Stock Part Balance Mod - v0.3.0 Alpha

[Superfluous J]

The File:

Stock Balance v0.3.0

Requires ModuleManager to work!

Changelog:

• v0.3.x: (Planned) More engines
  
• v0.3.0: Added Poodle and 48-7S, arguably the 2 least balanced engines.
  
• v0.2: First release. Added Inline Reaction Wheels.
  
• v0.1: Initial file, unreleased. started with Photovoltaic Panels.

Planned changes:

• More engine balancing, so everything has a time where it's useful and times where it's not.
  
• Structural masses.

Feel free to propose changes, either general or specific!

The Problem:

As I get more and more in to perfecting my space ships, I am continually coming across obvious best parts. The 48-7S and the LV-N are obvious ones. I also come across parts you should essentially never use. The Poodle, for example. I came on here looking for a mod that would try to balance the parts, and when I couldn't I thought to myself, "Why not make one?"

The Big Idea:

So here it is, in its tiny, tiny glory. This is actually more of a statement of intent and a call for help than an announcement, which is why I'm in the Dev thread and not the Showcase. The idea is to reason out what is unbalanced about a part (or set of parts) and then propose a new set of values that is hopefully better balanced, and then we (or I) can update the mod to include those new numbers. The mod will be a simple ModuleManager file containing the changes. No new parts will be proposed, and no non-stock parts will be included (bug the author if you think it's not balanced!)

See my first reply for a concrete example.

Edited February 2, 2014 by 5thHorseman

[Superfluous J]

Example of reasoning:

My first balance attempt is in Photovoltaic Panels. It's weird to me that the smallest panel that you unlock first is the most efficient, and that the subsequent panels are not better for lighter. Especially the Gigantor. If you can make it with stock parts, I figure, then you should be able to just put a Gigantor there and have it do the same job for prettier and less part counts.

So, the OX-Stat weighs 0.005 and gives 0.75 units per second. The OX-4x panels weigh 3 times as much and give 2 units per second or 2.667 times as much.

My first thought is to define a single "panel" to weigh 0.005 and give 0.75 units. The OX-4x panels have 6 smaller panels, or let's say 3 times the panelage of the OX-Stat. So, they should generate 2.25 units of charge per second. They should weigh 3 times the OX-Stat (0.015 tons) plus some extra weight for the turning motors and paneling, let's say an extra panel's worth (seems reasonable to me, any thoughts?) so they would weigh 0.02 tons. So, they generate more power but weigh a little more too. They should be better in most situations than the static panels because thy rotate, but that seems reasonable to me.

The SP-x panels need a casing to make them stronger, and I figure howsabout one more solar panel's worth or 0.025 (what they weigh now). They have a benefit and a cost. Balanced.

The Gigantor is terrible though. It generates 9x the charge of the OX-4x panels but weighs 20x as much. I think for every 2 OX-4x panels in a Gigantor, there should be 1 panel's worth of structure needed for the frame. Plus, a Gigantor should be 10x as good as an OX-4x just to make upgrading your ships easier. So it should generate 22.5 electricity per second, and weigh 10 times as much as the OX-4x plus 5 times as much as the OX-Stat, or 0.2 + 0.025, or 0.225 tons. This means they would generate 100 electricity per ton, which is roughly twice as efficient as they are in the game but still not as efficient as the others. However, they'll be useful for large builds as they would save structural costs and (more importantly sometimes) part counts.

So to sum up:

[table][tr][td]

[/td][td]

Current

[/td][td]

[/td][td]

[/td][td]

[/td][td]

Proposed

[/td][td]

[/td][td]

[/td][td]

[/td][/tr]

[tr][td]

Part

[/td][td]

OX-STAT

[/td][td]

OX-4x

[/td][td]

SP-x

[/td][td]

Gigantor

[/td][td]

OX-STAT

[/td][td]

OX-4x

[/td][td]

SP-x

[/td][td]

Gigantor

[/td][/tr]

[tr][td]

tons

[/td][td]

0.0050

[/td][td]

0.0175

[/td][td]

0.0250

[/td][td]

0.3500

[/td][td]

0.0050

[/td][td]

0.0200

[/td][td]

0.0250

[/td][td]

0.2250

[/td][/tr]

[tr][td]

e/s

[/td][td]

0.7500

[/td][td]

2.0000

[/td][td]

2.0000

[/td][td]

18.0000

[/td][td]

0.7500

[/td][td]

2.2500

[/td][td]

2.2500

[/td][td]

22.5000

[/td][/tr]

[tr][td]

e/s/ton

[/td][td]

150.0000

[/td][td]

114.2860

[/td][td]

80.0000

[/td][td]

51.4286

[/td][td]

150.0000

[/td][td]

112.5000

[/td][td]

90.0000

[/td][td]

100.0000

[/td][/tr][/table]

Example of Implementation:

So we've got the values we want. Now to mod them. A quick ModuleManager file will do the trick:

@PART[solarPanels1] // SP-W
{
  @mass = 0.025

  @MODULE[ModuleDeployableSolarPanel]
  {
	resourceName = ElectricCharge
	chargeRate = 2.25
  }
}

@PART[solarPanels2] // SP-L
{
  @mass = 0.025

  @MODULE[ModuleDeployableSolarPanel]
  {
	resourceName = ElectricCharge
	chargeRate = 2.25
  }
}

@PART[solarPanels3] // OX-4W
{
  @mass = 0.02

  @MODULE[ModuleDeployableSolarPanel]
  {
	resourceName = ElectricCharge
	chargeRate = 2.25
  }
}

@PART[solarPanels4] // OX-4L
{
  @mass = 0.02

  @MODULE[ModuleDeployableSolarPanel]
  {
	resourceName = ElectricCharge
	chargeRate = 2.25
  }
}

@PART[largeSolarPanel] // Gigantor
{
  @mass = 0.225

  @MODULE[ModuleDeployableSolarPanel]
  {
	resourceName = ElectricCharge
	chargeRate = 22.5
  }
}

[Superfluous J]

I will admit this got less reaction than I expected. Going on regardless, though...

I have balanced the 3 Inline Reaction Wheels. I removed the discussion about SAS (because it's on all command pods and probe cores now so it's not needed in these) and given the 3 wheels different amounts of torque and weight. The basic IRW remains the same, but the other small one now has 2x the torque and the largest one has over 4x the torque of the smallest one, and weighs a full ton. The larger ones also pull more power.

Here's the code, though it's included in the file which I've linked to in the first post.

@PART[advSasModule] // Inline Advanced Stabilizer (Renamed)
{
  @title = Large Inline Reaction Wheel
  @description = The inline reaction wheel system uses a series of spinning discs that are going "Very Fast", to generate the torque necessary to control a spacecraft. This larger model has more wheels and therefore allows you to assert more control. Please do not attempt to service this device while it is running.

  @MODULE[ModuleReactionWheel]
  {
	PitchTorque = 40
	YawTorque = 40
	RollTorque = 40

	@RESOURCE[ElectricCharge]
	{
		@rate = 0.6
	}
  }
}

@PART[asasmodule1-2] // Advanced S.A.S Module, Large (Renamed)
{
  @title = Huge Inline Reaction Wheel
  @description = After many years of research, STEADLER Corps rocket scientists discovered that Kerbal crewmembers just can't be trusted to keep a spacecraft under control. The Huge Inline Reaction Wheel addresses that issue by holding the largest rockets steady. It is highly recommended that crewmembers are kept unaware of the presence of such a device, as experience shows that many Kerbals will see that as a 'challenge to their flying prowess'.
  @mass = 1

  @MODULE[ModuleReactionWheel]
  {
	PitchTorque = 90
	YawTorque = 90
	RollTorque = 90

	@RESOURCE[ElectricCharge]
	{
		@rate = 1.35
	}
  }
}

[landeTLS]

This does look interesting. Havent had time to test it yet tho. The balancing looks good

[manatee321]

Great idea! certain parts need to be tweaked to make them relevant (like your current work for the large ASAS module)

I shall be following the development and progressing of this mod

[Superfluous J]

Great idea! certain parts need to be tweaked to make them relevant (like your current work for the large ASAS module)

I shall be following the development and progressing of this mod

If you have any suggestions I'd love to hear them. I'm by no means an expert on this. Engines is going to be a big challenge but I'm going to start with the Poodle and 48-7S

[Orionkermin]

I know it isn't relevant yet, but what about the costs of parts? I'm sure that will be one of the aspects that parts are balanced by in future builds of the game so, might be something to keep in mind as well. Keep up the good work!

[landeTLS]

The stock engines seem quite well balanced to me. I assume squad spent most of their balancing work om them since it is a game about rockets and the engines are at the heart of that. The only engine that is a little wierd is the poodle. It has so low isp compared to the lv-909(not 100% sure about the name. But its the short little first upperstage 1.25m engine you get), maybe its isp could be increased a bit but not as high as the smaller one as its more convinient to place rather than a cluster of those. But some of the other components could use some more tweaking of their weight etc. Maybe the fuel tanks too? It takes very little to overdo things and make parts eather op or over nerfed. I dont have any more spesific suggestions at this time but ill post some if i think of them

[Superfluous J]

The stock engines seem quite well balanced to me. I assume squad spent most of their balancing work om them since it is a game about rockets and the engines are at the heart of that. The only engine that is a little wierd is the poodle. It has so low isp compared to the lv-909(not 100% sure about the name. But its the short little first upperstage 1.25m engine you get), maybe its isp could be increased a bit but not as high as the smaller one as its more convinient to place rather than a cluster of those. But some of the other components could use some more tweaking of their weight etc. Maybe the fuel tanks too? It takes very little to overdo things and make parts eather op or over nerfed. I dont have any more spesific suggestions at this time but ill post some if i think of them

Yeah the Poodle and 48-7S are the only ones that need any major tweaking. You shouldn't be able to get get Mainsail thrust for less weight and better ISP, simply by putting a bunch of 48-7Ses on the bottom of your ship.

I don't know what the final values will be, but I have some ideas on how to figure them out.

I don't know how much weight tweaking fuel tanks need. I think they're all pretty equal and should be. two 400s equal the mass of an 800 (for example) and I like that.

I know it isn't relevant yet, but what about the costs of parts? I'm sure that will be one of the aspects that parts are balanced by in future builds of the game so, might be something to keep in mind as well. Keep up the good work!

Until I know how money will factor into the game, I'm ignoring the costs. However it's easy to change so if anything seems grossly over- or under-priced, yeah we can modify those when the time comes.

[Weegee]

Ooh this looks interesting. Not sure if I'll be using it, but I will be keeping an eye on this thread. Keep up the good work.

[Krevsin]

I shall keep my eye on this and the results. I do agree the Poodle needs to see more use in my space program, so I'll see what will come out of this.

[sal_vager]

To me the Poodle seemed to be just an LV-T45 modified to fit a 2.5m tank and with an efficiency/power upgrade, that new base and panelling is darn heavy.

[landeTLS]

Yeah the Poodle and 48-7S are the only ones that need any major tweaking. You shouldn't be able to get get Mainsail thrust for less weight and better ISP, simply by putting a bunch of 48-7Ses on the bottom of your ship.

Ah yes. I knew i was forgetting something. That engine is so damn overpowered the only thing nerfing it is the insane partcount you get from using lots of them. Looking forward to seeing what you do with them. Im thinking just straight reducing the thrust to around 25-30kn etc. (I think i remember the default to ne 40 or 45kn)But feel free to play around with it. Remember to keep it usable as a high twr engine for 0.6m vessels

[cfds]

Is the Poodle really that bad compared to the 909? If you count the structural parts that you need to make clustering happen, the TWR of the 909 cluster is only a slight bit better but lacks electric charge generation, has less gimbaling range (not really an advantage in the age of magic reaction wheels) and ist more awkward to integrate into stacks.

[Captain Sierra]

So I pulled some numbers and found out the following.

Poodle:

A cluster of 4 LV909s can create almost the same thrust as a single poodle, for a half ton less. They also have higher Isp ASL. 5 LV909s weighs the same, and produces 30 more kN of thrust with greater efficiency ASL. Translation: The poodle is a bit too heavy (and way too ugly IMHO) to be the best choice for a 2.5m service module engine and it's low Isp ASL makes it a poor choice for a landing engine, even worse if you are using a thrust corrector mod. Using radial attachment ports to cluster 4 of them radially adds 0.16 tons to the mass. Using cubic struts adds 0.02 tons.

48-7S

Currently provides 30 kN of thrust for 0.1 tons with a vac Isp of 350 and an ASL Isp of 300. It is theoretically the third most inefficient of any stack mounted engine, third to only the Ant engine and the mainsail.

It takes 50 of them to equal mainsail thrust. Including the mass of the cubic strut, for clipping, the total mass required is 0.105*50=5.25 tons. A mainsail weighs 6 tons and has an Isp ASL of 280 and vac ISP of 330. It's 20 lower than the 48-7S in both counts. If this did not add 99 unnecessary parts to your craft, it would be unanimously superior to the mainsail.

Here is how rediculous that looks, and I used editorextensions to enable surface attach so no cubic struts, making it even better.

Now, let's compare it with some balanced 0.625m engines from RLA Stockalike.

The LV-T5 produces 5 kN of thrust, has an alternator (electricity producing), and has higher efficiency. When in-space dV is the requirement, this is better.

There is the Kingfisher, a scaled down skipper equivalent. This engine produces 15kN of thrust with roughly the same efficiency as the 48-7S, yet yields less delta V, partly due to higher mass. The RLA Mainsail equivalent (I don't remember the name) gives 30kN of thrust with more weight, IIRC no gimbal, and less efficiency, making the 48-7S the better choice for small core boosters even when the alternative SHOULD clearly be the better option.

I think the 48-7S needs to fall between the kingfisher and skipper. It's Isp needs to be 285 ASL and 340 vac. It should give 10kN of thrust and it's big feature should be an extended gimbal range. This reduces overall delta V and also makes it much less practical both in atmosphere and as a landing engine. It's advantage should be 2 degrees of gimbal range, where the other RLA engines (and stock small engines) have only 0.5 degrees.

Also, I think the LV-1 Ant, LV-1R Ant Radial, Mk 55 Radial, 24-77 Radial, LV-T30, Toroidal Aerospike, and Skipper engine need to be at least looked at. If you want me to make more posts like this, let me know so I can do more maths.

[Superfluous J]

So I pulled some numbers and found out the following...

SNIP

If you want me to make more posts like this, let me know so I can do more maths.

This is EXACTLY what I was looking for. Please, post more I was very leery to tackle engines because I know the basics (mass bad, isp good) I by no means feel knowledgeable enough to balance them well.

Based on Captain Sierra's post, here are the Poodle and 48-7S I'm proposing. I'm in dangerous territory here so please tell me if I'm way off on some point:

I feel the Poodle should be better in space based on the wiki saying "The primary intended role of this engine is for larger upper stages of rockets for orbital insertions, transfers, and landings" So it needs more thrust and... I have no idea what else So I've upped the thrust from 220 to 260, which I think makes it better than the 4 or 5 LV-909 option, as it should be. It already has a vacuum fuel consumption better than all smaller engines which also seems reasonable and a great 2.5 vectoring. Frankly I'm surprised the stock version of this engine is so poor and hope this small change is enough. If it's not, let me know, preferably with ideas of what to change.

The 48-7S I'm just using the numbers given of Thrust 10 (was 30, seems drastic but hey), ASL ISP 285 (was 300), ISP Vac 340 (was 350), and vectoring of 2 (was 1). This is a straight-up nerfing and I'm trusting the analysis but (due to my lack of knowledge I'm sure) it seems like too much. If anybody feels the engine is too nerfed, please propose different numbers.

// liquidEngine2-2    = Rockomax "Poodle" Liquid Engine
@PART[liquidEngine2-2] // Rockomax "Poodle" Liquid Engine
{
  @MODULE[ModuleEngines]
  {
	maxThrust = 260
  }
}

// liquidEngineMini   = Rockomax 48-7S
@PART[liquidEngineMini] // Rockomax 48-7S
{
  @MODULE[ModuleEngines]
  {
	maxThrust = 10

	@atmosphereCurve
 	{
		@key = 0 340
		@key = 1 285
 	}
  }
  @MODULE[ModuleGimbal]
  {
	gimbalRange = 2
  }
}

I've updated the first post with v0.3.0 which includes the above code.

Edited February 2, 2014 by 5thHorseman

[NathanKell]

I certainly *won't* be using this, for obvious reasons, but I do want to congratulate you for taking this on and wish you luck.

Note that the 48-7S was supposed to get a nerf and they gave it a buff instead in .23; IIRC they said it'd be actually nerfed in .24.

[jacobgong]

http://forum.kerbalspaceprogram.com/threads/67343-Simple-balance-changes-for-stock-parts

Good, I was getting tired of the modul manager notations and stuff, now I can just download your version instead.

Problems with stock parts:

-The radial engines are totally useless because you can just add a structure part and use an inline engine for better performance

-the 1 thrust parts are uselss, I gave them 5 thrust instead. Still pretty useless.

-Poodle and the 48-7s have the same problem but in reverse. poodle's TWR is lower than that other 50 thrust engine. so It is just better to get 4 of those instead of the Poodle. And the 48-7S has insane TWR way better than anything else, so you can prettymuch replace every other engine by stacking it.

-all three stock reaction wheels have the same 20 torque. However they have different masses at 0.2, 0.3, and 0.5. And it is the big one that weighs the least.

-mini cube is OP just because it weighs nothing.

In general, bigger parts should have better performance than stacked smaller parts. That means better TWR, less mass (structure and wings) and higher ISP. But for stock parts.

Edited February 2, 2014 by jacobgong

[Captain Sierra]

-SNIIIIIIIIP-

Well, I'll definitely load that up in my alternate install (unfortunately, my grand tour tylo lander depends on the 48-7S being OP, though I do see ways to remedy that, aka, more of them).

I'm mulling over some of the smaller engines and the radials now.

Note, my suggested balancing of the 48-7S was against RLA Stockalike engines and the use thereof would make these changes seem a bit less drastic.

the mark 55 radial is just a sad piece of crap. I'm kind of hitting a wall with where to go on that one. It basically needs some serious all-around improvements.

[NoMrBond]

The Rm55 (at 320 vISP) should have ~315 T/t, since it weighs 0.9t that should put it around 285 thrust (or around 200 if you account for the 3' vectoring) just to be in line with other (non 48-7s) stock engines

So, options are

Leave weight / thrust at 0.9t / 120kN and adjust the ISP up (for ~133 T/t @ 3') making it ~315/365 for sl/v ISP respectively

Leave ISP / Thrust at v320 / 120kN and drop weight down (for 220 T/t @ 3') making it ~0.55t

Leave weight / ISP at 0.9t / v320 and pick up the thrust (for 220 T/t @ 3') making it ~200

Or move it into a gap not currently serviced by other engines, the 3' gimbal/VTC eats into the engine stats heavily so you could drop that to bump other stats.

Edited February 2, 2014 by NoMrBond

[Absolute Human]

Do you know this thread?

The same idea is popping out everywhere! And I'm not an exception

I have a SpacePort page which contains basically the same changes, plus tweaked attachment nodes. Not that it is somehow related to you, but I think some of the thoughts were already meant to be implemented in the stock game quiet a while age. Its a pity that Squad keep leaving loose ends in their work...

*EDIT:

Oh, I see, you have. I read the thread more carefully, everything here is like a music to my ears.

I didn't make really careful mathematical calculations on my own changes, but the points and conclusions are so much like mine!

I personally also tweaked the inline reaction wheels, fixed some of the strange illogical values of their pod and core eqivalents, nerfed the 48-7S and rebalanced 24-77. I also changed some drag values from my point of logic. Haven't touched solar panels though. (I wrote more on the space port page? feel free to check if you want)

Edited February 14, 2014 by Absolute Human

[Superfluous J]

Do you know this thread?

The same idea is popping out everywhere! And I'm not an exception

I have a SpacePort page which contains basically the same changes, plus tweaked attachment nodes. Not that it is somehow related to you, but I think some of the thoughts were already meant to be implemented in the stock game quiet a while age. Its a pity that Squad keep leaving loose ends in their work...

*EDIT:

Oh, I see, you have. I read the thread more carefully, everything here is like a music to my ears.

I didn't make really careful mathematical calculations on my own changes, but the points and conclusions are so much like mine!

I personally also tweaked the inline reaction wheels, fixed some of the strange illogical values of their pod and core eqivalents, nerfed the 48-7S and rebalanced 24-77. I also changed some drag values from my point of logic. Haven't touched solar panels though. (I wrote more on the space port page? feel free to check if you want)

I had not seen that thread. Feel free to take and run with anything you find here. While I love the idea I never seem to have the time. I still want to tweak more engines and a few odds and ends in the structural areas (especially impact tolerance) but... Yeah time and all that

It seems I went with more reaction wheels on the large ones, while you went with less wheels on the small ones. I don't know which is more realistic but I figured why make things harder on myself? I never even thought about changing node attachment points. Clever.

Okay, out of time

[Absolute Human]

You had not? But the author have already posted the link here

http://forum.kerbalspaceprogram.com/threads/67343-Simple-balance-changes-for-stock-parts

Good, I was getting tired of the modul manager notations and stuff, now I can just download your version instead.

Problems with stock parts:

-The radial engines are totally useless because you can just add a structure part and use an inline engine for better performance

-the 1 thrust parts are uselss, I gave them 5 thrust instead. Still pretty useless.

-Poodle and the 48-7s have the same problem but in reverse. poodle's TWR is lower than that other 50 thrust engine. so It is just better to get 4 of those instead of the Poodle. And the 48-7S has insane TWR way better than anything else, so you can prettymuch replace every other engine by stacking it.

-all three stock reaction wheels have the same 20 torque. However they have different masses at 0.2, 0.3, and 0.5. And it is the big one that weighs the least.

-mini cube is OP just because it weighs nothing.

In general, bigger parts should have better performance than stacked smaller parts. That means better TWR, less mass (structure and wings) and higher ISP. But for stock parts.

Anyway, I didn't try to make things exactly mathematical. Just moved some values below or above than the others.

BTW I've found that some 2 meter parts don't have the apropriate

breakingForce = 200

breakingTorque = 200

that all the others have, and so tend to break off more often. Had to add them manually!

[Superfluous J]

You had not? But the author have already posted the link here

Yup. And I even have vague memories of reading the posts below that post but I don't remember going to it. One thing: I have a very bad memory

[Strikerklm96]

And do you plan to update the costs, I would love to help you and or give suggestions on some stuff. I read some guys article about how the engine thrusts, ISP's are all really unbalanced and he was right! I can't seem to find the post now, but I remember everything!

Edited July 20, 2014 by Strikerklm96