steve_v

So I see. But you didn't list versions, so it was only by chance that I stumbled on that updated thread. Again, you didn't list versions, so there was no way for me to know that. Here we are wasting time playing 20 questions. Nevertheless, that's what is required for a bug-free modded KSP experience. Perhaps you should consider using a mod manager? If you can't keep track of which versions your mods are compatible with, how do you expect anyone else to when they don't even know what versions you are running in the first place? I just gave you two easy ways of providing that information... Open your log file and search for the word "NullReferenceException", then look at the "at <foo>" lines following it, where <foo> is the name of a mod method or DLL filename. It's usually pretty obvious what it belongs to. In your case I see exceptions from OrbitalDecay, then KerbalEngineer, then a vessel getting NaN orbit parameters, and then a bunch more exceptions from stock KSP methods. One of those first two is quite possibly causing all the others, and the presence of MiniAVC is very possibly causing all of them. If there's nothing obvious and removing the mods I mentioned don't fix it, then you'll just have to do it the old-fashioned way , as we've all had to at some point: Remove a few mods at a time until you figure out what's causing it.

Looking good so far. Eh, It happens. This is why we test things.

There are a great many exceptions in your log, and it's not immediately clear where this disaster starts. First off, you still have at least one copy of MiniAVC.dll installed. It's broken and has been since 1.8.0. Kill it. Other than that obvious point, the first significant NRE appears to be: NullReferenceException: Object reference not set to an instance of an object at Planetarium.GetUniversalTime () [0x00027] in <c1858a3f77504bd1aaa946fdccf84670>:0 at WhitecatIndustries.SCSManager.OnDestroy () [0x0000b] in <90f6673545a1415893eaefb8b6f9851c>:0 UnityEngine.DebugLogHandler:Internal_LogException(Exception, Object) UnityEngine.DebugLogHandler:LogException(Exception, Object) ModuleManager.UnityLogHandle.InterceptLogHandler:LogException(Exception, Object) UnityEngine.Logger:LogException(Exception, Object) UnityEngine.Debug:CallOverridenDebugHandler(Exception, Object) WhitecatIndustries is a reference to OrbitalDecay, AFAICT. I also note you have Kopernicus installed, and AFAIK that's incompatible with 1.10.1 unless you're running RTB's "bleeding edge" fork. Since you haven't listed any mod versions, who knows. Please help yourself, by providing all the information listed here. Posting the information required to troubleshoot and following the self-help advice (especially WRT mod compatibility) will achieve considerably more than repeatedly bumping your thread. You also "forgot" to list mod versions. To quote the "how to get support" sticky linked above, a lazy screenshot is not useful: Note that a copy of KSP.log now largely fulfils this requirement, as it contains a list of directories under GameData and assemblies loaded (with versions). Alternatively, CKAN can export a list of installed mods with versions.

Different, but, uhh, arguably not fixed. Again exiting R&D, constant stream of: [LOG 02:32:33.506] [KerbalHealth] KerbalHealthScenario.OnSave [ERR 02:32:33.515] Exception handling event onGUIRnDComplexDespawn in class RDSceneSpawner:System.InvalidCastException: Specified cast is not valid. at KerbalHealth.KerbalHealthStatus.get_ExhaustionStart () [0x0003d] in <5e040da14d844cf7904e8ad2c94d5cc8>:0 at KerbalHealth.KerbalHealthStatus.get_ExhaustionStartHP () [0x00000] in <5e040da14d844cf7904e8ad2c94d5cc8>:0 at KerbalHealth.KerbalHealthStatus.Update (System.Double interval) [0x00354] in <5e040da14d844cf7904e8ad2c94d5cc8>:0 at KerbalHealth.KerbalHealthList.Update (System.Double interval) [0x0001b] in <5e040da14d844cf7904e8ad2c94d5cc8>:0 at KerbalHealth.KerbalHealthScenario.UpdateKerbals (System.Boolean forced) [0x003ca] in <5e040da14d844cf7904e8ad2c94d5cc8>:0 at KerbalHealth.KerbalHealthScenario.OnSave (ConfigNode node) [0x0001f] in <5e040da14d844cf7904e8ad2c94d5cc8>:0 at ScenarioModule.Save (ConfigNode node) [0x000a6] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at ProtoScenarioModule..ctor (ScenarioModule module) [0x0003c] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at ScenarioRunner.UpdateModules () [0x00055] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at ScenarioRunner.GetUpdatedProtoModules () [0x00005] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at Game.Updated (GameScenes startSceneOverride) [0x000a7] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at GamePersistence.SaveGame (System.String saveFileName, System.String saveFolder, SaveMode saveMode, GameScenes startScene) [0x00045] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at GamePersistence.SaveGame (System.String saveFileName, System.String saveFolder, SaveMode saveMode) [0x00000] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at KSP.UI.Screens.RDSceneSpawner.onRDDespawn () [0x0000a] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at EventVoid.Fire () [0x00127] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 Booting clean install and double-checking md5sums of DLLs now... Ed. Yep, def had the right DLL. Not seeing this in a fresh save though, 1.4.5 release is fine too. This seems to be specific to my main save (because it has KH 1.4.3/1.4.4 data?). Further testing will have to wait unfortunately, it's 3am here. Tomorrow I shall eviscerate my save of all KH data and see what happens...

I just tried to actually, properly play the game, for the first time this year. And of course, I tried to land a lightweight rover on Duna. Because that's a logical thing to do right? Anyone care to guess how that "roving" went? How the [REDACTED] is this crap still not fixed? How is using the debug menu in normal gameplay an accepted solution? Wheels have been broken since 1.1. Four [REDACTED] years. Four. Glowy suits. Surface experiments. Comets. 2 DLCs. And something as fundamental as being able to drive on a surface is still broken beyond all recognition. For anyone stumbling on this, yes. It's 2020. The game is at version 1.10.1. Wheels are still a complete disaster. I give up, the bugs are not getting fixed, the developers don't give a damn, and this game is a still a smouldering garbage fire of half-baked ideas, bugs and jank.

Absolutely. Personally I'd prefer they added nothing at all, and just went for releasing a game that us actually finished rather than this continuous development perpetual beta nonsense. 1400 open bugs, many of them new, and bug count showing no sign of decreasing this far into development and this close to the release of the sequel is, frankly, ridiculous. That said, if they are going to add stuff to the game, the proposed graphical improvements aren't silly candidates. Who knows, Squad might even manage to add clouds without breaking something else... Hmm, perhaps we are. May have gotten a little distracted by the raging argument over in the KSP2 potato-compatibility thread. My bad.

Update installed in existing game, now unable to exit R&D building. [ERR 22:15:31.156] Exception handling event onGUIRnDComplexDespawn in class RDSceneSpawner:System.InvalidCastException: Specified cast is not valid. at KerbalHealth.KerbalHealthStatus.get_MaxHP () [0x00069] in <9ed5380de297434582116086f445d991>:0 at KerbalHealth.KerbalHealthStatus.get_MarginalChange () [0x00000] in <9ed5380de297434582116086f445d991>:0 at KerbalHealth.KerbalHealthStatus.HealthChangePerDay () [0x004e5] in <9ed5380de297434582116086f445d991>:0 at KerbalHealth.KerbalHealthStatus.Update (System.Double interval) [0x00165] in <9ed5380de297434582116086f445d991>:0 at KerbalHealth.KerbalHealthList.Update (System.Double interval) [0x0001b] in <9ed5380de297434582116086f445d991>:0 at KerbalHealth.KerbalHealthScenario.UpdateKerbals (System.Boolean forced) [0x003ca] in <9ed5380de297434582116086f445d991>:0 at KerbalHealth.KerbalHealthScenario.OnSave (ConfigNode node) [0x0001f] in <9ed5380de297434582116086f445d991>:0 at ScenarioModule.Save (ConfigNode node) [0x000a6] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at ProtoScenarioModule..ctor (ScenarioModule module) [0x0003c] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at ScenarioRunner.UpdateModules () [0x00055] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at ScenarioRunner.GetUpdatedProtoModules () [0x00005] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at Game.Updated (GameScenes startSceneOverride) [0x000a7] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at GamePersistence.SaveGame (System.String saveFileName, System.String saveFolder, SaveMode saveMode, GameScenes startScene) [0x00045] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at GamePersistence.SaveGame (System.String saveFileName, System.String saveFolder, SaveMode saveMode) [0x00000] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at KSP.UI.Screens.RDSceneSpawner.onRDDespawn () [0x0000a] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 at EventVoid.Fire () [0x00127] in <948fceea813942b7ac6b6d1b2dc2d0a3>:0 [EXC 22:15:31.172] InvalidCastException: Specified cast is not valid. KerbalHealth.KerbalHealthStatus.get_MaxHP () (at <9ed5380de297434582116086f445d991>:0) KerbalHealth.KerbalHealthStatus.get_MarginalChange () (at <9ed5380de297434582116086f445d991>:0) KerbalHealth.KerbalHealthStatus.HealthChangePerDay () (at <9ed5380de297434582116086f445d991>:0) KerbalHealth.KerbalHealthStatus.Update (System.Double interval) (at <9ed5380de297434582116086f445d991>:0) KerbalHealth.KerbalHealthList.Update (System.Double interval) (at <9ed5380de297434582116086f445d991>:0) KerbalHealth.KerbalHealthScenario.UpdateKerbals (System.Boolean forced) (at <9ed5380de297434582116086f445d991>:0) Full log.

* Consoles. * People who don't, won't or can't install mods. * Designing things like this into the game will likely yield better performance. * People have been asking for stock clouds since alpha. "There's a mod for that" has never stopped Squad adding things to the game, I don't see why it would stop Intercept/T2 either. I'm not going to argue further, but I will leave you with a question to ponder: How many sequels have you seen over the years that don't improve on the graphics of the original, and given this one will likely be released 8 years after the original, do you really expect no increase in system requirements?

What the community wants will likely be irrelevant. Shiny graphics sell games.

Unpack the archive, read the readme therein. Probably 7zip, strangely enough, since it's a 7zip archive.

[D3D error was 887a0005] I don't run Windows any more (and haven't for years), but apparently error code 887a0005 is "DXGI_ERROR_DEVICE_REMOVED". which would point to a GPU or GPU driver crash. Make sure you don't have any apps running that mess with the graphics stack (overlays, shader injectors etc.), reinstall GPU drivers, check power and cooling, all the usual stuff. Other than that obvious advice, you'll probably have to wait for someone who actually runs Windows to comment. I don't really read stack traces from that OS, nor do I have a Windows machine to test on.

Ads? I didn't see no ads... Wait, there are people who still surf without an adblocker? True, though there you get the whole sign up with a "real" name to upload thing, and sign-in/sign-up nags and click-throughs on download. MediaFire with an adblocker is actually fewer clicks and less javascript than Dropbox... Personally I got fed up with all of the above some time ago, I host my own files on my own hardware now.

I'm not sure that this evil is truly the problem... But you'd best kill it with fire anyway.

You went and picked out the very cheapest power supply you could find with a brand someone might recognise, didn't you? Drumroll please.... Nope, it's the second cheapest 500W PSU from a mostly non-explodey brand on PCPartPicker. I'm pleasantly surprised. It's definitely much, much better than either of the mystery-meat modules mentioned previously. Congratulations, you've moved up from "potentially dangerous" to just garden variety dirt-cheap. You will of course get what you pay for, but at least it has a decent warranty. A very decent warranty for the price actually. Thermaltake does have a care to not drag their brand through the mud with complete garbage it seems. As with all cheap PSUs, information is scarce since nobody can be bothered to review them. It's probably a CWT design, kinda old, almost certainly group regulated, and not very efficient. Other than that there's not much to say, cheap PSU is cheap. CWT make the guts for many brands, with quality ranging from "probably not a fire hazard" all the way up to "pretty dang good". Take a look at the price tag, and you can easily guess at which end of the spectrum a particular model falls. If you're going to spring at least a couple hundred for a 1660 Super, then powering it with a sub-$50 PSU seems like a strange choice to me. You'll likely still be using the PSU when the GPU is but a distant and horribly obsolete memory. But you do you, within reason. If $50 is your price bracket, this isn't a silly buy. Personally, If I was looking for a reasonably priced 500W PSU then I'd probably drop ~$70 on the Antec NeoEco Zen, mainly for the 80+ gold efficiency, DC-DC converter, better protection, Japanese main caps, and gruntier 12v rail. But that's just me, and I kinda like Antec anyway. vOv.

Best buy house-brand, unknown OEM, no real reviews online... One year limited warranty. Store page specs don't match manual. No warranty mentioned on store page. Cooling fan type: "Other"... Overload protection: "No"... Uh Oh. A 2017? model (manual copyright date) with a floppy drive connector? So cost-down they don't even give you mounting screws in the box. Did I mention the One year limited warranty? Well, it's up to you I guess. If that's what you're gonna use that's what you're gonna use. I wouldn't touch it with a 40' pole, 80' if used. Actually I'd probably just pay someone to get it out of my sight. If you really must go ultra-cheap, for 5 bob more (new prices) you get a Corsair VS450. I mean it's literally the worst PSU Corsair make, only rated for full power at 30c, and I hesitate to recommend it to anyone, but still. A Corsair, with a 3 year warranty. Why roll the dice on some unknown brand for the sake of $5? Seriously, this thing looks a even more suspect than the Apex unit you started with. A lot more. We're going backwards. As for those "pretty good reviews", perhaps we should take a look at the bad ones (emphasis mine)? Jebus. RUN. If you must use it, I have one small request: When it blows up, open it and get some pics. I'm curious as to what is actually in there besides all the air visible in the promo shots...

T2 can't cover anyone or anything if their claim to the material is in itself a breach of copyright. If the mod still contains that old multiple-contributor GPL code for example (which it almost certainly does), then the onus is on T2 to prove that all the contributors have signed over their rights, and until they do that their claim to the mod as a whole is void. SpannerMonkey may well have updated things more recently, but regardless of what the EULA claims, he almost certainly cannot have signed over the rights to T2 because he didn't own all the rights to begin with. Cool, awesome. Also liable to spontaneously evaporate if ever competently challenged in a court of law. Has T2 proven beyond reasonable doubt that all the authors, past and present, have used the game or created content based on it since the EULA change? Have they determined that those updating the mod actually can hand over rights to it? Until they do, it's all lies and posturing. No matter how much legalese you imply someone must have agreed to, you cannot make them hand over what they never had, nor can you force them to travel backwards in time to get it for you. Many, many KSP mods incorporate code from multiple contributors, and/or code from generic C#/Unity resources completely unrelated to KSP. A good number contain assets whose original authors are unreachable or even unknown. T2's claim that all rights somehow magically transfer to them because they say so, without bothering to examine the original licence or contact the authors involved, is pure fantasy. Anyway, it's clear we're not going to come to any kind of agreement here. If you want to take that ridiculous, illogical, unenforceable monstrosity of an EULA unchallenged and at face-value, I'm not about to stop you.

Given we're talking about a mod released well before the EULA change, and the conspicuous lack of proof that the author(s) agreed to or even saw said EULA, I'm not buying even "slightly lawful". T2 granting "permission" regarding material they have no claim to or control over means exactly nothing at all, and if they can't prove the author handed over their rights, that's exactly what they're doing.

In the specific case of SM Armory, that's not entirely straightforward to determine as most of the github repo is gone. The oldest component (common library) remaining is GPL licenced and appears to have been since July 2017. With this library being GPL, it's probably safe to assume that the rest of the mod in question was as well, at least as far as plugins go. And it's using code from other GPL projects... Aren't viral licences awesome? I wish T2 all the luck in the world trying to retroactively appropriate GPL code, they're going to need it. As for the assets, at least some of them are ARR, specifically the legacy boomsticks parts. That makes redistribution pretty darn fraught whichever way you look at it, and unless T2 can prove that the original author agreed to the new EULA, they can't have those bits either. Ed. Actually, all models in the pack appear to be ARR: ^ Indeed, thanks to T2's bogus and overreaching EULA boilerplate, the entire modding scene is in a properly confusing position. There are many mods out there that simply can not legally be re-licenced, whatever T2 might want. Barring any reliable legal advice on the matter (I don't pretend to be a lawyer, though there is one in the family), it all really comes down to whether an EULA can override pre-existing licences, or indeed lay legitimate claim to independently created content that takes no assets or code from the game to begin with, and whether anyone can prove who did or did not agree to the new EULA... That and whether T2 is willing to risk the almighty ****storm that will undoubtedly ensue if they try. Personally I doubt T2's claims would stand up in court, but without precedent there's probably no way to be sure. I also highly doubt T2 would ever try to enforce their EULA claims of mod ownership, as doing so would obliterate the entire modding scene overnight. There is no "forum licence", there is only the licence the material is distributed under and the forum requirement to state it if you provide download links here. Legal forks and repacks are often seen as disrespectful around here, but that has nothing whatsoever to do with licencing. Unless those doing the redistributing can produce the original licence (as found in source code headers, github repos and release archives) and show they're abiding by it, the question of respect is completely overshadowed by the matter of legality. If you want to get into semantics regarding the word "stolen" that's your call, but I'm not going to play. Someone redistributing a work in contravention of licence terms is still copyright infringement regardless of the language used... And yes, people do get emotional when they see someone passing off another's work as their own. They should probably get equally emotional about T2's ham-fisted attempts to do the same.

The immediate upshot of which being: Mono/JIT is ludicrously slow pretty much all the time, and old-school C compiled with GCC still beats the pants off needlessly verbose C#, even when the latter is restricted to a subset and compiled with the latest shiny toys like Burst/IL2CPP.

The reason physics is single threaded is that in a connected-rigidbody system the calculations depend on each other, so they kinda need to be processed sequentially. If two vessels aren't connected, they can have physics on separate threads since the only real interaction would be collision checks. A KAS rope in "undocked" mode counts as separate vessels. A claw or docking port makes it one vessel, so one physics system and one thread. Any undocked separate vessels will pass through each other under timewarp though, so you're kinda screwed. The best option really is just keeping part-count down. Windows likes to shuffle threads around, and the meters you see are averaged. It's not a physics system that lends itself to massively parallel computation, which is what GPUs are good at. GPU physics is great when you have lots of independent objects, but to calculate the motion of one link in a chain you first need to calculate all the others. KSP vessels are essentially chains of physics objects, so CPU it is. I'd elaborate further, but a few minutes of this forum on mobile is about all I can stand. The editor is abysmal.

Indeed it does. The "tolerably well" bit was because we really shouldn't need a mod like this in the first place.

A great many mods have licences that long predate the introducton of that EULA. Take Two asserting that they retroactively own any and all material related to KSP doesn't change that in the slightest. Many mods have, and have had for many years, licences that explicitly forbid such unilateral transfer of copyright. Copyright law will trump an EULA any day of the week in any court you like.

Yeah... Thanks for that. I thought something rotten was going on with stylish, confirmation is good. Guess it's safe to reinstall your theme again now, it's been a while.

With mods, or not at all. You're playing Kerbal Jank Program. Maja's Parking Brake works around it tolerably well. Lisias has found and mostly fixed one of the causes, and is working on the other. Squad has offered a large helping of hype for new versions, but that's about it.