EngineeringWaffle

Also having this issue, the JSON fix worked. Had a big mission to Jool and have had to do a lot of save file editing to make it work. Between trajectories disappearing and docking ports locking up I've had to keep the saves folder open next to KSP because I'm in there so much fixing stuff. To be clear about what I was doing, I have a descent stage and a transit stage and was visiting all the moons of Jool. My descent stage would undock, go down to the moon surface, run experiments, take off, and then redock to the same docking port. I had this bug every time. One time both docking ports were bugged, the other time just one docking port was bugged. But the dock never worked without editing the save file.

Weirdly, I've noticed that you actually can set a maneuver node if you enter an SOI within the first SOI. So if you chart a course directly to Laythe (From interplanetary space), you will be unable to create any maneuver nodes around Jool, but you will be able to create maneuver nodes around Laythe.

A lot of the objections being raised seem to be of the format: "if non-fungible science rewards were added in an terrible and tedious way, the result would be terrible and tedious." Obviously adding non-fungible rewards would not be something you would just do completely randomly with no thought, it would need to be just as carefully thought out and carefully constructed as any other system.

So I want to be clear that I am not advocate for KSP devs to take my half thought through idea and implement it verbatim. I totally agree that if you had to unlock all parts by going to different destinations it would kill the games openness. At the same time, the only difference between Duna highlands and Duna midlands is that one is called the highlands and the other is called the midlands. And 3 missions to duna is the same as one mission to a more difficult destination. What I'm advocating for is some variation in reward and some kind of non-fungible rewards for select missions. Some examples: Perhaps most missions give you science that gives you the basic parts. But if you analyze all the different atmospheres on Duna, you get a special methalox hybrid jet engine that can work in low-oxygen atmospheres. Perhaps on Eve you could discover new explosive bolts you can use for very powerful decouplers. Maybe the arch on Minimus could unlock spiked rover wheels that keep rovers more firmly on the ground. Or if you make it to the pole of dres, you unlock a harpoon you can fire at really small bodies to reel yourself in and anchor yourself to the ground. Almost have the unlocks be like enchanted items. Not necessary for progression, but interesting to create cool and different builds. Another more surface level example is skins. I know, I know, but if completing a difficult challenge unlocks a cool paint job for your boosters, that could be a fun flex to unlock. Or, you could even do it where you can "skip" the tech tree and discover single parts early by visiting certain destinations. So if you survey the atmosphere of Duna before unlocking plane stuff, you unlock a couple of the plane parts. If you make it to Juno you can unlock the big hydrogen engines, but you don't get the big tanks until you unlock the tech tree node. There are a million ways, but the key thing is just that having science be only a currency is, IMO, very shallow. And what you are saying about wanting to play the game differently every time, I totally agree. I think the biggest barrier to that is probably that most stuff is mostly the same. Big tanks, little tanks, medium tanks. They are all just tanks. Big engines, medium engines, little engines. They are all just engines. Helium vs methanox vs ion is actually interesting. But there are few things like that in the game. I think putting powerful and weird parts that have tradeoffs behind different and difficult objectives could add a lot of spice. And then depending on what kind of play through you wanted to do, you might choose to visit different planets first. Honestly, a lot of people on here justify why their thing should be in the game with it being realistic. This is not a realistic game. But even then, they kinda did. Science conducted on the moon taught us what we would need to build, for example, a moon base. And doing those unmanned missions in the beginning taught us what the space environment was like and what you would need to survive it. It was only by launching and experimenting with the Saturn IV that we learned enough to make the Saturn V. The data gathered in space certainly comes back to earth where we actually develop the tech, but it certainly informs *how* it gets developed. The thing I do think they nailed with KSP 2 is just not providing any science for doing the same experiment twice. And not really rewarding the Mun enough to be worth grinding it. I hadn't really noticed tbh, I kinda forgot until you mentioned it just how much time I did spend grinding out science on the Mun, but you are totally right. At the end of the day, I just want to see more depth and personality in the science system. I think they could do so much more than they currently do with it.

Let me put it this way: The KSP1 implementation gets a 3/10 on challenge, a 2/10 on increasing depth of spacecraft design, and an 7/10 on adding personality. The current KSP2 system gets a 1/10 on challenge, a 1/10 on increasing depth of spacecraft design, and a 2/10 on adding personality. I think that one can do way way better than the KSP1 system. I think it could be made way way deeper. I think it could be given much more personality. Just a couple of examples We could have experiments that we actually do. Rather than just running a machine, we could have an experiment that is "Crash the inertial sensor into the surface of a celestial body and then collect the black box recording." Or "collect a sample of a dark rock that is scattered around the munar highlands" and then you have to build a rover to actually find one of the rocks. Or "determine the effect of prolonged exposure to a rocket engine on the Duna Icecaps." Let's actually do some stuff! We could have experiments that are actually heavy or big or otherwise impact vessel design. Like an atmospheric sensor that completes faster the more air you get into it. Like an air intake on a jet. Or a seismic sensor that is delicate, but also needs to rest directly on the ground, so you need to pull off a super smooth landing. Or a radiation sensor that is absolutely huge and consumes tons of power, but not very heavy. Just awkward to build a ship around and actually launch. I'm not sure how to add more personality, but more voice lines, animations, and experiment results would definitely be a start. I don't hate QOL. And I actually get that the KSP1 system wasn't the best itteration of a science system. But with KSP1 I could see the vision for the process to be deeper. I could see how every aspect oozed personality and begged for more tailored interaction. But also that it was limited by the dev hours required to pull off something more ambitious. I see KSP2 as giving up, just making it a button that increases your points.

I mean, I think one thing that is key is that KSP1 is not necessarily the gold standard. I think KSP1 did a better job of gamifying it and adding personality. But I also think you could do a lot better. I mean, yeah. And if they made minecraft diamonds 100x more common, you could just manually throw out 99/100 diamonds. And if KSP had a heat shield that weighed nothing and fully protected your craft you could just not use it. The logic "you could always just play the game as if X mechanic weren't an option" can be used to justify all kinds of terrible decisions. And the reason why I think going through and doing experiments manually is interesting is that it adds some challenge to timing, it adds something to ship design (you can't bury your science modules deep in your ship), and it adds personality in the form of the messages you get. If these same elements of timing challenge, design challenge, and personality can be added to the science system by some other means I would welcome that change, I'm not married to the KSP1 system. I just think that attention needs to be paid to their removal and not every QOL feature is actually beneficial to the gameplay experience.

I just wanna chime in and say I think there are key things OP got and is getting right that seem to be going over a lot of people's heads. 1) Running experiments was gameplay in KSP1. Perhaps you found it tedious, but mining for diamonds in Minecraft is also tedious. If you want to make the game less tedious you could just automate the whole thing. One button and you win the whole game! It was somewhat tedious, but it also IMO was somewhat challenging to get all the experiments to trigger as you passed over a tiny patch of lowlands. Or getting a kerbal to jump out of the ship to quickly "record his thoughts" while screaming through the upper atmosphere at 3,000 m/s. 2) Currencies in this game are super boring and tasks don't relate to rewards The fun quest rewards in Skyrim aren't the ones that give you $6,000 gold, they are the ones that give you a special item. That you do specifically because it's the way you get a particular bow for a build you wanna do. One of the things I think is missing from KSP is any quest rewards beyond currency (and science is a currency). Go to Duna to discover how to make helium engines, go to the mun to find a new type of rock to make bigger and better landing struts. Go to minmus to unlock some tiny engines. Now, how do you actually implement this? I don't know. There are a million ways. But it's pretty much just a fact that the science tree they gave us is pretty much the most boring way to do a tech tree, it's basically just a shop. At least in KSP1 there were interesting decisions about which branch you wanted to invest your points in. Here you can just invest in the main branch and then unlock whatever you need. 3) The personality that was present for science is honestly, completely gone. "You take a sample of water, it appears to dramatically increase the surface humidity of anything it touches." "You start say something dramatic and poignant about the plight of Kerbal-kind in this grand universe, only to be cut off by random radio chatter that the situation is normal" "Yep, it's dirt" This is what we got every time we shook up a mystery goo canister, or extended our magnetic boom, or examined our totally-not-store-bought thermometer. Now we get a report pop up with text in it too small for anyone to read, but a UI that's so large it obscures your whole screen and about 10 notifications saying we have already done that experiment because I happen to have backups of my science equipment. 4) Science gathering is no longer strategic and has no relevance to ship design In KSP1 you actually had to plan your trip and what science you were going to do. you could bring one science junior and a lab and run the experiment tons of times, or you could bring just the one science junior but you'd have to make it count. And if you wanted to do multiple mystery goo readings you had to bring multiple canisters. And so your first ship to the mun would just have one mystery goo canister, but the next trip you'd bring a few and a science junior and then the next trip you'd bring a processing lab and set it up on wheels with one science junior and try to get a bunch of biomes. It meant you changed your ship design depending on what you were doing. KSP2 you put one of every sciency thing you have on your ship and launch it. Now, if you don't like the science gathering aspect of the game, or if you just want to explore the sandbox while unlocking parts slowly, maybe none of this matters to you. And that's okay But still, I think we can admit that KSP2 science doesn't exactly have the same amount of personality it used to, and it doesn't create terribly interesting gameplay decisions. How that problem could be solved is an open question, and it's probably going to take more than 5 minutes to solve, so anything that is proposed here will probably have at least one major flaw or drawback. But just because there isn't an obvious solution doesn't invalidate the criticism. Also, while writing this I've realized: I have not seen a single silly kerbal thing the whole time I've played KSP2 . I haven't read any science reports, I haven't seen any particularly funny part descriptions. Really nothing. Which is kinda sad.

Something to note about this that makes severity higher is that if your delta V is zero, you can't plan maneuvers. So whenever anything causes KSP to not correctly understand your vessel or where it's getting fuel from or where it's thrust is coming from or whatever, you completely lose the ability to plan manuvers at all. I think that it makes sense to alert people when their maneuvers exceed their delta V budget, but still allow them to be charted. Especially given the delta V calculations seem buggy right now.

Yes, it seems to explode directly when physics are loaded. Sometimes when approaching the ship I will hear a bang and then everything is smashed. I obviously can't know exactly how far away I am from the craft at that time, but on laythe particularly I noticed that the craft would initially load floating in the air without any physics. As soon as physics loaded, the craft would fall and explode. I've also noticed similar things with large ships in orbit. I wanted to dock with my large craft and as soon as my smaller craft got close enough for physics to load on the larger vessel, it exploded into a debris field. The bug report you linked is interesting, but I haven't had issues with the craft sinking. I remember that bug from the release where I would leave my mun lander and come back and the mun lander would just be gone, sinking into the mun forever. This bug is much more explodey than the sinking into the ground bug. Let me know if going back to my save and taking screenshots/videos would be useful. If you feel it would be beneficial then I'll do it, it's just a real pain because it means screwing around on EVA for like 20 minutes trying to swim close enough to be in viewing range but not in physics range. I can also supply the save file, though I'll warn you that it isn't super consistent.

Autosaving on launch is something present in KSP 1, don't know why a similar system wouldn't be present in KSP2

Reported Version: v0.2.1 (latest) | Mods: none | Can replicate without mods? Yes OS: Windows 10 | CPU: AMD 5600X | GPU: Nvidia 3070 FE | RAM: 32 Not sure what is happening, but if I am burning prograde with my vessel at the peripasis, then my apopasis goes up. Which is expected. However I have a very heavy ship with the hydrogen powered engines, so I time warped at 4x speed. Then my apoapsis started going down. Back to 1x time warp, and my apoapsis starts going up again. It's like switching to 4x speed puts the engines in reverse. I do have engines on both sides of my ship (it's actually 2 ships docked together) but the engines facing retrograde are de-activated. I have the same number of engines on both sides of the ship. It's also a ship with a large number of parts that is somewhat laggy, so maybe that has something to do with it? I have a save file I can provide, but it is just over 20Kb, let me know if I can submit it another way. Also I don't have a craft file because of it being 2 ships docked together. Included Attachments: KerbalSpaceProgram22024-02-2419-58-26.mp4

VRAM is fine. It's high, but it's fine. I am not using any experimental or strange drivers and my resolution is 1080p. But setting my resolution to like 480p only gave me like 1 extra FPS. Now the ship itself is large. Quite large. It's a bunch of ships I docked together to go to Jool and visit all the moons AFAIK the part count is just too high for KSP2 to handle. I don't know if it's possible to see part count in the flight mode, but that's not really the point. The point is I pushed the game to the limit, but the game didn't push my hardware to the limit. Obviously it's possible to build something that will slow down your game. Anyone can open cyberpunk and crank the max raytracing reflections to 100 and tank their performance, whether they have a 4090 TI or a 2060 super. But when you do that and open task manager you see your GPU get maxed out and heat up. What I find strange is that the bottleneck doesn't seem like it's hardware. It seems like my CPU and GPU are totally fine and able to handle more, but the game just won't push them. For me, this is reminiscent of a problem I used to have with Minecraft on an old laptop where it would default to my intel graphics instead of my discrete graphics because Nvidia didn't recognize a generic Java app as being something it should enable itself for. So I'd be getting single digit FPS, but task manager would show no activity on my GPU. This is obviously not the problem, since the computer I'm currently using has no onboard graphics. I've also checked I'm not in power saving mode, that the game is recognized as a game, etc. but it just has the same kind of feel, like something is just not using the right system and bottle necking on something totally random. Maybe it's an Nvidia driver issue. But IDK, no other game behaves like this for me. And my card has been out for a long while now. But I don't even know how to tell if it's a CPU problem or a GPU problem.

Reported Version: v0.2.1 (latest) | Mods: none | Can replicate without mods? Yes OS: Windows 10 | CPU: AMD 5600X | GPU: Nvidia 3070 FE | RAM: 32 If you get far enough away from a landed ship obviously the ship unloads and stops having physics applied to it. Which is fine. But then if you walk back to the ship (especially if you are using time warp on EVA) then the ship will often just randomly explode. I had this happen a couple times on Duna, where the ship seemed to explode "in place." Often landing gear would get damaged and then the craft would tip over. I then landed in the water on Laythe and had a similar thing happen but this time the ship seemed to fall out of the sky from approximately 100m above the surface of the ocean, destroying it. On subsequent reloads I was able to position the camera to view the craft before physics was applied to it and noted that it was perfectly upright (not the orientation it was left in) and hovering about 100m above the ocean. When I got close enough it started having physics applied and fell. I unfortunately did not take screenshots, but I do have game save files I will upload as soon as I can. Included Attachments:

Has anyone established a relationship between specs and performance? I know this sounds silly, but when running the game with a big ship I was getting 2 FPS. I checked task manager etc. and nothing was being pushed. No CPU core was over like 30% or 40%, no GPU temperature spike and no part of my GPU was anywhere close to maxed out. Obviously potato PCs will have issues, but has anyone established if upgrading CPU or GPU will actually make an appreciable difference? It seems like it must be some kind of crazy bottle neck. For reference, I have a 3070 and a 5600X

No I mean a physical physical notebook. That you buy. In real life.