altitudes. Why is such a thing dangerous? Put it on a rather simple drone and you can sneak a payload across a border without giving most radars the slightest chance of picking up on it, the drone can then drop that payload straight down anywhere you want and still manage to get out of the territory undetected which makes the culprit nearly impossible to identify. Has military, terror and smuggling applications, I shudder to think what some people would use it for. This is only one of several projects, some of which are way more dangerous if misused, that make me prefer being a paranoid lunatic jumping at shadows over taking any minor data security risks. Exciting life though I suppose, also at some point someone will try the same stuff I do and they may have a far less accurate moral compass so it would be better if someone like me manages to restrict usage of certain stuff by slamming a patent on things that should only be used for humanity as a whole - that way only the good guys like Elon Musk get to use it and anyone else caught messing with it can be shut down before they sell DIY kits on ebay.

Now that the OP has been cleared up maybe we can make some use of this thread for OS related things not debated elsewhere? The title kind of forces it towards W10 though, an operating system I despise to a degree that can't be accurately described in a family friendly forum. Still running W7 on all of my active computers because the forced rebooting is unacceptable, I have things running 24/7 on some of my rigs and a single sudden reboot can ruin several ongoing projects - even a planned reboot usually requires a procedure that takes several hours to get back to full capacity. Does anyone know of any W10 version that doesn't force reboots? (I know there are some "hacks" that can stop it but just NO, that's not what I want!) Would appreciate it immensely if I could get a hold of a no-scummy-reboots version somehow. Since W7 doesn't even work on new hardware I can only hold out as long as nothing breaks, the next time a PC gives up for good here I'll have to use either W10 or convert to Linux which is an entirely different world of hurt for someone who uses a lot of software that lacks counterparts outside Windows... If anyone wants to know why I can't allow rebooting:

Basically sums up why this entire thread is a waste of time. I've been waiting for someone to shoot this thing down with some proper understanding of just how ridiculous the idea is to have voxel terrain on even a single planet out of the (probably) dozens in KSP2, there is no amount of wishful thinking that will make full voxel terrain work acceptably in a game of such a colossal scale until we all sit on brand new year 2029 hardware. Having said all this I'm going to hate myself for mentioning it but the thread about an asteroid impact event got me thinking about terrain height deformation to show proper cratering, now let's say height deformation eventually actually makes it into the game, it MIGHT be possible to simply lower/delete a region of the terrain and do plastered on voxels in that confined area to avoid running voxels all over the whole planet... never heard of any game using a mashup like this and it would take TONS of work to implement and produce an extremely limited (and nearly unimaginably bug prone) result but would put SOME voxel shenanigans in the realm of "not entirely disproven by everything forever", though still with the unavoidably awful performance-to-quality ratio that voxeling causes by its very nature. Now might be a good time to also mention that merely enabling decent quality height deformation in the game is such a monumental task that it would likely warrant becoming its own DLC to cover the huge development cost. I dunno about the rest of you fine folks but it leaves a bad taste in my digital mouth to make plans about DLC for a game that isn't even released yet. I could go on and on about how underestimating the development cost and performance cost of voxels has KILLED several promising games, but anyone still needing convincing that voxels is an awful idea for KSP2 isn't going to listen to reason anyway so I'd just waste my time and patience.

Hey guys and gals and bots and Borgs and lawyers and whatever other flavors of sentients might hang out here... isn't this thread missing a very major difference in what sets KSP1 and KSP2 apart? KSP represents largely "what has been proven" while KSP2 generally represents "what we think should be possible". [1] The more you look at this difference the bigger it becomes; we have made a huge number of real life rockets that run on a fuel and an oxidizer - we've yet to lose our minds enough to aim beyond the sky by means of nuclear explosions but we've done the math and it should work. Making a little bit of an assumption here I'd call KSP2 "the math SHOULD be right" and KSP1 "the math WAS right" which is a huge difference in approach. I've been strongly in favor of KSP1 adding more stuff that Kerbals (who generally don't die of age, their only gene pool chlorine is accidents) could use that humans with our fragile build and limited life span wouldn't be able to try if following all of our morally generated laws, however I stopped favoring this approach quite as strongly once KSP2 was announced as it finally lets us step beyond the relative safety of what has a counterpart in our own history and aims toward a reasonably likely future where we had little choice but to actually put our craziest ideas to the test. Playing KSP1 is not at all unlikely to end up being a kind of "classic and conservative" approach to space travel once we finally get our hands on the KSP2 universe that lets us throw caution out the window and immerse ourselves in the madness of science at its most experimental. [2] [1] When dealing with experimental stuff where the math seems solid we will inevitably end up with some stuff that gets disproven, please don't derail the entire thread because of yet another Metallic Hydrogen rant. [2] See [1] (and please SERIOUSLY DON'T get into the metallic hydrogen, WE GET IT already!)

Hey @GoldForest maybe it would be better to title the thread Your Multiplayer Co-op Adventures* to put emphasis on the wonderful individuality of our dreams about enjoying the game with our undoubtedly morally upstanding and competent friends? * Acronym similarity to anything ever is totally super coincidental and unintended, any ancient pop song that may be playing in your head is your own responsibility and totally unrelated to the intents and purposes of this post, you are hereby also informed that the cake is a lie and you should do a barrel roll after you add moar boosters.

I would like to strongly urge Star Theory to read the forum but not personally post anything other than "info nuggets" until near the launch of KSP2. The reason for this is simple, and has become a globally known rule among those who frequent forums of all kinds: LURK MOAR. There is never a shortage of opinions on forums, many of which are based on insufficient and/or incorrect information and/or logic. In addition to this an opinion that has been expressed doesn't need to represent a significant number of people. Thankfully this forum has a "like" system so that very well thought out (or sometimes amusing) comments can easily show their popularity - though unfortunately the system also tends to generate a lot of likes for comments that basically border on being plain nasty rebuts to other things posted but no system is perfect. There is a huge wealth of knowledge in this forum, and it's gravely fragmented and drenched in a sea of harmful misinformation and extreme opinions from members representing abysmal fractions of the playerbase. Star Theory should of course join the forum in full at any time if they feel confident they can clear all of the following criteria: Strong "6th sense" for knowing when a forumite does or doesn't know what they're talking about. Fully ingrained understanding that forumites are NOT average players; selection bias is very pronounced. Ability to calmly select what to reply to and leave as little room as possible for faulty interpretations when replying. Have a good eye for which types of forumites agree or disagree with an opinion, the like-system is often helpful but far from perfect. Not lose a single minute of sleep no matter what is posted, devs are people too but on their own forums they don't have the luxury of everyone remembering this fact. Until then I once again urge the "lurk moar" approach, and hopefully I have just wasted my time telling them something they already decided to do since it's the only approach that doesn't result in disaster when dealing with a forum of this size. Keeping interaction with the playerbase at a manageable level and not listen too much or too little or to the wrong forumites is something many studios have failed spectacularly at and in the end the games suffer.

I don't see myself ending up doing anything beyond rescuing stranded survivors left by my incompetent friends in various places... right outside Kerbin. Doubt they'd manage to get further. Other than that in an ideal world where I'd have a bunch of friends and they all had a measurable degree of skill I'd love to operate a giant spaceport they can visit! Build landing platforms just because it looks cool even if there's no gravity, extend some docking arms for bulkier contraptions, maybe even set up some shielded hangars and stuff. Sprinkle captured asteroids around for decoration. Oh, also it would be fun to build a huge race track around a colony, have players duke it out in nuclear powered rovers, all dirty tricks allowed!

I'm at the very least all for the idea of stage automation. Being able to have stages fire in response to specific conditions would be very nice for contraptions with a ton of small stages. Let's say we just want to burn and then decouple a few cheap stages to get somewhere on a tight budget, being able to tell the next stage to fire when a specific event happens (like a selectable tank runs out of fuel or there is a specific atmospheric pressure) or a set time has passed since a trigger was activated (such as 45 seconds have passed since the previous stage was activated) would be super nice. Being able to completely remove any stageable part from staging and setting conditions on it individually would also be really cool, though it would need the ability to have more than one condition and fire on both and and or conditions; like fire an upper stage booster if the decoupler attached under it has already been decoupled and the atmosphere is thin, or activate emergency landing parachutes and deploy aerobrakes if Main Tank fuel drops to zero or the craft is losing altitude too fast while in a low part of an atmosphere. Finally there needs to be a way to disable and enable all this stuff in flight, obviously. Of course if you pile a ton of conditional stuff on a ship you can program it to reach a stable orbit or even land safely all by itself but that's not really autopilot in the sense that anyone else did the work for you, the player still needs to figure out every last detail and the only difference to a flight is not having to do it in real time. (Well yeah, also being able to launch a copy of the automated ship if needed.) Essentially I'm in favor of the OP but don't agree with calling it "autopilot", neither do I like the idea of adding a "real" autopilot - the ships should just be more "programmable" by the player and not go play the game by themselves... unless there are such amazingly programmable parts that the player can personally rig a ship to go play by itself, that's the opposite of "cheating" - would be more appropriate to call it "impressive". As a bit of a sidenote some gamers have physical disabilities and KSP1 has certainly not been kind to people who have trouble timing their staging and such. Taking the quick reflexes requirement out of the picture would no doubt be welcome to some unfortunate players who'd rather their efforts not be thwarted by the same body that already causes enough issues in their lives. I personally know what it's like to end up with injuries that don't get much better in years and get stuck at home with nothing but a meager wallet and a PC and nothing your body can handle but play games, the last thing you want at that time is for an injury or disability to also get in the way of those games.

Finally we have the "like system" back, that was really about time. Figured it was better to leave the whole forum alone and wait for it to come back than go on a commenting spree quoting everyone who said something good, @IncongruousGoat deserved a like much sooner and I'm probably going to re-read a whole mess of threads looking for more comments I found to my liking. Getting down to business on the main topics of the thread it's very clear to me that no single one of the participants has a good enough handle on all three keys needed to make it work. Key1: The real life specific properties that must be matched to substantially alter a planet by way of an impact event. Key2: How this needs to be modeled (simplified) in a game to make it work properly. Key3: How much work goes into doing this at a professional level. For the record I'm missing Key1 to a degree that can't just be ignored, I already hinted at this earlier in the thread but to ensure no misunderstandings; I don't know the specifics well enough to get into that stuff beyond knowing that it takes way more than most people would think. What I can get more specific about is Key2 and Key3, at least to a level that can clear up a big pile of faulty assumptions already posted in the thread. Key2 is never going to be a complete match for reality, we will have to settle for something that looks like a very simplified version where there is no "real" cloud of shrapnel spraying back into space and no atmospheric phenomenon beyond shockwaves and minor cosmetic stuff. For a single planet being the only one possible to get hit by an impact you can get away with not having to model the impact site - you can ninja swap the entire model of the planet if it's one sitting close enough to its sun to only have a thin crust on top of a molten interior. The moment we include any other type of planet on the list of possible impact targets there is going to be a huge bump in how much modeling needs to be done to make it work. At a minimum there needs to be a crater generated where the impact was, this crater generation needs to be possible to apply to any spot on any of the presumably dozens of planets that will be in the game so they ALL need to have some way of getting their meshes deformed - we can't just switch out the entire planet mesh for a premade one. The next issue after that is how to make it work even if a player drops another asteroid in roughly the same spot, the crater generation needs to be really smart and not just keep raising a ring and making a hole deeper. The impact can't send millions of planetary and asteroid fragments flying back to space (mainly for performance reasons but other reasons as well) but there needs to be some remnants of the impact so there must be a scatter system in place that sprinkles fragments from both the planet and the asteroid in a huge area around the impact site. For planets with life on them there must be a scorching around the impact, removing any life forms in a radius. For planets where the impact hits an ocean there needs to be either a way to send a colossal wave from the site or the planet must be covered by dust and mist for a long while until everything has "settled down" and the planet is revealed again. Any icy planets would need to thaw near the impact for some time and then gradually freeze again over time. Key3 is where we see if this entire thing ever has a chance of being included in the game. For a single target planet with no other effects than replacing the surface; yes it can be done in a relatively reasonable time frame. For the "full version" we can with total confidence say that it would be a monumental undertaking that is absolutely impossible to get into the game before launch. Some of you have probably started to realise what a mountain of work is needed but I'm betting nearly all of you still think it's not even 10% of how much it actually is, if you thought of mods as a comparison just STOP because you clearly don't understand the enormous differences between what's acceptable for a mod and what's acceptable as official content for a game that has to live up to the expectations set by a global hit as its predecessor. As a somewhat dull but realistic conclusion the Asteroid Swarm Game Mode is not economically justifiable as anything less than its own DLC. I'm not enough "in the know" about development budgeting to start throwing exact figures around but I don't think anyone who understood my post here would expect to get ASGM included for free via patch anyway.

If we're going to be adults and not sugarcoat this I have no choice but to call your opinion very self centered and extreme. When developers make the call if something is better off as a mod or as core content (aside development cost) they have to look at what it adds to the experience and what it alters or removes from the experience, taking a stance that even if it's completely optional you still say it ruins the game for you is flat out unreasonable (like suddenly refusing to shop at a grocery store because they start selling one additional type of potatoes) and doesn't give your input in a discussion like this any significant weight. Rather than threaten refusal to buy the game if you don't get your way over a detail almost nobody would even bother including in their overall summary of the game it would be a lot more constructive to just point out that it could be its own game mode and not even appear as a toggle for whatever mode you'd be playing. With that established I hereby propose a new game mode; Asteroid Swarm - a mode that not only includes just that one giant asteroid but actually several of them spaced far apart in time. This makes it far more viable to also have some bonus ones that are too small to be expected to cause visible damage to any planet or moon they potentially hit, which means they can be captured and kept as ultimate trophies by the player without the extra workload of adding "impacted" models for all the planets just in case some player hits them. They'd still be far bigger than any asteroids in "Sandbox" or "Progression" and on trajectories that could threaten stuff in places the player is most likely to go, however making this swarm endless would turn it into a babysitting task so after some time it'd have to end properly and let the player know that the danger has finally passed. This entire game mode would add at least one type of long sought after "story", it gives ample reason for the Kerbals to actually go places which would really be a selling point to a chunk of players who don't like not having clear goals.

I'll admit this idea is interesting and well worth at least considering, though you'd really need something a lot bigger than a meek "mile" to pose a realistic threat to anything the size of a planet. For the sake of keeping things on a reasonable scale, at most any incoming really big asteroid could maybe melt the crust on a planet that already has enough residual heat that there is lava under the surface... I'm certainly not sufficiently familiar with the scale of any of this to start throwing numbers around, what I can figure out is that the asteroid needs to be of colossal size but that makes it impossible to divert unless done a very long time before the impact and that doesn't fit well with having it already in motion around the same sun as the target planet. The most reasonable thing I can think of would be that after a century (or something like that) on Kerbin this asteroid will pass by (way outside Eeloo) with great speed and be on a collision course with a planet orbiting another sun; players would have enough time to just barely push it off course if they quickly manage to catch up. Rather than have it go on some crazy trajectory if redirected it would be safer to just have it hit the sun - the target planet could just happen to be in the way if nothing alters its course. The reason the asteroid wouldn't be possible to keep as a trophy regardless of outcome would be that it is literally something that is supposed to kill a planet; either every planet in the game would have to be built to show the results of an impact of that magnitude so it doesn't matter where the players sends the asteroid - or the asteroid is too massive to control beyond choosing between an impact on that lone planet or safely hitting the sun right behind it. Certainly would be fun to see some truly big asteroids in general though, preferrably with enough geometry to be anything worth exploring up close (finding some neat crystal formations in a hollow would be epic) and an overall mass that makes them the ultimate trophies to put in orbit around whichever planet our little doofuses happen to make their home. Would make fine backdrops for stations too, anchoring an orbital shipyard to a glimmering giant rock would make it a real sight to visit instead of a mere stepping stone to other areas. Honestly there are few things as majestic as coming in to dock on something colossal, I learned that when I made a 30m wide orbital refueling station and sent a drone to switch out a detachable tank cluster... was the most memorable moment I've ever had with the game, even beat the classic "your first successful Mun landing" (seems like nobody ever feels anything tops that) by a slim margin. Ugh it still sends a shiver when I recall the suspense of slowly floating towards the station that covered the entire screen...

Well if that was what you meant I can't agree with you regardless, the "spring system" works as a replacement for firing the main engine for about 2 seconds to get clear of the base in an extremely low gravity environment. The key here is that if you tried building the same craft on Kerbin it would possibly topple over and explode on its own; you can't put it on just any launch pad but by fragile I clearly don't mean it would explode from a force equivalent of using its own engine. The surface gravity on Minmus that I stated as a specific example due to its ease of access from Kerbin (and instead of the Mun that has a much higher surface gravity) is just under 0.5m/s2. Now let's say we need 200m distance to the base to fire the main engine safely, reaching this height requires launching with the final speed an object would have when dropped from the same height. This speed comes out to about 14m/s if I did the math right, at 0.8G that's less than 2 seconds of acceleration and a distance less than 15m and should be a tolerable stress for anything that can just barely survive actually using something like the Orion. Honestly this is the first time I did the math on it properly and it's even more realistic than I expected, my first "guesstimate" was a little over 1G for 2-3 seconds which is still within a reasonable Kerbal "spring launch" territory. In practice it's probably no better than using a 3 second booster stage when it comes to getting clear but it's very Kerbal and doesn't leave a bunch of debris raining down on top of the precious colony full of loveable Kerbals. Heh, looks like I just convinced myself to actually vote for this spring launcher for more than just the comedy, hadn't even thought of the debris aspect until now...

I was in fact very clear with every aspect of my reasoning, the highlights don't mean you can just not bother reading anything else I've said - I use them to put emphasis on an important point so that a skimming reader can get a good idea of what my post is about and read the rest if the topic is of interest.

Surprised nobody has mentioned the elephant in the room yet, or did I miss a post somewhere? Some propulsion tech can't be used anywhere near a base if you want to keep the base. Rather than bog such a fragile and destructive craft down with additional tiny stages to get it away from a ground colony it would be nice to just give it a good "push" of enough magnitude that it can fire up after a couple of seconds and go on its merry way. Not sure I'd specifically choose a rail for this purpose but a truly Kerbal insanity would be a spring loaded launch platform, if they can build an Orion they can surely boing a sizeable ship 200 meters straight up from a tall equatorial spot on Minmus? Would be far less tedious to assemble fragile ships in a low gravity colony than fly load after load of materials to orbit and still need to push it away from the builder with something other than the main engine. Obviously the launch mechanism would have its structural limits (not too sure about being able to catapult an entire stellar exodus ship high enough to fire up safely) but it would certainly be in the Kerbal spirit of things to put an Orion on a catapult and call it a "safety measure". Launch clamps would probably have to be possible to autorelease at the end point though, getting the timing wrong there would still mean going to space but in many more directions at once. Oh for the record I'm against space elevators in all forms for numerous reasons that others already stated, in addition to several other gripes I won't even bother listing because the odds of space elevator nonsense being in KSP2 is so close to zero it's not even worth debating.

Something that could be interesting would be to stop treating "science points" as science and as points. Give KSC a proper databank where all scientific data is stored as raw data so that KSC scientists can write numbers on post-it notes and toss them in the air all day until two notes land next to each other and give the scientists an idea to try out something new. Rather than force a grind or force players to get specific data from specific places there is a variant that makes a bunch more sense: Scientists will figure stuff out whether you tell them to or not as long as they have diverse enough data, let's put that into a logical but still game friendly system. First order of business: "Click to unlock" is not how real science works. Your KSC scientists should always be actually using their peanut brains rather than sit and wait for you to collect a point of data and then instantly turn into geniuses for a split second and solve a giant physics riddle in one go because you told them to - which magically consumes the data. Even if you twist it into whatever it's been called in this thread, "engineering points" and such, this approach is laughably unrealistic. To figure out something new you don't always need case specific data - but it helps! Rather than having to fire a rocket in vacuum to instantly know how to make a great vacuum rocket or drive a rover with a seismometer to the north pole to figure it out; collecting general data should provide a good base to know how to figure any stuff out over a long enough period of time, then on top of that using a rocket in a vacuum should provide a great boost to how fast another model of vacuum oriented rocket can be developed. Putting this together we would have a tech tree that gradually unlocks itself and you just tell your scientists what to focus on the most, but if you don't go out there and take measurements and try out parts this progression will take hundreds of years for each part. Each different rocket you test makes the next rocket unlock faster, each rover wheel you drive across a new surface makes the next rover wheel unlock faster, you probably get the picture. As for whether to have the player always dictate what to research or let the kerbals go at it as they desire is a matter of taste, I'd suggest for normal difficulty just let them go at it if the player hasn't told them what to do next and for high difficulty the player needs to keep the kerbals from losing focus by always telling them what to work on next. A queue system for research would pair well with this.

This thread is in need of some major expectation management. * Just because there is something to be found under water doesn't mean the game will ship with submarine parts, it only implies there are ways to make a craft heavy enough to get around down there at all. * Putting an ice shelf on an actual bit of ocean is probably doable without too many issues but making it possible to drill a hole in it is just as unrealistic as making the terrain itself deformable; at best it's a mountain of work and a giant performance hog for a very minor feature. You can expect stuff like that won't be in KSP2 for at the very least the next several years and is extremely unlikely to ever be implemented - in part also because once the game ships the physical properties of planets can barely be modified at all without risk of destroying a number of unlucky players painstakingly built colonies. They have yet to showcase the colony construction in detail but we have been told colonies obey physics and it seems likely that the process of constructing them is similar to how we construct crafts. Unlike crafts however the colonies built on a surface seem to anchor themselves to the ground in some fashion, if this is restricted to only specific "foundation-type" pieces or is automatically applied to any piece that has nothing supporting it is unclear at this time. Given that we don't know how the anchoring works (or if an unanchored colony is even registered as a valid colony) we have no way of telling if a freely floating colony can be created, neither do we know if colony parts come with pressure constraints that make them impossible to submerge in an ocean. Personally I'd love to have a colony at least partially submerged to get a nice view of what might be down there, however the only thing implied so far is that "something" is down there which means it could very well be nothing more than some surface rocks with a color variation and some occasional easter egg. TL;DR It's unwise have any expectations about ocean colonies and ocean features yet.

Finally the Non Atmospheric Constructor suggests I have a NAC for this, hahaaa.... haaa...... yeah. Anyway there are only so many variations that actually end up being creative; without adding meaningless words you have the location which I figure can be described with [Orbit, Space, Vacuum, Non Atmospheric or Micro Gravity] and the function which boils down to some variant of [Assembly, Building, Construction/Creation, Making, Forging or Fabricating] and the only fitting word I can think of for something that is an open volume (as opposed to something like the VAB) would be a "zone". With these very limited options that are apparent after a good minute of thinking about the possibilities or a few minutes if you need a dictionary (not everyone has a large english vocabulary) I still completely seriously have trouble understanding how Star Theory couldn't at least nail a really good candidate between all of them over the course of however long they've been working on the game. So yes @GoldForest I really meant that it kind of shocked me a little, that part wasn't a joke. Not sure if anyone else feels that it's a bit premature to have this thread going but I for one would like to see the thing we're trying to name up close and in action in order to come up with something that will be truly relevant, I'd much prefer to infuse its unique flavor of appearance and function into the name if at all possible. Anyway I've binge watched 15 hours of anime and need sleep, no more suggestions from me.

VAZ. Vacuum Assembly Zone. Feels a lot like the sibling of the VAB.

MGF. Micro Gravity Forge. Simple, short, clean. And boring but matches the SPH in style.

For lack of anything better to do I'll make some real suggestions, now that I've mostly recovered from the shock. BOB. Big Orbital Builder. Soft start, it's still not too serious but easy to say and makes sense.

OLGA. Orbital Large General Assembler, because that thing should enable a ridiculous build volume. Also I don't actually like this name. SAZ. Space Assembly Zone. I don't much like this name either. GARY. General Assembly Rocket Yard. Yes this is just getting worse. SCAM. Space Craft high-Altitude Maker. Bad enough yet? COW. Constructor Off-World. ..... Why am I even posting these? Because I don't believe for a second that Star Theory haven't already decided on a name unless they publicly state they need help coming up with one, the only thing this thread could hope to accomplish is give them an option to upgrade what they have already decided if someone actually manages to come up with an incredibly good name - or far more likely just give us people with too much time on our hands a few good chuckles. If anyone thought any of the above jokes was anything but awful then feel free to repost that one, I'm not upvoting any of them and only posted for a few uninspired giggles so that's why I deliberately went out of my way to go against the rule of only submitting one suggestion per post. Edit: Wow it does seem like they actually for real haven't decided yet, thanks @TBenz for link to proof because I can still barely believe this. Not gonna listen to that an extra time to make sure though, the audio was a dozen flavors of discomfort for my ears.