subyng

Members
  • Content Count

    125
  • Joined

  • Last visited

Community Reputation

29 Excellent

About subyng

  • Rank
    Spacecraft Engineer

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. I would love to contribute to optimizing performance. This mod is fantastic and deserves to run smoothly! Are there any already identified areas where there are memory bottlenecks?
  2. @linuxgurugamer Sorry, I kept forgetting about it. I've made the PR. I have also added some lines for spin stabilization but have left it commented out for now. The basic logic is that it calculates the instantaneous centrifugal force based on the engine position between two frame updates. I also started to add a *very* simplified model where I assume all the fuel is located in the CoM of the vessel, then calculate the time it would take for the "point" of fuel to reach the engine, given it's velocity due to centrifugal acceleration, such that it would take more time than a single physics tick to spin stabilize your fuel. I'll try to get that part working sometime this week.
  3. I've noticed that losing/regaining a commnet connection with Kerbalism can sometimes cause significant stuttering and high memory usage, forcing me to restart after maybe 30 minutes of play time. Would seem like there's some kind of garbage buildup related to Kerbalism's commnet implementation.
  4. Do you have a list of which engines correspond to which real life version? I'm planning to make some configs for Engine Ignitor and I'd like to have some reference!
  5. Great, I'll do it tomorrow! I just did some quick testing and it seems to work brilliantly! In addition to it taking into account the direction of your engine vs the direction of your acceleration, the following cases also seem to work correctly: If you are accelerating up then cut engines, once you beginning falling back down you will be experiencing negative G's, therefore your fuel will be unstable. However, if you are in an atmosphere, you will start feeling positive gees due to air resistance, and your fuel will be stable. If you have negative gees, but you point your engine point forward (i.e. you keep your ship facing prograde), your fuel will be stable. When you are in freefall, as long as you thrust in any direction, your fuel will be stable. While I'm at it I might try to see if I can look at determining centrifugal acceleration so you can spin stabilize your fuel.
  6. @linuxgurugamer I think I found the culprit for why the fuel stability is acting a bit janky. On line 311 in EngineIgnitor.cs, you're doing Math.Cos(a) but Math.Cos() takes in an angle as radians, and the variable a is in degrees. The other thing is that, this angle works great when in atmosphere (it's 0 when your rocket is pointed straight up, 90 when pointed at the horizon), but when you're in orbit it doesn't really make sense. If you're in free fall, the direction in which you would feel a g-force would be along the direction of thrust, your orientation with respect to the planet shouldn't matter. The geeForceVector that you're calculating isn't actually in the direction of gees, it always points down because you're doing Vector3d geeForceVector = vessel.obt_velocity - vessel.lastVel - vessel.graviticAcceleration / TimeWarp.fixedDeltaTime; Dividing graviticAcceleration (about 9.8m/s) by fixedDeltaTime (0.02s) gives you a huge value of ~490, and that vector is always pointing in the direction of gravity, i.e., toward the centre of the planet. I can make a pull request on this if you'd like.
  7. Damn, that also might explain why I have no KK ground stations visible (open or unopen) despite enabling extra stations and using KK ground stations in the settings.
  8. Finally the Mk2 pod gets some love! Great job!
  9. Just to clarify , I meant the science system in stock, I like how science isn't instantaneous in Kerbalism. But yep, I am using monthly budgets and construction time as well!