Lisias

I have kinda a grudge with 1.11. I had detected and fixed a pretty annoying problem on 1.8 <= KSP <= 1.10.1 where Kerbals and Crafts would drift when parked no matter what. I nailed the problem to be spurious torques being applied to the parts randomly, and cooked a workaround. Problem: on KSP 1.11.0, Squad "fixed" the problem in a way that drifting was still an issue and my workaround was rendered useless so, frankly, I kinda gave up. Debugging, diagnosing and fixing things is a very laborious and task consuming task. So sticking with 1.7.3 ended up being the best cost/benefit to me even nowadays, falling back to 1.4.3 for quick stunts (as this thing loads waaay faster on my rig, and it's also one of the most performatic KSP - second only to 1.3.1 and 1.2.2!) Uh… Yeah, you are right! Anyone enjoying detective games and liking XVIII Century sea tales, Return of the Obra Dinn is a must. Seriously, one of the best games of that genre I ever played. https://www.gog.com/en/game/return_of_the_obra_dinn

Pinpoint this thread. The KSP.log you published may help them to pinpoint the problem. Most of the time us a pretty silly detail we, developers, let pass trough. -- post edit -- Check this post: Do you have DART installed? (On mobile right now, can't see myself). If yes, so there's no need to bother Parallax guys : remove DART and reinstall Parallax and things apparently will be fine!

Appears to be Parallax. [LOG 20:17:28.050] [OD] Enabling Body Kerbin: True [EXC 20:17:28.307] KeyNotFoundException: The given key was not present in the dictionary. System.Collections.Generic.Dictionary`2[TKey,TValue].get_Item (TKey key) (at <9577ac7a62ef43179789031239ba8798>:0) ParallaxOptimized.SubdivisionData.DetermineMaterial () (at <40e34d3c09734e828597afb2f90e2e92>:0) ParallaxOptimized.SubdivisionData..ctor (PQ quad, System.Int32 subdivisionLevel, System.Single subdivisionRadius, System.Boolean subdividable) (at <40e34d3c09734e828597afb2f90e2e92>:0) ParallaxOptimized.PQSMod_Subdivide.OnQuadBuilt (PQ quad) (at <40e34d3c09734e828597afb2f90e2e92>:0) PQS.Mod_OnQuadBuilt (PQ quad) (at <4b449f2841f84227adfaad3149c8fdba>:0) PQS.BuildQuad (PQ quad) (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.Build () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.SetVisible () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.UpdateVisibility () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.UpdateSubdivision () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.UpdateSubdivision () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.UpdateSubdivision () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.UpdateSubdivision () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQ.UpdateSubdivision () (at <4b449f2841f84227adfaad3149c8fdba>:0) PQS.UpdateQuadsInit () (at <4b449f2841f84227adfaad3149c8fdba>:0) (wrapper dynamic-method) PQS.PQS.StartSphere_Patch0(PQS,bool) PQS.ForceStart () (at <4b449f2841f84227adfaad3149c8fdba>:0) PSystemSetup.ForceInitPQS (PQS pqsIn, UnityEngine.Transform target) (at <4b449f2841f84227adfaad3149c8fdba>:0) PSystemSetup+<SetupLaunchSites>d__38.MoveNext () (at <4b449f2841f84227adfaad3149c8fdba>:0) UnityEngine.SetupCoroutine.InvokeMoveNext (System.Collections.IEnumerator enumerator, System.IntPtr returnValueAdd ress) (at <12e76cd50cc64cf19e759e981cb725af>:0) UnityEngine.DebugLogHandler:LogException(Exception, Object) ModuleManager.UnityLogHandle.InterceptLogHandler:LogException(Exception, Object) UnityEngine.Debug:CallOverridenDebugHandler(Exception, Object) [ERR 20:17:28.453] PQS Kerbin: Restarted Uninstall it to see what happens. If it works, reach Parallax maintainers for help. If it still borks, post a new KSP.log here and mention me and I will check it again! Good Luck!

Elon Musk's SpaceX hires 14-year-old whizz kid as a software engineer at Starlink https://www.thenationalnews.com/business/technology/2023/06/16/elon-musks-spacex-hires-14-year-old-whizz-kid-as-a-software-engineer-at-starlink/ Exactly what one needs to do to get hired as "Engineer" nowadays? (the kid graduated, at least. but… still…)

(late) @AlamoVampire I did i!

Dude... someone was drinking on the job... not enough the weird appearance, the fuel tank was placed between the pilot (over the engine) and the gunner. Ok, it was heavily shielded but yet... Ilyushon IL-20 (1948) https://forum.warthunder.com/index.php?/topic/514891-ilyushin-il-201948/

If this is a bug on Toolbar, yes. If this is a bug on something that Toolbar relies, it may affect something else with not so benign effects. Normalisation of deviance is a problem by itself. It worths at least a superficial investigation to figure out the possible causes. So it's something on code and, so, should be affecting not only Windows. Makes things easier for me, to tell you the true. Let's see what I manage to pull from this hat. Really? So the log is lying to me?

Are you talking about this one? [ERR 06:52:29.853] [Toolbar] [ERROR] button texture not found: KerbalLaunchFailure/PluginData/Textures/warning_24 The file is inside a PluginData, so KSP is not mangling with it for sure. Check if the file <KSP_ROOT>/GameData/KerbalLaunchFailure/PluginData/Textures/warning_24.<something> exists. Check the casing of the pathname, check the casing of the filetype (JPG instead of jpg, for example, or even Jpg). You are running Windows 10, that usually uses a case insensitive file system - but NTFS can be configured to be case sensitive… But it's a wild guess. What is your Windows Localization? Unity has a pretty nasty habit of replacing chars with dots, a living hell on PT-BR, as we have a lot of non-ascii chars on our alphabet and I bet this can be even nastier to non roman alphabets as Greek, Russian and Hebrew. I once got royally screwed because I created a new savegame called Temp without realizing I already a temp one on the file system (I use a case sensitive file system). Dude, files were being written on the wrong savegame, I don't need to tell more… I checked your log, and you are not using any fancy chars on the KSP's pathname, however. But it doesn't hurts to check.

Only "amateurs" get their hands dirty. Professional ones seed the discord to be spread by ingenuous people. A Person is inteliggent, People are dumb! But, and again, pure expeculation. I have no evidence of such practices - but, hey, such practices leave no evidences!

You are forgetting a significantly relevant factor: recently, TT2 fired a lot of people from P.D. and some subsidiaries, most of them management. These people are not happy, and you can bet your mouse that some of them have a grudge about the ordeal and acquired a personal (and professional) interest on the failure of this project. The gaming industry is not exactly above F.U.D., Astroturfing and similar scumbag's practices. I find some criticisms around here way harsher from what it would be reasonable even from disgruntled and vindictive users. Sometimes I wonder if there would be not some big Company interested on devaluating P.D. or even TT2 in order to try an hostile takeover, as Microsoft did on Nokia years ago (where is Stephen Elop nowadays?). But… This is all speculation from my part.

It will work, but I would not recommend doing this way as you would be hardcoding TweakScale into your parts, annoying people that don't want to use it. I would not recommend this solution neither, unless you are not keen on mixing your parts with Stock ones. TweakScale Support is meant to cope with the Stocks (and some common used by 3rd parties) BulkeHead Profiles. So it's way easier to cope with Stock and most Add'Ons out there if you stick with the scaleFactoers TweakScale gives to you. If you really wants to go Percentage way, just use free and call it a day. There's no point on giving to the user Scale Factors that don't match any other part of the game, forcing the user to just turn "Automatic Scaling" on Editor and hoping for the best. There's already https://github.com/TweakScale/Companion_NAP supporting NeistAir, and since this thing is part of the TweakScale Companion Program, anyone installing it already have such support. It's available on SpaceDock and CKAN under the TweakScale Companion Program: https://spacedock.info/mod/3202/TweakScale Companion (the "UberPaket")

Because History, unfortunately, doesn't rewards Planning, but Results. Boeing completely screwed up Starliner, besides having years of planning and "careful" execution (and a hell of a budget). SpaceX hacked and slashed their way into the problem with Dragon and nailed it, with a incredibly smaller budget and timeframe. SLS finally took over, but after decades and billions and billions of over budget. And they are still years behind schedule. It's possible to go Full Cascade and still get good results, but historically this development model failed more than succeeded because… Humans. People lie, people hide, people pretend - you only need one bad apple on a basket of good apples to lose them all. (and I'm kindly ignoring that people made honest mistakes all the time, some of them ruining the planning and scheduling - but, yet, most Companies just fail on admitting failure early on the Cascade's process, what would allow them to rework all the Planning and preventing getting stuck in development hell later, when it's finally impossible to hide that fatal failure).

Geez! You were bitten by an insidious MM bug that I'm hunting for years already! And I let it scape under my nose! Damn! There's a least one situation where MM fools itself on believing the ConfigCache is up to date. I detected one and fixed it on my fork, but recently I published a bad release of KSP-Recall that I initially attributed to a new mishap on the Cache validation, but could not found it by eye balling the MM's code so I reluctantly ended up attributing to sleep deprivation. You just proved to me that there's still one more bug to go! Can you please send me a copy of every log in your KSP rig? ConfigCache, MM Pacthes, etc. Essentially, everything under <KSP_ROOT>/Logs and <KSP_ROOT>/PluginData. And the Player.log too, for good measure. With this info, I can try to look for hints about what would be this missing use case on the MM' cache validation since right now your rig has whatever is the situation that triggers it! Thanks in advance! In time: https://github.com/net-lisias-ksp/ModuleManager/issues/20

Yes. Please make a list of them and I will update this issue: https://github.com/TweakScale/Companion/issues/19 I'm planning to have a window of opportunity to tackle down a lot of Companion tasks next month, and I will carry one benjee's ones too.

Did you updated something recently? I found a huge amount of KSPCF messages complaining about the order of the Modules on the craft file not matching the order on the prefab and doing changes. For this to happen, something changed on your GameData between the last time you played this savegame and now. I would suggest to revert any changes to see if the problem goes away. (Do full backups of your saves anyway, so if things go really south, you can try again without losing them!)

Yes. While doing support for my add'ons, more than once the user sent me their whole savegame so I could reproduce a insidious bug on my rig. I use MacOS, and I was able to run savegames from both Windows and Linux. As long you install all the add'ons you are using on the rig, you will be able to use the savegames in any platform KSP runs on. Happy salvaging!

I'm playing a game with way less bugs and problems than KSP 1.12 : I'm playing KSP 1.7.3. (hey, it's an improvement! I was playing 1.4.3 last week!!!)

Supply chain attacks are happening everywhere - if you think CF got a bad hit, you don't know what happened to Python's Pypi repository recently - not to mention Google Play, an eternal problem still without a reasonable resolution. SpaceDock had the same problem, by the way - these guys removed a lot of entries recently, one of them with more than a year being hosted there (granted, it was not malware, it was advertising. I know because I was the one who detected it and reported it). Digging around, I found that even the github accounts linked to that add'ons were removed!!! CKAN is so vulnerable to this problem as anyone else - all it takes to something bad to happen is a bait and switch: a nice add'on being indexed for some time, and then suddenly a new release with a custom coded malware embedded and things go south. Not trying to sweep CF's issues under the rug, but I consider naive - or plain dangerous - thinking that CKAN is immune to this problem. They are not. I don't blindly trust even Github, I have a branch where my binaries are committed using my digital signature for reference. If something bad ever happens, we have a digitally signed commit with the original binary for comparison and inspection. I think SpaceDock (and CKAN) should consider asking Add'Ons to be digitally signed - this would prevent Supply Chain attacks by 3rd adversarial parties, and if an author misbehaves you can flag as suspicious anything signed with that key [instead of banning the whole account due a single misbehaviour].

(sigh). Ok, I will do all the work for you: https://docs.curseforge.com/#what-is-eternal I will say it again: If CKAN are not being allowed to index KSP add'ons on CurseForge is because authors choose not to allow it, this is not on CurseForge's shoulders. I'm one of the authors that choose not to allow it - everything I do is already published on github and some on SpaceDock, I see no reason to use CurseForge's infrastructure (and money) if there're alternatives to do the same job. — — POST EDIT — — (sigh²) From my original link, I presented you with the game and dependencies API. The Upload API were depicted after that first two. Following the links on the right of the page, one can reach this page: https://support.curseforge.com/en/support/solutions/articles/9000207877-project-distribution-toggle Where the gore details are available:

I beg your pardon? https://support.curseforge.com/en/support/solutions/articles/9000197321-curseforge-upload-api#:~:text=The CurseForge API allows users,versions and dependencies from CurseForge. In a way or another, it's up to the author to decide to allow direct downloads from CurseForge or not, it's not a CurseForge decision. Since my add'ons are published on github and SpaceDock, I choose to disallow direct downloads on CurseForge for them. It was my decision, not CurseForge's. CurseForge had its problems, as everybody else, but since it as bought from Twitch things are improving. Exactly my point (emphasis are mine): That's conditional on someone else doing something for you and also not what I was asking (avoiding breaking things versus simple testing before release), so I will take that as a "No." If future KSP-Recall releases will not be tested with Procedural Parts, then maybe a conflict is called for here after all. Source. Exactly what I did.

CurseForge is more friendly to Authors, as they have a business model and can afford to properly pay for moderation and curating. But you need to cope with ads both on site and Installer - someone need to pay for the party. CKAN has no ads, and I don't have the slightest idea how they pay their bills. They are clearly overwhelmed, as they can't guarantee any level of stability once a update is made when many add'ons are updated at once. The current policy is insufficient and tends to overburden responsible Authors with support caused by misbehaving 3rd parties - I once had to agree on doing free QAS for someone else to prevent bugs on THEIR add'on from breaking mine in order to prevent a precipitated and unfair Incompatibility flag on mine. Their technical advising is not rarely questionable, and sometimes deleterious- found some pull requests from them on 2rd parties that created worst potential problems than the one they aiming to fix - but, hey, this one is in the maintainers shoulders that merged the pull requests. On their behalf, they cared to try and this can be a good thing when both sides are willing to learn from their mistakes. To be fair, CurseForge don't even try to do any of that - but they don't pretend it neither. My best guess is that CKAN is terribly understaffed, and so the few available hands need to cope with technical issues they are not familiar with. CKAN has also a still long list of very, very annoying bugs that they are, apparently, struggling to tackle down. Authors have a way better treatment by CurseFurge, by the way. Not directly related to users, but it's still a factor. If you are using MacOS with 32 bits applications support, CKAN works on GUI. Otherwise you will need to run it in command line or in a Text User Interface at best (there's a command-line option for it). Telling you the true, it's not bad - I even prefer it sometimes. Curse Installer is the opposite, it only works on recent MacOS - Mojave is not supported (and I'm locked on it).

Users! Makes no sense on investing time (and modding is a huge time investment!!!) on a thingy with less than 100 users while I get way more results on supporting my add'ons on a game with 1260 users at the same time. Also note that even at the best times, (Sunday), the proportion is still abysmal: So it makes absolutely no sense on modding a Game that: Have almost no users online, sometimes literally [edit: the zero points are glitches on the Steam Charts, KSP1 is getting some now (2023-0623)] It doesn't runs on any of my machines with a minimally acceptable performance And I'm not willing to commit resources on a new rig right now. TL;DR: I can't even run the thing. Have no published API So I need to Reverse Engineer the thing myself And this is not exactly allowed under some Legislations Not to mention the huge additional time investment on it There's absolutely no Demo Projects published by the Developes, so Reverse Engineering is really the only way to go And, adding offense to the insult, I need to learn how to build a Unity Game in order to be able do deliver something - no kidding, I need to learn the whole pipeline. On KSP, all I need is Notepad, Paintbrush and sometimes Visual Studio CE to compile a DLL It just doesn't worth it at this point. — — POST EDIT — — This was also discussed on the thread below:

Hi! What happened is that you uninstalled the messenger, so the problem ceased to be reported. But it's still there. Install KSP-Recall and TweakScale, reproduce the problem, and then post here a link to the KSP.log published on DropBox or something, and I will be able to pinpoint the culprit. My best guess is that there's something triggering the infamous Assembly Loader/Resolver bug on KSP, and TweakScale makes heavy use of that thing!