smjjames Posted April 21, 2015 Share Posted April 21, 2015 *gets the sarcasm detector*ANALYSIS INCONCLUSIVE*go the movies*Lol sarb XDReally though, would adapting to the new aero be simply a matter of plugging in the new numbers? And then checking if things work obviously. Link to comment Share on other sites More sharing options...
sarbian Posted April 21, 2015 Share Posted April 21, 2015 Exp started 2 weeks ago. Since then I spent all my evening and half of my weekends "plugging in the new numbers". The landing sim partially worked for the first time yesterday and there are plenty of other broken stuff. Link to comment Share on other sites More sharing options...
smjjames Posted April 21, 2015 Share Posted April 21, 2015 Oh not as easy as I thought it would be then. Link to comment Share on other sites More sharing options...
futrtrubl Posted April 21, 2015 Share Posted April 21, 2015 The current (old) stock aero was easy when it came to predicting drag as orientation just didn't matter and it was just based on mass of your ship. Now it does (will), so quite a bit harder to predict. Link to comment Share on other sites More sharing options...
SpacedInvader Posted April 22, 2015 Share Posted April 22, 2015 (edited) So this inability to successfully execute maneuver nodes with MechJeb is really starting to frustrate me. I've reduced the node tolerance down to 0.7m/s which is better, but still permits the vehicles to turn in pursuit of the maneuver node before shutting off. In order to prevent any wild chasing of the maneuver nodes, I have to reduce the tolerance down to 2-3 m/s. While both settings are better than the alternative of MJ simply burning endlessly until it's either out of fuel or crashed, this doesn't explain any sort of why this is happening. Something has changed in the last couple of days and I've been unable to determine what it is and how to fix it correctly. I know for a fact that something has changed because vehicles which used to perform just fine with a tolerance of 0.1 must now have their tolerances reduced to the previously mentioned 0.7 or even higher. In response to what Sarbian said about larger ships needing a less restrictive tolerance, I have the following two pictures:As you can see, neither of these is what I could consider a large ship, clocking in at under 10t, and yet both need relatively low tolerances for dV (the latter has been to the mun 4 times with a tolerance of 0.1m/s and yet now it wasn't even working correctly with the 0.5m/s of tolerance seen in that picture) despite their small sizes. I should also mention that in both cases, attempts to achieve specific orbits were often as much as 2-5 km off where in the past MJ had been able to produce nearly perfect orbits.I don't think that MechJeb is to blame here, as I've completely uninstalled / reinstalled the addon several times, and have tested it alone on a stock installation and it's worked just fine. That being said, I'm at a complete and total loss as to what else might be affecting it's ability to properly execute maneuver nodes. Here is a list of my installed mods:000_Toolbar000_USIToolsAIES_AerospaceAJEAnimatedDecouplersArcanumIndustriesARTASETB9_AerospaceBackgroundProcessing-0.4.0.1.dllblackheartChattererCITCMESCoherentContractsCommunityResourcePackCommunityTechTreeConnectedLivingSpaceContract_science_modifierContractConfiguratorContractPacksContractsWindowCrossFeedEnablerCrowdSourcedScienceCustomAsteroidsDDSLoaderDeadlyReentryDiazoDistantObjectDMagicOrbitalScienceEditorExtensionsEngineIgnitorEnhancedNavBallExtraplanetaryLaunchpadsFalconTechFASAFerramAerospaceResearchFieldExperienceFinalFrontier.datFirespitterFreedomTexFS_License.txt.txtftmn_newFuseboxGCMonitorJSIKASKerbalConstructionTimeKerbalJointReinforcementKerbalStatsKerbalStockLauncherOverhaulKineTechAnimationKittopiaSpaceKlockheed_Martian_GimbalKopernicusKOSMOSKSPAPIExtKSP-AVCKWRocketrylicense.txtLICENSE_DDSLoader.txtLICENSE_KittopiaTech.txtLICENSE_Kopernicus.txtLICENSE_KopernicusTech.txtLICENSE_ModuleManager.txtLICENSE_OuterPlanetsMod.txtMagicSmokeIndustriesMechJeb2MechJebAndEngineerForAll.cfgMM_License.txtmodlist.txtModuleManager.2.5.1.dllModuleManager.2.5.13.dllModuleManager.2.5.8.dllModuleManager.2.5.9.dllModuleManager.ConfigCacheModuleManager.ConfigSHAModuleRCSFXMunarSurfaceExperimentPackageMyModsNASAmissionNavyFishNearFutureConstructionNearFutureElectricalNearFuturePropsNearFuturePropulsionNearFutureSolarNereidNovaPunch2OpenResourceSystemOPMOPTORSResourcePackOuterPlanetsMod.versionPlanetShinePreciseNodeProceduralFairingsProceduralPartsRCSBuildAidRealChuteRealFuelsRegolithRemoteTechResGenRLA_StockalikeSane StrategiesSCANsatScienceAlertSDHISETISETIgreenhouseShipManifestspacetuxSquadStageRecoveryStationScienceStockBugFixModulesTechManagerTextureReplacerThrottleControlledAvionicsThunderAerospaceToadicusToolstoolbar-settings.datTrajectoriesTriggerTechTweakableEverythingTweakScaleUmbraSpaceIndustriesUniversalStorageVenStockRevampVirgin KalacticVNGWaypointManagerWildBlueIndustriesWombatConversionsYarbrough08000_FilterExtensionsI know it's a long list, but I'm really hoping someone here might know if anything in there could have a negative effect on the way MJ handles manuevers. Again, I know for a fact that this is something that has changed recently, though as far as I can remember, the only recent variable changes to the mix were some edits to MJ's attitude adjustment page. That said, if I can't find the culprit, I'm going to have to start again from fresh installation of the game and start working back to where I am now. I feel like it's going to be a headache, but considering all of the different mods I like to run, I'm probably going to have to start obsessively keeping a changelog for my installs so that when something like this crops up, I can identify quickly all of the recent changes that could have caused it.On that note, I'm really hoping someone can help me out here as I'd really hate to lose weeks of game progress because I can't track down the source of this issue.Thanks.PS: Before someone tells me to just accept the lower tolerances, I don't actually think that is an option. I rely heavily on MechJeb because the bulk of the fun in KSP for me is in designing the vehicles rather than hand flying them everywhere. At these reduced tolerances I can only imagine the difficulties in trying to match orbits with an asteroid or other celestial body, especially when I've got a limited number of engine ignitions thanks to the Engine Ignitor mod. As I see it, I must find a way to correct for this issue or I'll be forced to start over from scratch in the hopes that I'll be able to track the variables well enough to revert it back to working the next time it occurs.PPS: @Sarbian, I know you're pretty busy with getting MJ ready for v1.0, but as a request for some future release, I'm wondering if you might include some sort of break on the number of times MJ ignites / kills the engine. While using engine ignitor, you have a limited number of ignitions, but MJ doesn't make any accounting for this, sometimes rapid firing the engine 10-15 times in a row while on a descent trajectory and leaving you without an engine at a point in the flight where that will guarantee you're going to become a crater. Again, this is just a thought for some future release, but I'm sure I'm not the only person who uses MJ who might want something like this to exist. Thanks. Edited April 22, 2015 by SpacedInvader Link to comment Share on other sites More sharing options...
smjjames Posted April 22, 2015 Share Posted April 22, 2015 I just set it to 0.1, which is the lowest tolerance I believe. Actually, I think that's the default and lowest tolerance. Link to comment Share on other sites More sharing options...
SpacedInvader Posted April 22, 2015 Share Posted April 22, 2015 (edited) I just set it to 0.1, which is the lowest tolerance I believe. Actually, I think that's the default and lowest tolerance.I consider 0.1 to be the highest tolerance because it is the most restrictive. Anything less restrictive (0.7 or 3 from my examples above) is a lower tolerance because it's easier to achieve. If I set the tolerance right now to 0.1 m/s, the maneuver node will start to move up towards "Radial +" at about 2-3 m/s left on the node, and then will never get lower than 0.2 m/s, leaving MJ to burn fuel radially until there is no fuel left to burn.EDIT: I just put together a little imgur album to show exactly what happens when I try to use 0.1m/s tolerance currently:Javascript is disabled. View full albumAs you can see, everything is fine until MJ gets to the end of the burn, at which time the node slides up and pegs itself on Radial+ with 0.2m/s left on the node. MJ will continue to burn like this until there is no more fuel left. This behavior is the same across all of the craft that I've tried to use since this problem has cropped up. Edited April 22, 2015 by SpacedInvader Link to comment Share on other sites More sharing options...
SpacedInvader Posted April 23, 2015 Share Posted April 23, 2015 So it appears that I've solved my own problem and also understand why it might be so obscure. I use Kerbal Construction Time, which includes a simulation mode so you can test your vehicles without having to wait on them to be built. This mode contains something similar to hyperedit (or hyperedit itself) to allow you to jump straight to another planet to test your vehicle there, which is what I was doing to test my Munar lander. Apparently, MJ doesn't like this sort of jumping around, and as a result jumping to the Mun to test resulted in the autopilot acting a little kooky.After giving up temporarily on solving the issue and wanting to get back to the game, I decided to just fly the mission to see how it turned out and everything went just fine, autopilot included. After the fact, I went back and attempted to simply hyperedit the vehicle up the the Mun to see if that had the same effect as the simulation mode, leading to the behavior described above. As I'm guessing many people don't do much maneuvering after using hyperedit (putting stations in orbit is the most common use that I know of), this control issue probably hasn't surfaced many times in the past. Though I feel that MJ probably is responsible to some degree, I would think that this is an issue that isn't worth the effort of fixing as the situations where it would crop up are highly limited and in most cases wouldn't be critical to anyone's missions. Link to comment Share on other sites More sharing options...
sarbian Posted April 23, 2015 Share Posted April 23, 2015 oh, yes. It is something I know of. You have to quick save/load for MJ to be properly initialized when moving SoI like that. I'll have to find the root cause one day because it could hide bugs that would only show when players do a single session without saving or quitting the flight scene. But low priority indeed. Link to comment Share on other sites More sharing options...
Starwaster Posted April 23, 2015 Share Posted April 23, 2015 (edited) The current (old) stock aero was easy when it came to predicting drag as orientation just didn't matter and it was just based on mass of your ship. Now it does (will), so quite a bit harder to predict.No, it really wasn't. I know people like to say that but the issue with pre 1.0 stock aero was that it applied drag as an acceleration instead of a force. (i.e, if calculated drag = x then x was applied as x m/s instead of x newtons of drag)In a way, you could say it ignored mass. (which if you're applying force, you can't do)@Sarbian: Re maneuver node chasing. I have an idea to eliminate node chasing. At the time that MJ first starts execution, have it hold position instead of actually locking onto the node. You can try that yourself manually either with stock SAS or MJ so that basically you're just ignoring that the node other using it for initial orientation. Really there's no need to change orientation regardless of whether the node stays put (which it won't. It's going to dance when it gets to the end of the burn) Edited April 23, 2015 by Starwaster Link to comment Share on other sites More sharing options...
sarbian Posted April 23, 2015 Share Posted April 23, 2015 it applied drag as an acceleration instead of a force.Somewhere in NYC someone will smile when reading that Link to comment Share on other sites More sharing options...
Gaalidas Posted April 23, 2015 Share Posted April 23, 2015 I'm totally lost with that NYC comment.Then again, what else is new when I'm in the picture.Either way, node chasing has always bothered me. I tend to hover my mouse over the "cancel maneuver" button when letting mechjeb do my burns so I can reduce the need for the craft to suddenly pull away and then have to reorient to deal with a miniscule overshot or something when the necessity of accuracy isn't so necessary such as when aligning for an escape burn or returning from a moon. Link to comment Share on other sites More sharing options...
gosnold Posted April 25, 2015 Share Posted April 25, 2015 I had an issue in 0.25 and 0.90 with Mechjeb not working in a KSP fresh install (I am on Windows 8.1). The part showed up and could be put on a rocket but the button did not show up. Changing the permissions of the KSP folder so that it did not ask for confirmation when adding/deleting stuff solved the problem. Link to comment Share on other sites More sharing options...
Starwaster Posted April 25, 2015 Share Posted April 25, 2015 I had an issue in 0.25 and 0.90 with Mechjeb not working in a KSP fresh install (I am on Windows 8.1). The part showed up and could be put on a rocket but the button did not show up. Changing the permissions of the KSP folder so that it did not ask for confirmation when adding/deleting stuff solved the problem.Where is KSP installed? If it's in ProgramFiles (or ProgramFiles (x86)) then move it to a different folder. I installed mine to \Games which I created for that purpose. (ProgramFiles can cause perms issues) Link to comment Share on other sites More sharing options...
Gaalidas Posted April 25, 2015 Share Posted April 25, 2015 Just more evidence that Windows 8 is another step in the wrong direction when it comes to a user's ability to control his/her own computer. Link to comment Share on other sites More sharing options...
Lawthfar Posted April 25, 2015 Share Posted April 25, 2015 Quick questchon. Any idea weather Mechjeb will work in 1.0? Link to comment Share on other sites More sharing options...
billkerbinsky Posted April 25, 2015 Share Posted April 25, 2015 Quick questchon. Any idea weather Mechjeb will work in 1.0? see post #9402 above; the MechJeb developer has access to an early 1.0 prerelease and is working on porting it. Link to comment Share on other sites More sharing options...
Cdr_Zeta Posted April 27, 2015 Share Posted April 27, 2015 Hello,Wow - need I say more I was just gonna say...I applaud and kudos all around and can wait patiently for Mechjeb to updated to 1.0...In career mode it will actually take me a good week or more to even get to Precision Engineering to get the part. Good Luck with Mechjeb update and thank you !!Cmdr Zeta Link to comment Share on other sites More sharing options...
smjjames Posted April 27, 2015 Share Posted April 27, 2015 Sarbian, when will MJ be ready for 1.0? Link to comment Share on other sites More sharing options...
BrutalRIP Posted April 27, 2015 Share Posted April 27, 2015 Sarbian, when will MJ be ready for 1.0?was wondering that myself Link to comment Share on other sites More sharing options...
StoryMusgrave Posted April 27, 2015 Share Posted April 27, 2015 When it's done. Don't rush these people. Comments like this drive devs mad, and could cause them to hang it up and quit. Link to comment Share on other sites More sharing options...
sarbian Posted April 27, 2015 Share Posted April 27, 2015 Or delay the release by an hour. See you after I watch GoT. Too bad you asked. Link to comment Share on other sites More sharing options...
daawgees Posted April 27, 2015 Share Posted April 27, 2015 Or delay the release by an hour. See you after I watch GoT. Too bad you asked.That literally just made my day. LMAO!The current episode is great! Enjoy! Link to comment Share on other sites More sharing options...
Mishkin_007 Posted April 27, 2015 Share Posted April 27, 2015 Or delay the release by an hour. See you after I watch GoT. Too bad you asked.how nice. Link to comment Share on other sites More sharing options...
smjjames Posted April 27, 2015 Share Posted April 27, 2015 When it's done. Don't rush these people. Comments like this drive devs mad, and could cause them to hang it up and quit.Sorry, was caught up in the hype, didn't mean to be an ass. Link to comment Share on other sites More sharing options...
Recommended Posts