|
meowmeowmeowmeow posted:Why is this easier than doing loads between bodies but not modeling the deflection? You could get axial/shear/moment loads between bodies, see if it exceeds a limit and if so break the joint vs adding on the spring-hinge stuff and looking at deflection. You'll lose the modeling of a joint starting to lose position leading to increased loads but you'd fix the wobbly rockets issue while still having them come apart. You could do this, but then load forces would not propagate between parts -- as long as the joint between a part and the rest of the craft can withstand the forces generated by that part alone, everything is A-OK. If you say "why not propagate the forces", well now you're writing your own physics engine modelling internal stresses. CrypticTriptych fucked around with this message at 01:29 on Mar 9, 2023 |
#
?
Mar 9, 2023 01:26
|
|
|
|
| # ? May 24, 2024 15:19 |
|
|
queeb posted:Looks like the technical director was fired or quit, always a good sign! hahahaha ksp2 is never going to be a good game 
|
|
#
?
Mar 9, 2023 02:08
|
|
|
There's no such thing as a rigid body in the real world, any time you're doing rigid body physics you're basically in fantasy land trying to tune something to be close enough to the real world to meet people's general expectations. That's why you get all sorts of hosed up results when someone clips one rigid body inside another. Of course the problem with using springs is that real-world deflections are on the order of micrometers, and you need very fine timesteps to avoid massively overshooting when you're simulating how far something deflects.
|
|
#
?
Mar 9, 2023 02:08
|
|
|
Counter point: the technical director is by and large responsible for the product that was shipped, maybe he’s responsible for the state of affairs being not good.
|
|
#
?
Mar 9, 2023 02:17
|
|
|
Are there any actually interesting gameplay decisions that result from wobbly rockets? As far as I know, the only solutions are "use weaker rockets to launch smaller payloads" or "slap a bunch of fuckin struts all over everything". The latter is boring as gently caress, and the former kinda sucks.
|
|
#
?
Mar 9, 2023 04:48
|
|
|
Vizuyos posted:Are there any actually interesting gameplay decisions that result from wobbly rockets? As far as I know, the only solutions are "use weaker rockets to launch smaller payloads" or "slap a bunch of fuckin struts all over everything". The latter is boring as gently caress, and the former kinda sucks. I guess if you only want to build perfect rockets and you automate all of your gravity turn with Mechjeb then wobbly rockets don't appeal, but seat-of-the-pants flying is a large part of what makes KSP fun for me. I love trying to make it to orbit with a design that I only discover has less than ideal flight characteristics once I am already blasting through the lower atmosphere.
|
|
#
?
Mar 9, 2023 05:17
|
|
|
meowmeowmeowmeow posted:Why is this easier than doing loads between bodies but not modeling the deflection? You could get axial/shear/moment loads between bodies, see if it exceeds a limit and if so break the joint vs adding on the spring-hinge stuff and looking at deflection. You'll lose the modeling of a joint starting to lose position leading to increased loads but you'd fix the wobbly rockets issue while still having them come apart. What CrypticTriptych said, but also, a system where a thing breaks instantly if force > joint strength goes to poo poo in video games where time between updates might be inconsistent and calculations during a scene load or whatever can be wildly inaccurate. Deflection & springs have give to them, they're modelling an analog system. That also resists a momentary fuckup. Without that you have a brittle result and need to make strength high enough to withstand whatever your most common error is -- which likely means unreasonably strong. Vizuyos posted:Are there any actually interesting gameplay decisions that result from wobbly rockets? As far as I know, the only solutions are "use weaker rockets to launch smaller payloads" or "slap a bunch of fuckin struts all over everything". The latter is boring as gently caress, and the former kinda sucks. There are two background questions: a. How big is a big payload to you? b. Is KSP primarily a game about designing rockets or a game about flying them? To me, KSP is a rocket design game, and semi-realistic physics are the challenge you're fighting against. (Which could be substantially improved -- just because I defend wobbly rockets doesn't mean I think it's perfect.) Secondly, yes, you can build pretty big rockets that launch pretty big loads with no mechjeb and a pretty minimal amount of control input. It takes design iteration but can be done. In fact the principles of designing a rocket that is stable and flies a smooth & efficient gravity turn to orbit are really not particularly dependent on scale. Bigger rockets just have much less wiggle room. OTOH if the game is about flying rockets to you, and the VAB is where you add boosters until you have 7km/s dV for your Duna mission, then no a wobbly or breakable rocket doesn't add much.
|
|
#
?
Mar 9, 2023 06:20
|
|
|
queeb posted:juno feels very mobile game like, but i guess thats because it is, since its just a renamed simple rockets 2, which some of you may already have in your steam library. Juno is not as good as KSP, but you can now launch 3d rockets on the toilet. The recently added career and tutorial system is actually very good at teaching the game. It's also dirt cheap on mobile.
|
|
#
?
Mar 9, 2023 08:17
|
|
|
I like the flight sim of KSP more than the spaceflight. Putting Jeb on top of a ramjet and seeing how fast he can go is fun, and what's really fun is when the engine explodes and I have to very quickly learn to handle a glider
|
|
#
?
Mar 9, 2023 08:37
|
|
|
swordfish duelist posted:Holy poo poo this is a mess. New thread title?
|
|
#
?
Mar 9, 2023 09:39
|
|
|
Vizuyos posted:Are there any actually interesting gameplay decisions that result from wobbly rockets? As far as I know, the only solutions are "use weaker rockets to launch smaller payloads" or "slap a bunch of fuckin struts all over everything". The latter is boring as gently caress, and the former kinda sucks. 
|
|
#
?
Mar 9, 2023 10:47
|
|
|
right-click, autostrut: heaviest part, add some more boosters, repeat
|
|
#
?
Mar 10, 2023 13:07
|
|
|
There's a new KSP2 Dev Insight on graphics: https://forum.kerbalspaceprogram.com/index.php?/topic/214806-developer-insights-18-graphics-of-early-access-ksp2/ TLDR: Most performance hits come from terrain graphics, so they're going to rebuild the terrain system. They are also switching to the High-Def Render Pipeline in Unity rather than Universal Renderer. I'm very early in learning Unity stuff, but... that's a pretty huge conversion, right? 
|
|
#
?
Mar 10, 2023 22:12
|
|
|
quote:I’m Mortoc, the new graphics programmer on the team.  What happened to the old guy...
|
|
#
?
Mar 10, 2023 22:23
|
|
|
Aertuun posted:
There was never a first graphics programmer, citizen. Mortoc has always been the sole graphics programmer
|
|
#
?
Mar 10, 2023 22:37
|
|
|
Alpenglow posted:They are also switching to the High-Def Render Pipeline in Unity rather than Universal Renderer. I'm very early in learning Unity stuff, but... that's a pretty huge conversion, right? It'd be way more of an issue if it was the other way around, since universal stuff can load into the HD pipeline. I think.
|
|
#
?
Mar 10, 2023 23:21
|
|
|
I hope it works for them but I've never heard of switching to the HDPR pipeline being used as a fix for a performance problem for a game with KSP's level of graphics. Usually HDPR is less performant but looks better than URP, not the other way around. HDPR can also be pretty tough to debug and comes with its own set of portability issues that can be tricky, but I don't work there and they may have some brought in some real sharp folks that can make it work better.
|
|
#
?
Mar 11, 2023 00:38
|
|
|
First KSP2 patch is due on March 16 unless something show-stopping comes up in QA. https://forum.kerbalspaceprogram.com/index.php?/topic/214809-approaching-patch-one/ Looks like they are improving resource flow logic, which has been a big cause of performance issues.
|
|
#
?
Mar 11, 2023 00:55
|
|
|
lmao the KSP reddit is losing its mind, but not about KSP2 or bugs. No, a 350 reply thread about communism; is it good or bad?  But about this question Klyith posted:a. How big is a big payload to you? For me, the baseline for heavy lift is launching a full Rocko 64 tank to LKO. It's my platonic ideal of a cargo mission: big enough to provide a good amount of fuel for several missions but not so huge it needs an elaborate oversize rocket. Back in the day before ISRU that was how I restocked my fuel depot, and I had a very reliable launcher (the Big Gulp) that was partially recoverable / reusable. So that's 36 tons plus a bit extra for docking ports, RCS, extra dV for orbit docking maneuvers, etc. I've done heavier launches of course, but not often.
|
|
#
?
Mar 12, 2023 06:08
|
|
|
Klyith posted:lmao the KSP reddit is losing its mind, but not about KSP2 or bugs. No, a 350 reply thread about communism; is it good or bad? Oh, I know this one. It's good
|
|
#
?
Mar 12, 2023 06:23
|
|
Klyith posted:But about this question For me it's close but a little bigger than that. For me it's a rocko 64 with a small engine and maneuver thrusters, it's own probe core, small fuel tank for its own engine, thrusters and monoprop maneuvering set. essentially a self contained mobile skiff that meets with a refuelling station on it's own and deorbits itself independent of the SSTO that gets it there. Call it 42 tons or so instead of 64. But close. For lower density craft it's what can fit in 2 rocko 64 length bays in an SSTO, with a total mass of around 50-64 tons, which is about everything I want to get into orbit, I pretty much launch everything in SSTO space planes when possible.
|
|
|
#
?
Mar 12, 2023 09:55
|
|
|
Klyith posted:lmao the KSP reddit is losing its mind, but not about KSP2 or bugs. No, a 350 reply thread about communism; is it good or bad? jesus christ https://old.reddit.com/r/KerbalSpaceProgram/comments/11ooc8j/kerbal_space_program_communist_flags_by_me_for_my/
|
|
#
?
Mar 12, 2023 21:47
|
|
|
marumaru posted:jesus christ God, I love Something Awful dot com.
|
|
#
?
Mar 13, 2023 05:30
|
|
|
USSR won the space race lol
|
|
#
?
Mar 13, 2023 06:43
|
|
|
The USSR had an early lead, so their newspapers usually reported on it as a space scoreboard. First satellite! First dog in orbit! First man in space! First woman in space! First probe to crash into the moon! So when the US outraced them to a manned moon mission, it was Soviets 5, Americans 1, even though it was clear the Americans had accomplished something they couldn't.
|
|
#
?
Mar 13, 2023 07:06
|
|
|
marumaru posted:jesus christ Wow people in there really drank the kool aid. People are trying really hard to get really specific about why they don't like communism while avoiding turning the mirror on themselves. "America never committed genocide on that scale." Like wow, ok, the mental gymnastics in that thread.
|
|
#
?
Mar 13, 2023 07:25
|
|
|
Chamale posted:The USSR had an early lead, so their newspapers usually reported on it as a space scoreboard. First satellite! First dog in orbit! First man in space! First woman in space! First probe to crash into the moon! So when the US outraced them to a manned moon mission, it was Soviets 5, Americans 1, even though it was clear the Americans had accomplished something they couldn't. The USSR really is special in simultaneously attracting just the worst people as supporters and detractors.
|
|
#
?
Mar 13, 2023 09:40
|
|
|
The subreddit has some amazing content too.
|
|
#
?
Mar 13, 2023 19:41
|
|
|
on average the KSP reddit is really good, though lately all the KSP2 drama has made people significantly more crappy and negative the communism thing is just some classic internet though
|
|
#
?
Mar 13, 2023 19:56
|
|
|
it's also funny to see people there realize that the KSP1 volumetric clouds mod's clouds look better than the $50 KSP2's
|
|
#
?
Mar 13, 2023 21:24
|
|
|
That's not the normal clouds us plebs can just download. You gotta be a Patreon subscriber for those clouds, and they cost $5
|
|
#
?
Mar 13, 2023 21:29
|
|
|
Otacon posted:That's not the normal clouds us plebs can just download. You gotta be a Patreon subscriber for those clouds, and they cost $5 not sure who made more money from the KSP 2 release, Take2 or Blackrack
|
|
#
?
Mar 13, 2023 21:42
|
|
|
I got whiplash jumping from "KSP Mods should be free" to "SHUT UP AND TAKE MY MONEY" the first time I saw those clouds.
|
|
#
?
Mar 13, 2023 21:45
|
|
|
Otacon posted:That's not the normal clouds us plebs can just download. You gotta be a Patreon subscriber for those clouds, and they cost $5 Although it's supposed to be a patreon exclusive only during "early access", once the mod is finished it'll be released for free.
|
|
#
?
Mar 13, 2023 21:46
|
|
|
Klyith posted:not sure who made more money from the KSP 2 release, Take2 or Blackrack edit:  https://www.youtube.com/watch?v=YrlE4sckhMo&t=7s Hadlock fucked around with this message at 21:56 on Mar 13, 2023 |
|
#
?
Mar 13, 2023 21:54
|
|
|
Wow, that's pretty amazing work. I might actually pay for that horse armor.
|
|
#
?
Mar 13, 2023 22:06
|
|
|
Hadlock posted:edit: try this one on for size https://old.reddit.com/r/KerbalSpaceProgram/comments/11pjpdv/blackracks_volumetric_clouds_mod_goes_insane/
|
|
#
?
Mar 13, 2023 22:48
|
|
|
Otacon posted:That's not the normal clouds us plebs can just download. You gotta be a Patreon subscriber for those clouds, and they cost $5 it'll be free Soon(tm), but it's not like it has DRM. it's definitely worth the  /2 though imo
|
|
#
?
Mar 13, 2023 23:12
|
|
|
https://forum.kerbalspaceprogram.com/index.php?/topic/215095-patch-one-is-go/ First patch is live. Direct link to patch notes here: https://forum.kerbalspaceprogram.com/index.php?/topic/215108-ksp2-patch-notes-v0110/ Luneshot fucked around with this message at 18:13 on Mar 16, 2023 |
|
#
?
Mar 16, 2023 18:03
|
|
|
|
| # ? May 24, 2024 15:19 |
|
|
Fairly promising first pass for fixes, hopefully this alleviates a lot of the doom and gloom
|
|
#
?
Mar 16, 2023 18:59
|
|
