Tig

I'm assuming the crash fix came from this change to the code (Support lost texture event): https://github.com/ihsoft/KIS/commit/aec6da26f218311695bafe5b193709382967d691#diff-9fd2acd34859195030978517c3414ef4 If you don't mind me asking, is this an issue with c# in general or just unity's interaction with it. I've been trying to make a mod of my own (will have to see if that actually continues) and coming from a c/c++ background sometimes managed code is a bit tough. I just don't "trust" it, lol. I read the comments in the code... I guess what I'm saying is that I don't get it... I thought all of those issues just got handled and don't understand why your changes were even necessary in the first place. Specifically, this comment in the code: /// This method <i>must</i> be called if an icon becomes unusable. Otherwise, all the cached Unity /// objects will live and take memory till the scene is reloaded. Some of the internal counters /// will also not get updated as expected. With my background that scares the crap out of me. I thought that was part of the whole it-just-works-managed-code-magic. If I'm asking more than I realize, feel free to tell me be an adult and look it up myself. I don't mind. But if you or someone else that knows could point me in the direction of "its a c# thing" or "its a unity thing" I'd appreciate it.

Thank you for letting me know. Sometimes its hard to tell with kerbal's errors whether or not they are truly an issue.. especially with more complex mods.

I've heard about a bug in Unity on some systems with the log file. You can use the command line -logFile <pathname> to put it wherever you want. Some use it just for the convenience of not having to dig through the appdata folder after one of the updates. I did some more testing.. I'm very suspicious that the crash is happening when the garbage collector runs after having removed a container or two. I don't think its the number of containers so much as it is the timing of when the GC runs after having removed them. Even with padding the heap the game will crash for me if I add and remove a few containers and then just do nothing, waiting for a GC. The game is crashing for me (after having deleted containers) shortly after the current heap size exceeds the prior maximum by 30-40ish MB which in my experience is when the GC gets run.

Hey @IgorZ maybe this will help a little? Steps: Loaded up KSP, I got the IMC-22K on and 10 times before I crashed. Reloaded, Using MemGraph, linked below, and padding the heap with 1GB I got the IMC-22K on and off 39 times. Then crashed. Reloaded, did not pad using MemGraph. 11 times. Crashed. Changed MemGraph's settings in padheap.cfg from total of 1024 to 2048, so now adding 2GB. Reloaded, padded, 69 times (lol, sorry, its true ). Crashed. Now, maybe I got a little faster with my clicking... but I don't think enough for that. I should note this is on my work-in-progress modded 1.6. But I've been playing and modding for long time. No errors in my logs except for a few type mismatches from IFS/KSPIE. As for MemGraph, Sarbian and Padishar and now Linuxgamer did all the hard work, but I volunteered as lead guinea pig, lol. Hope it helps. Thanks much for KAS and KIS, I don't play without them. Edited: Also should note, nothing of any value in the crash log or the KSP.log or output_log except the following from output_log.txt (which I'm sure you're getting as well, but for the sake of completeness): ========== OUTPUTING STACK TRACE ================== ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E501C93) 0x00007FF79E501C93 (KSP_x64) ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E2E1D03) 0x00007FF79E2E1D03 (KSP_x64) ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E2B5F68) 0x00007FF79E2B5F68 (KSP_x64) ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E5EC382) 0x00007FF79E5EC382 (KSP_x64) ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E5ED235) 0x00007FF79E5ED235 (KSP_x64) ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E5FD611) 0x00007FF79E5FD611 (KSP_x64) ERROR: SymGetSymFromAddr64, GetLastError: 'Attempt to access invalid address.' (Address: 00007FF79E7A431C) 0x00007FF79E7A431C (KSP_x64) 0x0000000041B54FC2 (Mono JIT Code) (wrapper managed-to-native) UnityEngine.Object:DestroyImmediate (UnityEngine.Object,bool) 0x0000000041B54F1F (Mono JIT Code) UnityEngine.Object:DestroyImmediate (UnityEngine.Object) 0x0000000059B967E5 (Mono JIT Code) KIS.KIS_IconViewer:Dispose () 0x0000000059BD4464 (Mono JIT Code) KIS.KIS_IconViewer:Finalize () 0x00000000055384B9 (Mono JIT Code) (wrapper runtime-invoke) object:runtime_invoke_virtual_void__this__ (object,intptr,intptr,intptr) 0x00007FFAF97282F1 (mono) mono_gc_is_finalizer_thread 0x00007FFAF984BD7E (mono) mono_unity_class_get 0x00007FFAF9728625 (mono) mono_gc_is_finalizer_thread 0x00007FFAF9793B87 (mono) mono_thread_interruption_request_flag 0x00007FFAF9848E77 (mono) mono_unity_class_get 0x00007FFB3BCB3DC4 (KERNEL32) BaseThreadInitThunk 0x00007FFB3D573691 (ntdll) RtlUserThreadStart ========== END OF STACKTRACE =========== **** Crash! ****

1

Go into CKAN, at the top it says File Settings Help. Click Settings, go down to "Compatible KSP versions" Click the box next to 1.3 so it has a checkmark if there isn't one already. Remove checkmarks from all other versions (1.4, 1.5, etc) Click Accept. ----- CKAN should now only load mods or the versions of mods that are 1.3 compatible. As a note some mods that stopped being updated before 1.3, like 1.2 for example, might still work with 1.3 but no promises. Also some mod authors don't specifically change the minimum version, so if a mod is failing try rolling it back to a time when its max version was in 1.3. This can be accomplished by clicking on the mod in question, then in the right hand frame click on versions. Find the prior version you want and double click, a dialogue box will ask you to confirm installing the earlier version. Hope that helps.

Hi @FreeThinker I'm getting a number of errors in the KSP.log from type mismatches. In a number of the MM patches the moduleID field has a string like moduleID = IFSCCmeshSwitcher. In the error log it says it is expecting an integer. Quick check of the code shows that moduleID is defined sometimes as a string and sometimes as an integer, but I could easily be misunderstanding what's going on. I hope that's helpful. Thank you for the mods. snippet from the KSP.log (no errors in the output_log): [LOG 00:05:24.217] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/CT2502/CT2502' has no database record. Creating. [LOG 00:05:24.222] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/CT2503/CT2503' [ERR 00:05:24.246] Invalid integer value! Field moduleID, value IFSCTmeshSwitcher on object of type InterstellarFuelSwitch.InterstellarMeshSwitch [LOG 00:05:24.374] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/CT2503/CT2503' has no database record. Creating. [LOG 00:05:24.379] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/CT2504/CT2504' [ERR 00:05:24.404] Invalid integer value! Field moduleID, value IFSCTmeshSwitcher on object of type InterstellarFuelSwitch.InterstellarMeshSwitch [LOG 00:05:24.487] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/CT2504/CT2504' has no database record. Creating. [LOG 00:05:24.492] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/EC2501/EC2501' [LOG 00:05:24.516] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/EC2501/EC2501' has no database record. Creating. [LOG 00:05:24.520] DragCubeSystem: Creating drag cubes for part 'EC2501' [LOG 00:05:24.546] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/EC2502/EC2502' [LOG 00:05:24.567] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/EC2502/EC2502' has no database record. Creating. [LOG 00:05:24.572] DragCubeSystem: Creating drag cubes for part 'EC2502' [LOG 00:05:24.598] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/EC2503/EC2503' [LOG 00:05:24.618] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/EC2503/EC2503' has no database record. Creating. [LOG 00:05:24.623] DragCubeSystem: Creating drag cubes for part 'EC2503' [LOG 00:05:24.645] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/IST2501gas/IST2501gas' [ERR 00:05:24.653] Invalid integer value! Field moduleID, value IFSISTGmeshSwitcher on object of type InterstellarFuelSwitch.InterstellarMeshSwitch [LOG 00:05:24.670] PartLoader: Part 'InterstellarFuelSwitch/Parts/TankRevamp/IST2501gas/IST2501gas' has no database record. Creating. [LOG 00:05:24.678] DragCubeSystem: Creating drag cubes for part 'IST2501gas' [LOG 00:05:24.708] PartLoader: Compiling Part 'InterstellarFuelSwitch/Parts/TankRevamp/IST2501lqd/IST2501lqd' [ERR 00:05:24.716] Invalid integer value! Field moduleID, value IFSISTLmeshSwitcher on object of type InterstellarFuelSwitch.InterstellarMeshSwitch

Well I do appreciate that, like I said I intended no personal offense. I just didn't want to see people buy low priced cpus that are bad for KSP, and also didn't want people to waste money on a $500+ cpu. I can believe that with 1866 ram. I think that you'll find that with increased ram speeds you get diminishing returns. Downclocking from 3600 to 3000 I lost *maybe* 1 fps. If you don't mind I'll throw 2 cents in here. I think maybe a current price point chart would be most helpful. I mean the reason most people don't buy a 2080 TI gpu isn't because the don't want one... I'll list one from AMD and one form Intel where I can and bold the one of the two I would suggest. Usual caveat applies: this is for KSP only, it in no way suggests that the bolded cpu is better in all circumstances. CPU price Intel AMD Special Note Absolute cheapest ($60) None Athlon 200G Save $35/50 and buy the R3/i3 if possible Approx $100 i3-8100 Ryzen 3 2200G iGPU in the R3 is vastly superior to the i3's if that matters to you. 60-70% of a RX550 Under $200 i3-8350K i5-9400F Ryzen 5 2600x The R5 is better at almost any multi threaded use over the i3. Just not KSP. Under $310 i5-9600K Ryzen 7 2700x i5 is faster and ~$50 less. Over $310? I truly don't recommend anything. But if you want to spend more than $400 go with the i7-9700K, i9-9900K or the i7-8086K. Once overclocked the i7-8086K should give you the best performance since they were selectively binned. AMD over $310? Nothing. When 3rd Gen Ryzen hits the market that may change. It should be noted that you might easily find 4th through 8th gen i5s and i7s for good prices... but that fluctuates very quickly. Again to ward off any fanboy claims, every intel cpu I've suggested beats its AMD rival out of the box on single thread performance. Also, if its a K they typically overclock more on a percentage basis. One additional note, if you really need to spend as little as possible right now, skip the discrete graphics card, buy the R3 2200G and use the integrated Vega 8 GPU. It's shockingly impressive for an iGPU and demolishes the UHD 630 iGPU in most modern intels.

I'll try to be as nice as I can, but the hardware "advice" in the tutorial is terrible. Astonishingly bad. Sorry. I'll try to give some good advice. KSP is still heavily limited by one major thread. From a practical standpoint you can treat it as a single-threaded game. It also is not very graphically demanding; stock KSP will work fine with any mediocre discrete gpu. ---------- As such the only thing you really need to worry about is the cpu, specifically IPC (instruction per clock) and clock speed. It is positively ridiculous to suggest 6+ core cpus in order to play KSP. Period. Hyperthreading? Again, simply ludicrous. So then, what cpus are the best for KSP? Any modern intel K processor or a 2nd generation Ryzen (also 3rd gen Ryzen when it releases). The best is still any intel K processor that is overclocked. You do NOT need an i7. You do NOT need more than 4 cores. 3rd gen Ryzen might be better when released... no one knows for sure yet. But a K intel is better than 2nd gen Ryzen. If you have to go cheap with intel go with a non-K, an i3 or even a Pentium G series... Suggesting Bulldozer and Piledriver AMD cpus? They have GARBAGE IPC. This isn't even up for debate. IPC ~40% less than intel core proc, ~30% less than Ryzen. The fastest was the FX-9590, 200+ watt TDP and it gets wrecked by sub 30 watt U-series laptop cpus. A FX-6300 as a suggestion? really? A FX-8350? really? Threadripper, sigh. Please, please delete those suggestions. The cheap list for AMD cpus is Ryzen 3 and Athlon 200GE. ---------- I'm not trying to offend anyone. If you've got what you've got and money is an issue you do the best you can. But suggesting that people invest in an AM3+ system from 2012 is pouring money in the toilet. There are cheaper, faster and more future proof platforms readily available. ------- On ram, I'm not aware of any testing showing that ram speed makes any significant difference in KSP. But please don't do this: Please don't overclock your ram unless you really really know what you're doing. If you need DDR4 3000 ram, don't get 2400 ram and OC it. Errors from OC memory are not fun to deal with. Just spend the extra $10 and get 3000 ram. And even then only for some other game - KSP doesn't care. No offense, but anyone on a 7/8 year old FX-6300 doesn't have the qualifications to tell other people what they should do with their modern cpu. To anyone interested, Ryzen in general doesn't overclock as well as Intel core. Yes, you can overclock it, but you'll typically run into thermal/voltage limits faster than you will with an intel K. Good cooling is always necessary. The rest is Silicon lottery... ------ What I would suggest: - An 4 core intel i5 cpu (or i3 8th gen), with a strong preference for a K if you're comfortable with overclocking... it doesn't need to be the latest generation, the difference between a 4th gen and 7th gen i5/i7 at equal clock speeds is tiny. - 8GB of ram, 16 if you're modding - A good but not ridiculous motherboard socket. Not AM3+. Not LGA2066. AM4 good. LGA1151 good. If you're overclocking make sure the chipset supports it. - A decent discrete gpu, suggestions as of the date of this post under $100 would be a gtx 1030 or rx 550. If you're modding, for under $200 I'd go with an rx 570 or 580. ----- Caveats: 1) My advice applies to KSP performance. If you're playing other games then you may want a different processor. If you're encoding video/playing with large databases/etc. you might want a different processor. 2) KSP doesn't need a powerful gpu, that said some mods can quickly increase the demands placed upon your gpu. Stock KSP at its slowest will almost always be due to cpu limitations. Typically occurring during launches as you approach the speed of sound around 300m/s. 3) I'm not an intel/amd/nvidia fanboy. Over the last several years its simply true that intel has had better IPCs and higher clocks. Ryzen closed the gap significantly. I'm very hopeful 3rd gen Ryzen will be a game changer, better pricing for everyone if intel has high end gaming competition no matter what you choose. I suggested mostly amd gpus because as of today's date they have the best price/performance... that can change quickly and can be very different between countries. 4) There is a big difference between Ryzen 3 and 3rd gen Ryzen. Its going to confuse people.

Assuming you want to accomplish this with a small amount of delta-v one of the better ways to do it is by using Jool/Jupiter. This gives you two advantages: 1) The gravity assist from Jupiter will lower the delta-v needed to increase your orbit out to Pluto 2) You can use the large planet for a really cheap inclination change (as you approach Jupiter use a small normal/anti-normal, depending on the direction of the inclination change, burn so that you encounter Jupiter above/below its equator) As a side note, although in theory it would be possible to intercept a large inclination planet without matching its inclination, its darn difficult... like trying to shoot a moving car, except the car is traveling 500mph.

I know that Tweakscale is a dependency for Interstellar. But, are any of the patches in the Tweakscale directory necessary/strongly recommended for Interstellar? Put another way, does Interstellar rely on the patches in the Tweakscale directory in order to function and/or balance with other mods? To be clear I'm talking about the MM patches in GameData\TweakScale\patches, not GameData\WarpPlugin\Patches, the latter I'm keeping of course. Thank you for all your work on the mod.

Hey @DMagic thank you for all the mods. ScanSat 18.3 is causing microstutters approximately every second in game. In any location, even just driving a 15ish part rover around KSC causes the stutters. Its not gpu related, fps is unaffected. Its definitely CPU, you can see the counter flash yellow very briefly approx once every second.. There is zippo in the logs other than a lot of CurrencyConverter spam...is that part of ScanSat? Its not related to GC either, I've done a lot of testing for MemGraph (and GrumpyCollector before it), the GCs for me are around 2 minutes with padding. This is definitely every second and is not correlated at all with GC spikes. Reinstalling 18.2 and the stutter disappears. No other changes other than the rollback. Nothing at all in the logs. Sorry this isn't more helpful; I didn't save my logs from 18.3. I can certainly go back and reinstall them if you'd like but there's nothing there (other than CC spam)... I promise Thank you again, let me know if there's anything I can do.

Indeed I did. I tried to write it in a way that a newer player might understand. The irony of quite a bit of rocket/orbital science is that if you understand how to properly state a question with all proper terminology then you're likely capable of answering the question yourself. But see I find that inclination is the hardest part for newer players to deal with. Differences in inclination can greatly increase the needed delta v budget, and is more difficult/unexpected for beginners. Moho probably is the best example. This was my intuitive sense, I wasn't sure how to prove it... I mean, that the worst time to transfer would typically occur at the greatest inclination difference. Not to say that eccentric orbits don't change dv (your PE vs AP discussion demonstrates that) but just a few degrees of inclination can substantially increase dv.

I think this is a little lost in translation and I think the word that is throwing everyone off, including me, is "maximum" I hate to put words in the OP's mouth, but I think he is asking the following: In the set of all future optimum hohmann transfers between Planet A and Planet B what is the worst case scenario for the delta v required? In other words his question inherently involves non-circular orbits/orbits with inclination changes. (For circular orbits without inclination changes the set of delta v required for all future hohmann transfers between A and B would all be equal, of course) For instance, each year there is some optimal minimum delta v required to go from low orbit Kerbin to low orbit Eeloo. Create an array with the index representing the year and the value equaling the optimal minimum delta v that year to transfer from low orbit Kerbin to low orbit Eeloo. The theoretical array has an index from 1 to infinity. What is the "maximum" number in said array? I think that's what the OP is asking. If its not, well, I still think my version is somewhat interesting... (at least its solvable, lol) PS. I know the optimal transfer window from Kerbin to Eeloo is not exactly one year apart. But dammit, its close enough to make my point.

Short answer: No. Sorry. Longer answer: The command line arguments are actually for Unity, not KSP per se. They direct some aspects of the Unity player but are not visible to the game... I know there was some discussion in the Unity community about creating some sort of API to pass custom command line arguments along but I don't think the feature has been implemented. Squad has zippo control over that.

After I wrote my post I wiped out the entire installation of KSP told it to download again on Steam and went to bed. Guess what? Now it magically works... So I guess we will chalk this up to "user error" lol. On a more serious note, one other issue I had was CKAN crashing which I think has to do with the spacedock.info certificate expiring. I can't download from spacedock without changing my system date. I'm wondering if that caused my issues. I don't know if/how that would affect MJ - but in any case I've downloaded all the mods direct myself, reinstalled everything manually and now it works (with or without the patch provided). So, @sarbian thanks and sorry for the time waste. To anyone else with this issue, I personally am going to skip using ckan until spacedock resolves thier certificate issue. I'm not positive that's the issue but skipping ckan seems to work for the time being. ---- Updated info: Reinstalled CKAN and downloaded different mods. As soon as CKAN gets to a mod that comes from spacedock it crashes. Installed the mods that did download. The mod that was downloaded *immediately before* the one from spacedock does not work...

@sarbian I think I may be having the same issue... I'll try to be a bit more clear on my issue. There is no MJ icon on the spacecenter view and the MJ parts do not show up at all in the VAB/SPH. I did this on a clean install with MJ as the only mod. Installed both with CKAN and manually, same result. No MJ, no MJ parts. Log showed MJ loading, no errors. No joy with 2.6. Also tried a couple of 2.6-xxx dev versions as well, same result. Also tried both 64 and 32 bit just to see, still no MJ. I'm on Windows 10 if that matters. I then installed a couple of other mods (just to make sure I hadn't suddenly forgotten how to mod KSP ) like DMagic and TAC-LS. They worked fine, parts in the VAB/SPH, icons and settings available for the mods where appropriate, but still no MJ. Weird/interesting/maybe important note: I started a career game, and the AR202 was right where it should be in the tech tree. I cheated in some tech and funds and got the AR202 in the tech tree. But again, in the VAB, no AR202... strange. No log, no help link: https://filebin.ca/3o8vPsQuFmn1/output_log.txt I'll say in advance though that I don't see anything in the log file giving me any hint as to the issue. I've done my fair share of modding and tried all the dumb stuff (reinstalling ksp, turning off AV, etc) already. Wouldn't bother you, but I'm really at a loss on this one. As always thank you for all your mods, but especially MJ, MM and the albeit short lived GrumpyCollector PS. I said a couple times that there is no MJ icon because iirc there is one in the spacecenter view and the VAB but I really haven't played much since 1.1.2 so perhaps my recollection is incorrect.

Hey Padishar, I did find a correlation with those long lag spikes I'm sometimes experiencing. Seems to be related to a high number of page faults. From Process Explorer: And a graph shot: The red circles are mine. Each one was accompanied by a long delay (10-20seconds) - sometimes in flight, sometimes on a scene transition, sometimes when purchasing a part in the VAB. Memory goes from ~7.1-7.3GB and jumps up to 9.0-9.1GB. On the first screenshot you can see a normal scene and then the huge page fault jump up to 16,000+ page faults. Now process explorer doesn't differentiate between soft/hard faults (either that or i'm ignorant on how to make it do so ), but in any case my total ram usage never exceeds my 16GB physical ram. So, I assume they are soft faults? But then the I/O spikes as well... But, not that much... the I/O spikes are about 50-60MB. Kerbal and swap file are on ssd's... i can't imagine a 10+ second delay being caused by the swap on an ssd (Assuming that some are hard faults that need disk access in the first place). The graph is about 50ish minutes. So, its infrequent, and not consistent in occurrence. Sometimes 30 min apart, sometimes 5 min. The *only* time I can say with certainty its happened more than once is when I've purchased a part in the VAB. 2 times for sure, maybe 3. Other times I buy parts and no lag at all... sigh. At least I know when I'm being less than helpful... lol. These intermittent issues are hard to nail down. Thought I'd pass it along - tried to get it to do this in stock, but haven't made it happen yet. It's somewhat random even on my modded install. I don't know if its mods or just that in general these issues take a lot longer to occur with stock due to the lower total memory usage and slower heap growth. Or maybe I'm just not noticing in stock because the delays are much more brief. Idk. If I see it I'll certainly report back. I didn't notice this behavior before MemGraph - but I know well enough that doesn't mean squat. As always, thanks again for the mod; I wouldn't be playing the game without it. And I really hope Squad is reading this thread. PS. Clarification on my language, I was taught/learned that "soft page fault" = its already in memory, the OS just needs to tell it where, "hard page fault" = not in ram, gotta go get it from the disk. But I've noticed some people use those terms differently. I have no idea who/what is right (i've seen some people use major/minor) so I thought I'd clarify how I was using the terms. EDIT: Caught it going even higher, nearly 120k page faults, i saw deltas before/after this shot in the 30-60k range. Memory & I/O spike about the same. I think I just alt-tabbed out to Procexp faster this time. (Ignore the total user time, I was doing other stuff on the net running kerbal in the background to see if that had any effect on total memory growth/heap size. Short answer: no)

Glad you figured it out and glad it's solving this nasty issue for lots of us. For anyone confused by the numbers, my personal recommendation would be to double or triple @Padishar's numbers after the colon in padheap.cfg if you're running 16GB of ram. Watch your ram usage with GCMonitor and adjust as you see fit. I wouldn't change his numbers much with mods and 8GB of ram, but again the mods - how many/which ones - can have a profound effect. There is NO one-size-fits-all settings with MemGraph to lessen with the stutter issue. I'm just trying to give you a good starting point for anyone confused... hopefully it helps a few. I wouldn't recommend just adding a zero to the end unless you're in @Galenmacil's range of 32GB+ ram. ( hehe, saw your post before your edit )

Hey @Padishar, thanks for the update. Going to do more full testing on different sizes when I have time, busy week... I've done some limited testing though and just wanted to give the TLDR version since you know my settings and the way I test by now Everything works *exactly* the way you would have theorized in advance. More heap = longer interval = longer stutter frame in ms, but the increase in said frame is slight compared to the increase in the intervals. Thx again. PS. To all other users, I'll say this a bit stronger than the mod author, this mod is in no way for people with 4GB of ram, and frankly be careful with your MemGraph settings with 8GB of ram if you're running stock. Mods will run you out of 8GB of ram very quickly in my experience with MemGraph and GCMonitor. I'm seeing my total ram usage (not heap) grow even in stock with every scene change still and mods exacerbate that issue, at least for me. The moment you run out of physical ram by using this mod, which I've tested, you'll see a big drop in performance even on an ssd. (I have a Samsung EVO, maybe the swap on an m.2 would be fast enough, i don't know, but I'll take donations towards buying one for testing purposes. )

How/when did you meet my relatives?

Excellent. Most of the major mods are updated. To the extent they aren't updated to 1.1.2 sometimes they already work with it. Heck some 1.0.5 mods if they're just parts still work. If you're not sure just go to the mod's forum thread - somebody else already tried it to see if it works. Here's my mod list if you're interested. I believe they all work well together. Not on my list but also worth checking out are Mission Controller 2, Interstellar, Kerbal Construction Time. Also MechJeb is very very popular; I don't think you need both MechJeb and Kerbal Engineer Redux. I use KER right now, but they're both excellent and imo, having one of the two is necessary. I literally would not play the game without one of them. If you have any issues with stuttering, as I do, MemGraph is invaluable. "# of Mods you can use" = "# of Mods that crash your game" - 1 Okay joking aside, performance and number of mods is very system specific. My mods run very well on a i5-6600K @ 4.6GHz and a HD7950 @1100 MHz, 16GB ram, Win 7x64. CKAN and KSP-AVC do a great job of keeping you up to date. But they are tools and you need to understand what they are doing in case something goes wrong. So my advice is to make sure you understand how to manually install a mod. There are *many* places that is already discussed. Also, Module Manager is the keystone that makes all the other mods work together. If you have any desire to create your own mod, mod your installed mods, or understand why one mod is modding another mod then Module Manager is crucial. Learning its basic syntax is for slightly more advanced modders, but well worth the time. If you have any experience with code you'll pick it up quickly.

Although its been answered. Its not a stupid question at all. I started using the platform agnostic "Mod" instead of "Alt" because pre 1.1 I tried out Kerbal on Linux and was cursing at why I couldn't load up the debug. Different "Mod" key. Took me 10 minutes to figure it out. *facepalm*. You'll get no judgment from me, lol. Thank you that makes perfect sense now. Looking at my graph above I perceive, albeit somewhat poorly defined, small stair-steps in my MemGraph chart. I'm wondering if that's happening as it runs out of larger block sizes. I can see how managed memory languages were probably really popular with guys/gals coming from C/C++ into something like C# for some applications. But if there's anything I've learned from this discussion its that even if your memory is "managed" if you don't understand memory management and specifically how your memory management works you can really screw things up still. I've seen memory issues in other managed games, perhaps there is a new generation that never had to deal with unmanaged languages and some of them are exacerbating the problem. At least with C/C++ if you messed up your memory management typically the consequences were very bad very quickly - hence the bug got found and smashed. Now it seems like managed languages can allow people to use very poor techniques, especially in how/where they create, but it still works... and if the program is small enough the issue is not perceptible. Or maybe its just that I'm crusty and bitter that I had a professor who had us learn SPARC assembly... And with 101 mods! Although to be fair some are dependencies, plus about 10 custom MM cfgs. So true, I wrote that in my post because so many people think it is always a crash warning... to be fair, with windows 9 out of 10 times (or more) it probably is a crash. I wonder how many of the "my game is crashing, i quit" reports are just people being impatient with this warning and not waiting a few seconds. Autosave interval is set to 5 minutes, although there is something in the settings.cfg i hadn't noticed before called AUTOSAVE_SHORT_INTERVAL, its set to 30. I changed it to a few different values, didn't appear to have any effect - found a few other posts with the same results. In any event I also use the S.A.V.E. mod, but its set to 20 minutes intervals. I use fraps mainly for the frametimes; I'm big into the frametime analysis and happy to see that more and more gaming websites are focusing on the metric. Its subjective, it always is with fps, but my personal preference is for consistency over average speed. In theoretical gaming world I'd rather a rocksolid 40 fps over 45 fps that is bouncing between 35 and 55. Totally subjective. And yes, the integer does bother me with fraps. Especially since its not only the overlay but the freaking logs as well, which i just don't understand since the frametime log is very precise. Sure, I could recreate my own more accurate fps from the frametime log but why make me go through the hassle... sigh. ----- Thank you as always. If I may, what aspect of Kerbal/Unity/Mono is causing the huge increase in total memory being used? From the beginning of my play session to the end the HWM grew about 800MB or so (from a max of ~3000 to 3800ish) so unless I totally misunderstood you on how blocks are freed and new pages allocated that shouldn't be the issue (or rather only 800MB of it), but my total memory went from 6.8GB to 13.1GB. What else is in there? I thought the vast majority of the rest of the memory being used was for models/textures. My understanding was that all that gets loaded at the start and then should just remain static in memory. But clearly I'm missing something... about 5 gigs of something.

Hey @Padishar, Thought I'd move further testing of MemGraph to well, uhh, the MemGraph thread. But to anyone interested in garbage collection in Kerbal/Unity/Mono and the steps taken to try and at least stem the bleeding, the Grumpycollector thread is a goldmine of info, even though the mod itself is not useful for the average user, it helped point everyone in the right direction. So after actually playing the game for a while I thought I'd report in on my experience. First, the game remains incredibly more playable, but I've noticed a few I-wouldn't-call-problems, but I thought they'd be of interest to you. After a play session of about 3 hours or so - many launches, many landings, many scene changes - I checked in on MemGraph and GCMonitor. HWM is moving between ~3100-3850MB. Obviously quite a bit more than where I start out (hitting Mod-End 3 times and HWM ~2500-3000). My stutters still stay in the 55-60 second range, and my fps is locked at 60 - except of course when it dips on a stuttered frame. Stuttered frames seem maybe a bit longer - I wouldn't be surprised if they were that's traditionally been the case as the heap grows That said, same pattern as before, even if they are longer the relative increase in ms compared to the heap size is small GCMonitor - Perhaps most interesting, my memory usage has climbed to 13.1GB, lol. For reference in my short testing sessions, my usage was about 6.8GB. Windows reports a working set of 10.6 and commit of 11.2 As I'm typing this windows now reports 8.5/8.7, GCMonitor hasn't changed. Long scene transitions: Moving from the MC or the Lab back to the main screen sometimes takes longer than usual. In a few instances the typical windows program crash pops up, simply clicking "wait for program to respond" actually works and Kerbal comes back a few moments later. The problem is my favorite type: Non-repeatable intermittent... it does appear to get worse the longer I play. But, sometimes the transition is immediate even 2-3 hours in. Strange. Some exceptional memory usage spikes Brief moments where MemGraph reports 40-50MB/sec demands, see below Unfortunately, I don't have hard data for you on frametimes. I wasn't running fraps, I've heard anecdotally that it can affect Kerbal's preformance, no idea if its true - but since I was playing for my own enjoyment I thought I'd leave it off. If Kerbal is running and I start up fraps, then Kerbal immediately crashes guaranteed. Don't know if that's common or an issue with just me, but in any event no precise data on frametimes for you. sorry. But I got a picture of MemGraph - looks pretty clockwork on the intervals still: Note the scale of 64MB, yeah those are 40-50MB spikes. And the trend line is pretty clear; the memory demanded per second is rising with time, but each GC resets the "clock" if you will. Is there some reason for the positive slope in between the GCs? I guess what I'm asking is: why isn't the trend between GCs flat? Some growing array/list that instead of getting appended to is being reallocated brand new every frame? Just seems unfortunate with the KB/s being a function of time, its turning the the total heap between GCs into a quadratic. ick.

@Xyphos I read your post and the post important thing I wondered was whether it was Minecraft music or Kerbal music in the background of your dream. I'm... weird.