Jump to content

Captain828

Members
  • Posts

    14
  • Joined

  • Last visited

Everything posted by Captain828

  1. Yeah, I've noticed the same. I guess it's also dependant on how powerful the PC is. As a stress test, I actually have a test craft filled with chutes that hangs KSP for a few good seconds.
  2. No worries man, take care of yourself! It could be that older versions worked fine but here we are with a newer version of KSP, RC and SR.
  3. Given that there are no issues with just RealChute installed this is purely a StageRecovery issue. Did you try with the option off as well?
  4. @vardicd Does the fix work for you? I'm still seeing the calculation being done with the setting on or off.
  5. Yeah, sorry about that, I wanted to keep my reply short for brevity sake but I see now I should have been more explicit: while I first experienced this when I had my full collection of mods, I then went one by one to find the culprit. In the end I found it was from StageRecovery that was doing some expensive calculations for every change in the VAB when RealChutes were present. I then double checked by, again, removing all mods and testing with just these two which resulted in what I wrote above. For the logs, I checked before and they had nothing of help in them.
  6. @linuxgurugamer While searching for why I'm getting such poor performance in the VAB I stumbled into a performance issue with this and RealChutes. The below was on stock 1.11 and just the two mods. With something as simple as this https://www.dropbox.com/s/8arkhxymmaiv1yd/screenshot 2021-01-26 23.33.21.png?dl=0 (I actually had Restock in this one) when I play around with the fuel sliders: SR + RC chutes: below 10FPS SR with no chutes (RC installed): minimal input lag (and maybe 1-2FPS drop if I play with the fuel sliders like a mad man) SR + stock chutes (RC not installed): silky smooth SR with no chutes (RC not installed): silky smooth just RC: silky smooth From a glance it seems clear that SR is trying to do some more complicated calculus for RC chutes. Looking at the source code, a quick option would be to defer all immediate calculations in the VAB when RealChute is detected and only run it on-demand when the player presses recalculate. The best option would be to profile and see which calculus causes this. Let me know if you need any logs for this or more information.
  7. I did have all the latest versions. After a careful check and redownload of all 100+ mods I use, I've found the culprit being StageRecovery. With something as simple as this https://www.dropbox.com/s/8arkhxymmaiv1yd/screenshot 2021-01-26 23.33.21.png?dl=0 when I play around with the fuel sliders: SR + RC chutes: below 10FPS SR with no chutes (RC installed): minimal input lag (and maybe 1-2FPS drop if I play with the fuel sliders like a mad man) SR + stock chutes (RC not installed): silky smooth SR with no chutes (RC not installed): silky smooth just RC: silky smooth From a glance it seems clear that it's trying to do some more complicated calculus for RC chutes. I will post my findings to the SR mod as well. I'm surprised no one else has found this issue - @stupid_chris can you replicate this?
  8. Not sure if it's just me or some other mod I have but if I have even the simplest of crafts in the VAB with a RealChute on it any kind of fuel tank changes (playing with the fuel sliders) kills FPS. The more chutes I have, the worst it can get, even dropping to 2FPS. The size of the ship actually doesn't affect this at all just the number of chutes. Before I test this out without any other mods and getting logs, is anyone else experiencing this on 1.11?
  9. I can confirm 0.6.0 gets no B9 warnings/errors with the latest version of SystemHeat (0.3.5) installed.
  10. According to https://wiki.kerbalspaceprogram.com/wiki/1.2 that's not the case. There are a number of other mods that use g0 in their ISP-related calculus like Mechjeb, KER and kOS and all of them use the 9.80665 value. That only shows up for close-approach on rendezvous targets.
  11. @Snark I just noticed in the source code that you have public const double KERBIN_GRAVITY = 9.81; which you use as g0 in the fuel consumption formula but shouldn't you be using the standard gravity, as in 9.80665?
  12. Is this build 1036 on CKAN? EDIT: nvm, it is but you need to update Jenkins to pass 1.11 for the CKAN spec (ksp_version_max). He has - check the dev version on CKAN, 2.11.0.0-1036.
×
×
  • Create New...