Jump to content

Lisias

Members
 View Profile  See their activity

  • Posts

    7,454

  • Joined

  • Last visited

 Content Type 

Everything posted by Lisias

  1. Lisias
    I think you nailed, sir! Now I know what's happening!!! That would not help. There's code in need to be written to cope with the new KSP features!!
  2. Lisias
    Modules that implements IPartCostModifier and IPartMassModifier (and perhaps others) are being called by zombies copies of B9PS Modules, and if any of these called modules make a call to Unity, you have a NRE Exception inside it. Such zombies happens every time you switch Scenes (launch a craft, revert to editor, etc). On every scene where the current craft is destroyed, you earn a lot of zombies on the system. And every one of these ones may call other living modules from dead contexts. It's a mishap on B9PS . Since not all modules make calls to Unity from their interface implementations, not all of them are deadly affected by this problem. But all modules are affected in a way or another - at least, by wasting CPU and memory on dead objects lingering around.
  3. Lisias
    ANNOUNCE KSP Recall 0.0.7.3 KSP-Recall 0.0.7.5 is on the Wild, featuring: A (dirty) workaround for the KSP 1.11.x faulty calculation of Funds on recovering crafts (Issue #12) Fixes a deployment mishap for CurseForge A new module, Refunding, as well a new Resource was introduced for KSP 1.11, to counter attack the miscalculation mentioned above. When recovering the craft, the badly calculated refunds will still be there, but an additional Resource called Refunding will be present stealing giving back the losses. Users of EVERY add'on that implementsIPastCostModifier and are running on KSP 1.11.x NEED to install KSP-Recall - or your Career will be seriously hindered. Affected (known) add'ons are: Darth Pointer's Pay to Play FreeThinker's Interstellar Fuel Switch allista's Cargo Accelerators All Angel 125 Add'Ons that uses WildBlueTools Nathan Kell's Modular Fuel System (and Real Fuels) IgorZ's Kerbal Inventory System KOS Kerbalism And many, many others - perhaps Squad's own modules (who knows?) This Release will be published using the following Schedule: GitHub, reaching manual installers and users of KSP-AVC first. Right now. CurseForge. Right now. SpaceDock (and CKAN users). Right now. The reasoning is to gradually distribute the Release to easily monitor the deployment and cope with eventual mishaps. -- -- POST EDIT ---- A lame mistake was detected and fixed. KSP-Recall 0.0.7.5 was published.
  4. Lisias
    Sounds like cryogenics boiling up. Fuels like Hydrogen can't be stored on fueltanks for too much time, they just boil up and leak into space. Choosing the right fuel for the mission (what limits the engines you can use on the mission) is the reason Real Fuels were made. Rule of thumb: Cryogenics (chilled kerosene, LOX, etc) are excellent for first and second stages. Once you establish a parking orbit and start to plan the rest of the mission, it's best to rely on non-cryogenics fuels: kerolox, hypergolic, nuclear, ion, etc. If you really don't want to cope with these details, you can try removing Real Fuels from the system. It will probably screw up any vessels already launched, however - make backups while experimenting things. (I use S.A.V.E to make things automatic).
  5. Lisias
    ROMBUS. Dr Werner is proud. (If you are Brazilian, i can hear your laughing from here!)
  6. Lisias
    I'm playing on a MacMini 6,2 - I7-3615QM, Intel HD 4000 and 16GB . It's way from being a beefy machine, but it does the trick as long I keep the Textures quality at 50% and the Anti Aliasing low or off to save memory bandwidth. Try to lower the Texture Quality, it will probably help. About TweakScale, it deactivates itself on unescaled parts, so if the blowing craft has no part scaled, I can tell you for sure TweakScale is not the culprit. There're a recent bug fixed on B9 Part Switch (still unpublished, sadly) that was injecting bad contexts on TS (and many more too, but TS is sensitive on this problem because it needs to call Unity on that calls), leading it to bork - but this is only a problem if you have the Forum's B9PS installed.
  7. Lisias
    Somehow, this post came to my timeline, so I thought to myself it would be a good idea to add my bits on it! Scaling engines on Career can potentially spoil your game. Make things too easy - in real life, economics and limitations on material engineering prevents us to resize engines at our will, so if you want to follow a "plausible path" on your game, avoid scaling engines. Wheels, tanks, wings, control surfaces, fuselages (a pity we can't scale crewed parts correctly, but this is hardlocked on KSP's guts, nothing I can do), lights, even parachutes, that are the things I scale up and down without a second thought. On the other hand, on my SandBox games, dude... (I'm TweakScale maintainer, by the way )
  8. Lisias
    ANNOUNCE KSP Recall 0.0.7.2 BETA is on the Wild, featuring: Further ehances the workaround for the KSP 1.11.x faulty calculation of Funds on recovering crafts (Issue #12), (hopefully) solving the problem on a way that would work for every Add'On. A new module, Refunding, as well a new Resource was introduced for KSP 1.11, to counter attack the miscalculation mentioned above. When recovering the craft, the badly calculated refunds will still be there, but an additional Resource called Refunding will be present stealing giving back the losses. Now works too with PayToPlay! And without double charges happening due a mishap while launching the craft! DANGER, WILL ROBINSON, DANGER This is a preliminary BETA version intended to be tested on the field. The kludge, I mean, the workaround worked on controlled environments, but I need to test this thing on the field in order to check both the effectiveness and safety of the stunt. PLEASE, PRETTY PLEASE USE THIS ONLY ON DISPOSABLE SAVEGAMES. Add'On authors that make use of the IPartCostModifier please report your findings on the Issue #12. I need your feedback. This Release will be published using the following Schedule: GitHub, reaching manual installers first and users of KSP-AVC. Right now. CurseForge, will not be published. SpaceDock (and CKAN users), will not be published.
  9. Lisias
    Are you sure? My empirical tests pinpoints 1.4.0 as the source of the problem! Did you see how the lifting forces are terribly unbalanced on KSP >= 1.4.0 ? I ended up managing to control roll on all KSPs (buggy or not) by carefully adjusting the Authority Limiter on flight, but I must admit that on 1.3.1 this was way easier, probably because the lifting forces are equally distributed on the surfaces. Also, check the SAS, it's way easier to control the thing using it. Well, in a way or another I never did any tests on anything older than 1.2.2, so I will give this a shot. -- -- POST EDIT -- -- You see, you may have a point! I did the following test on KSP 1.1.3: The behaviour of the Lifting Surface themselves on KSP 1.1.3 didn't differed from my tests on 1.3.1 (and recently 1.2.2), but the SAS behaviour appears to be more... consistent, and the craft is easier to control even without trimming the Authority Limit. I think you found something - whatever the problem is, apparently SAS is involved on it....
  10. Lisias
    ANNOUNCE KSP Recall 0.0.7.1 BETA WAS DITCHED I found a mishap on it and issued a new released with it fixed.
  11. Lisias
    "I'm insane and you are my insanity." -- James Cole
  12. Lisias
    I spent the last two days trying to fix a problem using a very simple craft on career. Once I managed to work around the problem, I thought to myself... "Hell, why not?" And launched the thing! But then I realised that on that early stages of Career (that problem only affected Career), the Kerbonaut was unable to EVA his way into safety.... Boy, this is going to hurt... -- -- -- I'm the KSP-Recall maintainer and I wish to ask for help on testing a work around for a problem affecting a lot of Add'Ons (technically, everything that uses the IPartCostModifier). If you know of an Add'On that is having problems on getting back Funds on recovering crafts, please advise on the KSP-Recall's thread!
  13. Lisias
    ANNOUNCE KSP Recall 0.0.7.0 BETA in on the Wild, featuring: A (dirty) workaround for the KSP 1.11.x faulty calculation of Funds on recovering crafts (Issue #12) A new module, Refunding, as well a new Resource was introduced for KSP 1.11, to counter attack the miscalculation mentioned above. When recovering the craft, the badly calculated refunds will still be there, but an additional Resource called Refunding will be present stealing giving back the losses. DANGER, WILL ROBINSON, DANGER This is a preliminary BETA version intended to be tested on the field. The kludge, I mean, the workaround worked on controlled environments, but I need to test this thing on the field in order to check both the effectiveness and safety of the stunt. PLEASE, PRETTY PLEASE USE THIS ONLY ON DISPOSABLE SAVEGAMES. Additionally, the solution doesn't works, YET, for every add'on on the wild - currently PartModules that changes the cost of the part on the flight scene may be missed. The next Release, 0.0.7.1, will support them but now I need to check how this thing will behave on the field first. Add'On authors that make use of the IPartCostModifier please report your findings on the Issue #12. I need your feedback. This Release will be published using the following Schedule: GitHub, reaching manual installers first and users of KSP-AVC. Right now. CurseForge, will not be published. SpaceDock (and CKAN users), will not be published.
  14. Lisias
    NOTAM Datum perficiemus munus The new and nasty problem on KSP 1.11.x described on this post now have a functional (but not yet confirmed to be safe) workaround. I managed to counter-attack the current KSP's faulty calculation, but not without a cost: I had to heavily mangle the part's resources in order to brute force my way on the problem. Some parts will have the Recovered Funds mangled in order to force the desired result. THiS IS NOT THE FINAL SOLUTION, I still have some small tricks to try that may improve a bit the final results - but the mechanism was stablished and it works as intended. The solution will be available on KSP Recall - and it will work for every affected add'on, and not only for TweakScale. https://forum.kerbalspaceprogram.com/index.php?/topic/192048-ksp-18-ksp-recall-0061-2021-0209/ -- -- POST EDIT -- -- The issue was transfered to KSP-Recall's issue tracker. new Link: https://github.com/net-lisias-ksp/KSP-Recall/issues/12 -- -- POST POST EDIT -- -- The kludge workaround was published for beta testing! Anyone willing to risk his/her SAS FOR SCIENCE is more than welcome to test the thing. (please use it only on disposable savegames, ok?). -- -- POST POST POST EDIT -- -- KSP-Recall 0.0.7.3 RELEASE is on the wild. This one was properly tested, some fixes applied, and it's good to be used. Happy Refundinds!
  15. Lisias
    I want to point out that "dedication" is not the same as "professionalism" or even "intentions". [Hired] Professionals have duties to fulfil, orders to carry on, schedules and goals to meet - they need to do what they are told to do, otherwise they risk losing the job. I don't.
  16. Lisias

    This and that

    Lisias replied to VoidSquid's topic in The Lounge

    Yeah. I try to do not feel (too much) guilty when some of them dies... I'm a bit less "compassionate" however.... I try and retry missions as "training", in these ones dead Kerbals are resurrected - but then I say to myself: "ok, this one is the real deal", and anything going wrong is permanent. But, boy, I feel guilty when this happens... Yep, I know the feeling. Didn't worked by SAS out all that years just because I'm stubborn - besides being stubborn as hell surely played a role on all of that! Uh... I had posted here before! My posts here are an exercise of self-censorship - 3 or 4 posts of mine the way I would really like to write them, and I get banned from the whole Internet.
  17. Lisias

    This and that

    Lisias replied to VoidSquid's topic in The Lounge

    Curiously, pretty similar to a certain entertaining device where we pretend to send green critters into Space (or After-Life, most commonly...) Well, sooner of later the "King" will chop some heads off - he wants a working Cathedral, after all. Take measures to prevent yours from being part of the "Party". Sooner or later, the money ends and someone is going to pay. The Milking Cow will eventually dry off....
  18. Lisias
    To tell you the true, I'm afraid of reporting this and making things worse. Fact: someone squashed the whole IPartCostModifier mechanism, by using the cost from the part's prefab while recovering the craft. Fact: this is not the first time the living part's data is squashed by the prefab counterpart - effectively rendering the living part data useless. Question: where else this is happening on KSP 1.11.x? By fixing this one, what else will break since apparently now is the norm to have living data squashed by the prefab? Question: how Squad is managing damaged parts on recovery? Is this borked too? If not, what are they doing? And why they are ignoring the IPartCostModifier so?
  19. Lisias

    This and that

    Lisias replied to VoidSquid's topic in The Lounge

    That's the thing: There're Cathedral Builders, and there're Wall Builders. We need Cathedral Builders, as someone need to know where, how [when] and why the walls will be built. We need Wall Builders, because badly built walls will collapse and ultimately we will have no cathedral. Software is like cathedrals. Some are yet more complex, by the way. So we need Cathedral Builders and Wall Builders too. And when the Software is constantly failing, crumbling or plain crashing, I think it's more than clear where the problem is! The Wall Builders, Issue Solvers from now on, are committed to the task, not to the product. They will solve the problem with the less effort possible, because they have a lot of issues to solve and "wasting" time on anything that is not directly related to the issue under their nose is, ultimately, counter-productive for them - they are on a schedule. It's up to the Cathedral Builder, that on Software are a role fulfilled by Designers, Manager and Team Leaders, to tell the Issue Solvers how they should close the issue without creating new ones later. They need guidance. It's up to them to guide the programmers about how to solve the problems, otherwise they will take any shortcut they can in order to deliver the task sooner - and once the commitment ends with the task, so the responsibility. Or so they think - because once the Cathedral collapses, everybody will get their heads chopped off by the King.
  20. Lisias
    Essentially how it happened to me too! But I started to watch KSP videos on 2016, I think... And bought my copy on 2018! Welcome aboard! Most of the time!!!!
  21. Lisias
    Waiter, there's a Poof on my sou... Hum... Wrong thread!
  22. Lisias
    Need help? https://github.com/net-lisias-ksp/ImpossibleInnovations/blob/master/GameData/ImpossibleInnovations/Parts/Structural/II-poof/poofing.cfg
  23. Lisias
    It's probably not going to be fixed on 1.11.x no matter what I do, so I will fix it myself the way I can. Waiting for bug fixes after a x.y.1 version was issued is, historically, an exercise in futility - and I have users suffering from this problem NOW.
  24. Lisias
    This broke irremediably TweakScale on Career. You are now punished by using TweakScale by losing more Funds on recovery, instead of getting some back. There's no way to fix TweakScale to cope with this, so it's irremediable for TweakScale (and any other add'on that changes the value of the part after launch) but I think I found a way to "exploit" Career again in a way to overcome the punishment. TweakScale et all will still be broke on KSP 1.11.x, but perhaps I can counter-break them on KSP-Recall. What's not a fix, it will be a hell of a kludge - but it's what I think it can be done, as fixing the problem is beyound my reach if I want to keep TweakScale published on this Forum. Everything, absolutely everything that does the right thing and uses IPartCostModifier should be being affected by this problem.
  25. Lisias
    KSP 1.11.x broke irremediably (at least, under forum rules) TweakScale and probably every add'on that changes the value of the craft after launch (as add'ons that implement damages). The IPartCostModifier interface is being ignored on recovering the craft, besides still being called on recovering the craft. This is not an accident. The KSP Dev Team choose to fetch information from the PREFAB instead of using the value of the already called part's IPartCostModifier interface. The full story is on TweakScale's thread and github.
×
×
  • Create New...