Jump to content

Nnimrod

Members
 View Profile  See their activity

  • Posts

    199

  • Joined

  • Last visited

 Content Type 

Posts posted by Nnimrod

    [1.8-1.12] Ferram Aerospace Research Continued: v0.16.0.5 "Mader" 03/04/22

    in KSP1 Mod Releases

    Posted · Edited by Nnimrod

    I've got a question. 1.12.3 with expansions, FAR 16.0.4 and kerbal engineer 1.1.9 installed, nothing else.

    I load up the stock Mallard, raise the landing gear, set altitude to 10km, and mach number to .535. Calculate data/stability derivatives gives me a reference area of 122 ms^2, an AoA of 7.02544 degrees, a forward velocity of 157.864 m/s, and coefficients of lift and drag of 0.84 and 0.0632. With an atmosphere density of 0.288 kg/m3, that comes to a lift force of 367,762 newtons, or 37,501kg weight. Far says that this if for level coordinated flight, so lift force should be exactly equal to weight, but KSP says the weight of the mallard in this configuration is 39,046kg. I tried a few different speeds/altitudes, and came up repeatedly with a lift force of only 37.5 tons.

    Is this a rounding error or something, or am I missing something? Also, is there a way to see FAR's outputs with more decimal places? Or even better have it print a range of values to a text file? Like for instance the precise values of the AoA sweep?

     

    edit: It looks like my inaccuracy stems from the density value. I pulled them from the KSP wiki. I'll try using some values produced using the USSA equations on Ohiobob's site, because apparently that's how KSP's atmosphere works in stock now. Which is pretty cool. 

    float curve calculator in a spreadsheet?

    in KSP1 Mod Development

    Posted

    I was bored, and thought I'd make a little spreadsheet to calculate a bunch of parameters for planes like maximum sustained turn performance, climb rate, stall speed, etc. FAR is doing the heavy lifting, because it calculates Cl and Cd, but I also need to calculate thrust available, thus this thread. I use google sheets, and I haven't done this before.

    Affordable joystick for this game?

    in The Lounge

    Posted

    13 hours ago, Spacescifi said:

     

    Gamepad controllers are cheaper still if you don't mind.... just putting it out there.

    Don't know where you live.... but in the USA $100 is not much money, especially when working full-time.

    So if you are not then I could see how it could be expensive to you, when it is not so much if you have the money for it. But if you don't.... then work for it.

    Not old enough? No problem. You could teach english online to some other country, or tutor some kids online.

    So many options that will pay you in experience that will serve you well even when not playing KSP.

    haha $100 is not much money :D Don't I know it. Inflation is a helluva drug. I can afford more, but are joysticks really that expensive? I didn't even do any research before posting here. The last joystick I bought was for Jane's USAF, and it uses a gameport connector (I think).

     

    If joysticks are just that expensive then give me your recommendations disregarding the $100 preference.

    Question about how KSP loads resources, and trying to save on memory usage.

    in KSP1 General Mod Development Help and Support

    Posted

    I'm making a tech tree mod, and placing and balancing every part. For balance reasons I don't want any parts there that I haven't put there, and for memory reasons I don't want the game loading any resources it won't be using. How should I handle mods that add many parts that I won't use, but which add some parts that I absolutely want? Like for instance BDB. The point being that I don't want to load a bunch of textures and models that I won't be using.

     

    Option 1. Don't use those mods, and instead redistribute only the assets required within my mod. The issue with this is that not all mod's licenses/authors allow for this. Although it definitely saves memory.

    Option 2. Include a patch that deletes all parts except those from my mod. I make the other parts mods dependencies and I don't redistribute any art. This is maximally compatible, but is KSP going to waste memory loading resources for parts that aren't even being loaded?

     

     

    Perhaps something could be done with Janitor's closet? It's late and I'm sure I'm missing something, but I figured since I'm considering the possibility that I might actually publish it, I ought to do things right this time.

    How does KSP performance scale with different hardware (And OS?)

    in The Lounge

    Posted

    Title basically, does anyone know if KSP scales any differently than might be expected with DDR3 vs DDR4? Or with different generations of processors? Ryzen and up CPUs first intel? What about win 7 vs 10? Different generations of nVidia cards? Do RTX cards see any benefit?

    I suspect that it probably scales about the same as any other physics heavy unity game, but if anyone has some experience benchmarking it or knows of an article on it I'd appreciate it, thanks.

     

    Just as a starter tidbit, I had for years 16gb of very fast DDR3 (2666c11, Hynix MFR) with my 4790k, and recently popped in two more sticks of slower memory (2133c11, Samsung rev. B I think) along side them, and running 32gb at the slow speed seems to result in greater stability. I haven't crashed once with the 32gb configuration.

    Feasibility of an entryCost mod?

    in KSP1 General Mod Development Help and Support

    Posted · Edited by Nnimrod

    So one of the biggest, and least addressed issues with balanced and engaging career gameplay is failure to adequately model the economic challenges that real space programs/rocket companies face. With that in mind, could a mod that checks for various things which modify entryCost, and then returns a modified entryCost be made?

     

    Consider two example engines, engine A and engine B. Engine B is a related descendant of engine A. The mod checks to see if engine A has been researched, and if it has it checks for a module called "EntryCostMod" or something like that within engine B's .cfg. Inside "EntryCostMod" it finds engine A = .45, and if engine A has been researched, it multiplies engine B's entryCost by the specified value (0.45). This would make an awful lot possible.

     

    For bonus points, some way to handle contracts to buy a certain quantity of parts would be amazing, but difficult I think. Perhaps a reasonable surrogate for this would be to have a bit of code that keeps track of every time you buy a part, and to have a module in the part .cfg  called "QuantityDiscount" or something that looks at the running list of how many times you've bought a part, and uses a simple formula to apply a discount to the cost of that part such that buying the part over and over again makes the cost asymptotically approach a specified portion of the original cost.

     

    Thanks for taking a look. EDIT: Forgot to mention, but if the quantity discount bit is possible, would it also be possible to make it work for resources like fuel?

    [1.8.x-1.12.x] Module Manager 4.2.3 (July 03th 2023) - Fireworks season

    in KSP1 Mod Releases

    Posted

    3 hours ago, TranceaddicT said:

    No, you identify the module to be patched with @MODULE[]. Using 'name = ' would be to change the module name ... you DO NOT want to do that.

    REMOVE: name = ModuleEnginesFX

    Expected result: ModuleEnginesFX will not use AtmCurve and replace any existing velCurve with the KeyValue pairs you've provided.  Note: If the velCurve doesn't exist nothing will be replaced.

    Change "@velCurve" to %velCurve" to create a velCurve if it doesn't already exist. 

    
@MODULE[ModuleEnginesFX]
{
  @useAtmCurve = False

  !velCurve
  @velCurve
  {
  	%key = 0.0000, 1.0000, 	0.0000, -0.0039
  	%key = 0.0500, 0.9998, -0.0039, -0.0058
  	%key = 0.1000, 0.9995, -0.0058, -0.0086
  	%key = 0.1500, 0.9991, -0.0080, -0.0127
  	%key = 0.2000, 0.9984, -0.0127, -0.0187
  	%key = 0.2500, 0.9975, -0.0187, -0.0277
  	%key = 0.3000, 0.9961, -0.0277, -0.0409
	%key = 0.3500, 0.9941, -0.0409, -0.0604
	%key = 0.4000, 0.9911, -0.0604, -0.0891
	%key = 0.4500, 0.9866, -0.0891, -0.1317
	%key = 0.5000, 0.9800, -0.1317, -0.1945
	%key = 0.5500, 0.9703, -0.1945, -0.2872
	%key = 0.6000, 0.9559, -0.2872, -0.4242
	%key = 0.6500, 0.9347, -0.4242, -0.6265
	%key = 0.7000, 0.9034, -0.6265, -0.9253
	%key = 0.7500, 0.8571, -0.9253, -1.3666
	%key = 0.8000, 0.7888, -1.3666, -2.0184
	%key = 0.8500, 0.6879, -2.0184, -2.9810
	%key = 0.9000, 0.5389, -2.9810, -4.4028
	%key = 0.9500, 0.3187, -4.4028, -6.3743 
	%key = 1.0000, 0.0000, -6.3743, -0.0000
  }

    Also: it would probably be better to bang the velCurve before adding it back. This would ensure the velocity curve is the one you've provided should any velCurve exist that extends beyond key, 20. (Not that I think there are any that go beyond key, 6.)

    You also forgot your commas between values.

    Thanks very much. I didn't know there needed to be commas between values. I'll try this tomorrow.

×
×
  • Create New...