Jump to content

Part Costs in KSP: Why Are They So Weird? And What Should Probably be Adjusted


Recommended Posts

Part cost isn't something that is normally brought up with regards to balance or realism. Part cost is only important in Career, and even in career, it isn't really that big of a concern, so its understandable why it isn't brought up that much. Still, it is very much a part of the game that should be looked at, as there is a lot that can be improved upon.

Before diving into KSP costs, it's good to look at the costs of components in real launch vehicles.  Now, pricing launch vehicles is surprisingly difficult, but fortunately just relative costs for parts is all we need for now. For the planned Vulcan launch  vehicle, it is estimated that the engines and related components are around 65% of the cost of the first stage. For the Falcon-9, propellants only make up around 0.3% of the costs of the launch vehicle. Now of course, we need to keep in mind that these values might not be accurate for all launch vehicles, especially things that are not conventional rockets, but hopefully its good enough to give a rough idea. From this, it seems engines  are the most expensive part of the launch vehicle, and particularly the first stage, with fuel being only a minor contribution.

Looking at the first stages of various stock and dlc vehicles, engines only comprise around 15-55% of the first stage cost, lower than what one might expect. Fuel costs are much higher than expected, between 5-30% of the first stage. These two departures from realism could be resolved by simply reducing the cost of liquid fuel, oxidizer, and probably also monopropellant for good measure. In fact, liquid fuel is currently 4.4 times as expensive as oxidizer (0.8 funds/unit vs 0.182 funds/unit), so even just reducing liquid fuel and possibly also monopropellant would work (monopropellant has a cost of 1.2 funds/unit).

Now you may have noticed that those price fractions varied a lot, and the reason for that is aerodynamic parts, with control surfaces in particular. As a point of comparison,  the smallest and cheapest control surface is the Elevon 4 (the small one), which has a steep cost of 400 funds. This is more than a LV-909 Terrier (390 funds), and two-thirds the price of a Mk1 Command Pod (600 funds). Now, Elevon 4 costs the exact same as an Elevon 1 (the normal one), so its really only tied for cheapest control surface, but that should show that something is wrong with these prices. The Elevon 5 (slanted one) is more expensive than the Elevon 3 (large triangle one), despite giving less lift and being unlocked in the same tech node. The AV-R8 winglet costs more than the Tail fin, despite the Tail fin giving more lift and deflection, and being unlocked in a tech node of the same price.

Now, these are not the only oddly high prices with regards to aerodynamic parts. Wings are slightly less expensive than control surfaces, but not by much. The AV-T1 in particularly is especially crippling, costing a whopping 500 funds for a small wingless. This is 5 times the cost of similarly sized wings unlocked not soon afterwards, so they are clearly a rip off. Now, the probably-too-high costs of some of these wings can be excused by the fact that they are capable of withstanding reentry heat, not something most aviation wings need to handle, which can be seen with the cheaper but less capable FAT-455 series of parts. Nosecones, especially the smaller ones, are also probably a bit to costly, with costs approaching those of engines in the same size.

Moving to structural parts, for the most part they are rather inexpensive, which would be expected, but there are some oddities. The Structural Fuselage actually costs more than the Mk1 Liquid Fuel Fuselage its based on (when empty), despite it presumably being far easier to manufacture without an integrated fuel tank. The size adapters are also pretty costly, being around as expensive as nosecones. The multicouplers have some odd prices too. The 1.25m tri and quad couplers are both more expensive than their 2.5m counterparts, with the TVR-2160C Quad-Coupler costing 2000 funds, compared to the TVR-400L Quad-Adapter costing just 800 funds.

There are more examples of oddly high/low prices. Mk16 parachutes are nearly as expensive as the Mk1 command pod, radiators cost close to or more than the ISRU parts they are designed to cool, J-90 Goliaths are less than twice the price of the J-33 Wheesly while having a built in air intake and side mount, the Clamp-O-Tron Jr. being over twice as expensive as the regular Clamp-O-Tron, the list goes on.

Hopefully this is enough to clearly show that prices need some revisions, if not a total overhaul. It would nice if this would be done in KSP, but knowing this community, I'm sure people would much rather the time be spent on something more exciting.

Edited by wafflemoder
Link to post
Share on other sites
22 minutes ago, wafflemoder said:

Part cost is only important in Career, and even in career, it isn't really that big of a concern

Build big!

22 minutes ago, wafflemoder said:

Before diving into KSP costs, it's good to look at the costs of components in real launch vehicles.

Look at real dV - it is why launching from Kerbin looks cheap.

25 minutes ago, wafflemoder said:

For the planned Vulcan launch  vehicle, it is estimated that the engines and related components are around 65% of the cost of the first stage. For the Falcon-9, propellants only make up around 0.3% of the costs of the launch vehicle.

Cost of development?

26 minutes ago, wafflemoder said:

From this, it seems engines  are the most expensive part of the launch vehicle, and particularly the first stage, with fuel being only a minor contribution.

Only at the begining - later on when You "mass" manufacturing them (like soviets) they are dirt cheap.

30 minutes ago, wafflemoder said:

The size adapters are also pretty costly, being around as expensive as nosecones

This is reasonable. They are hard to manufacture in comparision with tanks. Try to roll a cone. Of course after You cut sheet in proper way that led You to trash lot of material. And in such aplication direction of rolling (on sheet) is not free, so You can cut only in certain area on the sheet and throw away the rest.

33 minutes ago, wafflemoder said:

There are more examples of oddly high/low prices. Mk16 parachutes are nearly as expensive as the Mk1 command pod

Reasonable. Such parachutes are extremaly expensive due to certification.

34 minutes ago, wafflemoder said:

radiators cost close to or more than the ISRU parts they are designed to cool

Heat exchenger to work in vacum are extremaly expensive because they are not mass manufactured - we do them stacking layers by hand. This are very expensive hands and eyes, because You have to inspect every plate for cracks, then test whole batch, choose one if any is good (if not set another batch in furnance) and so on.

37 minutes ago, wafflemoder said:

Hopefully this is enough to clearly show that prices need some revisions, if not a total overhaul. It would nice if this would be done in KSP, but knowing this community, I'm sure people would much rather the time be spent on something more exciting.

Does pricing play any significant role in KSP?

Link to post
Share on other sites
20 hours ago, vv3k70r said:

Look at real dV - it is why launching from Kerbin looks cheap.

Its not the actual cost of the vehicles, and this is arguably just to not hinder new career mode players. The prices of parts are weird in relation to each other. 

21 hours ago, vv3k70r said:

Only at the begining - later on when You "mass" manufacturing them (like soviets) they are dirt cheap.

That is true, but that also applies to every other part in KSP, not just engines. Relative pricing should still hold if all parts are manufactured at equal rates. The engines in cars and aircraft are good examples of this in action, as both are very much mass produced and engines still make up a large fraction of the total cost.

21 hours ago, vv3k70r said:

This is reasonable. They are hard to manufacture in comparision with tanks. Try to roll a cone. Of course after You cut sheet in proper way that led You to trash lot of material. And in such aplication direction of rolling (on sheet) is not free, so You can cut only in certain area on the sheet and throw away the rest.

The actual material costs should be minimal in comparison to the manufacturing process.  Rolling cone shapes also isn't that much harder than a cylinder. You'll need a new form, but that goes for tank diamters as well, so itsn't likely to increase the cost that much. 

21 hours ago, vv3k70r said:

Reasonable. Such parachutes are extremaly expensive due to certification.

Yes, parachutes would be expensive to certify. But the same goes for a command pod, which is also larger and much more complicated.

21 hours ago, vv3k70r said:

Does pricing play any significant role in KSP?

Not really, and that's kind of the point. You'd expect the price of your vehicle to be more than a passing thought in a gamemode about paying for your rockets, both for the players and the developers, but that isn't really the case right now.

Link to post
Share on other sites
1 hour ago, wafflemoder said:

The engines in cars and aircraft are good examples of this in action, as both are very much mass produced and engines still make up a large fraction of the total cost.

Largest fraction of the cost is in revenue (even in form of wages). Socioeconomic machine that is orginized to deliver is rare and thus expensive (in terms of exchange). But product itself is dirt cheap. Transaction price for insiders have significant discount.

1 hour ago, wafflemoder said:

Rolling cone shapes also isn't that much harder than a cylinder.

How many cones from this special materials (let say inconel or hast alloys) have You done? Because if You make it wrong cost is on You (and so on customer to cover risk, this material arent cheap) and there is much more aplication where You roll cylinder then cone (so less people can do that) and most machines have significant disadventages in rolling cones so machines for them are more expensive, less used (so more expensive) and have other issues (like machine size to part size) that make it even more expensive (and so less usefull because of service cost for other aplication - they are to valuable) - this drive price by multipler.

1 hour ago, wafflemoder said:

But the same goes for a command pod, which is also larger and much more complicated.

No - if parachutes do not open You know who is to blame, if command pod manufactured in west arangment burn - You need to investigate what company made responsible part and if its certification cover usage in this particular issue (quite often not, issues came from complexity).

1 hour ago, wafflemoder said:

You'd expect the price of your vehicle to be more than a passing thought in a gamemode about paying for your rockets, both for the players and the developers, but that isn't really the case right now.

Pricing is just for funny gameplay where You are not sending flotilla of crafts.

Link to post
Share on other sites

Totally agree with the OP, the part prices make zero sense for the most part (along with a lot of other part values like mass, storage capacity, thrust, etc and the tech tree in general).

There's been a few halfhearted attempts by the devs to balance the parts, but given their reluctance to break people's craft files (even prior to 1.0) the game never got the deep balance pass it needed and we're probably forever stuck with these bad placeholder values.

At least the values are easy to change using modulemanager.

Edited by Lord Aurelius
Link to post
Share on other sites

Yeah I would love a full-on balance pass of part prices, masses, and even in a few cases exact dimensions.

I'd love this to come hand-in-hand with an actual integration of the part upgrade system that's sat unused (by the base game) for years.

I'd also love to be able to set a HUGE initial cost for parts. Like how you can be forced to purchase the part before using it now, only much steeper. I want to have to actually consider whether or not I want that new engine. Part test contracts should give no payout and maybe even cost money, but unlock the  part for "free"

Link to post
Share on other sites

Yeah. As much I would like a total overhaul of the entire game balance, that is a lot of work. Part cost should be one of the easiest to rebalance since its basically self contained, which is why I focused on it.

Link to post
Share on other sites
On 12/31/2020 at 6:20 PM, vv3k70r said:

No - if parachutes do not open You know who is to blame, if command pod manufactured in west arangment burn - You need to investigate what company made responsible part and if its certification cover usage in this particular issue (quite often not, issues came from complexity).

I would really love for you to cite a source that would suggest one to believe that parachute certification is going to exceed the cost of command pod manufacture. Like you said:

On 12/31/2020 at 6:20 PM, vv3k70r said:

Largest fraction of the cost is in revenue (even in form of wages).

Which would suggest that, based purely upon manpower, R&D, parts, complexity, etc, the command pod must be more expensive than any parachute.

 

Link to post
Share on other sites
4 hours ago, MarsUltor said:

I would really love for you to cite a source that would suggest one to believe that parachute certification is going to exceed the cost of command pod manufacture. Like you said:

Go and buy one (or anything that have cert for aerospace) and then see the diference to price of material and work costs.

I can quote my invoices - if there is no need for paperwork You get discount, if You want same paperwork to cover it You get multipler on price.

4 hours ago, MarsUltor said:

Which would suggest that, based purely upon manpower, R&D, parts, complexity, etc, the command pod must be more expensive than any parachute.

Pod do no job, parachute does. You can issue a claim if parachute fail. Pod have no job in which it can fail and cause responsibiliti - it is chassi.

It is very diferent in papers if Your part is acording to order (to drawing - as chassi) and if it works exactly as stated (parachute). You can see if it is in case of drawing when You get it and You can complain. With parachute You will know in last, very exapensive moment and someone have to cover this risk?

I know it is hard to grasp if You are not manufacturing stuff. In case of part by drawing if it pass QC (let say by ISO chain) responsibility is moved to those who ordered. They get what they want and they accepted. It do not work with part that have to execute certain action.

Edited by vv3k70r
Link to post
Share on other sites

Around when 1.0 was released, Squad used usage metrics (the ones we think mean they want to rule the world) to rebalance the game, if my memory serves me right. Since then, many parts were added.

  • The parts are priced for the game’s sake, not to be realistic
  • Newer parts follow another logic

I suspect that the only thing that really matters is how interchangeable parts compare. It’s not like you’re going to add a Mk I lander can because its a better deal than a Mk 16 chute, if you need a parachute.

Link to post
Share on other sites
Posted (edited)
2 hours ago, vv3k70r said:

Pod do no job, parachute does. You can issue a claim if parachute fail. Pod have no job in which it can fail and cause responsibiliti - it is chassi.

If basically any part of the pod fails, it doesn't matter if the parachute works or not, you will die before you can come down, if not on the way up. And you can absolutely pin-point what part in the pod is responsible for the failure, that's part of what telemetry is for.

Even for a dumb chasis (which a pod very much is not) it is still larger and hence requires more material and a larger workforce to construct, and so should cost more.

Edited by wafflemoder
Link to post
Share on other sites
8 hours ago, wafflemoder said:

If basically any part of the pod fails, it doesn't matter if the parachute works or not, you will die before you can come down,

No. If for example radio com do not work - You can fix it or use spare. If lights do not work You can fix it or use spare, if... and so on. There is redundancy.

And then You can ask manufacturer of specific part what fail and make claims.

But chassi itself have no claims after You checked and accepted (in case testing procedure, like xray where corectl) - responsibility is on You. There is nothing to fail - it is just a block of metal. You can only misuse it, but it is on You.

8 hours ago, wafflemoder said:

Even for a dumb chasis (which a pod very much is not) it is still larger and hence requires more material and a larger workforce to construct, and so should cost more.

No. Most expensive parts of the ship is not the hull. However it takes most of work hours. But these are cheap workhours in compare to drive, engines, pumps (for balancing) and so on till electronics & nav. In manufacturing we have political incorect names for such a jobs. Margins on hull and on other parts divede which country produce them.

"Should" is a word of will. World is as it is and not as we presume it "should be".

Hence in game these prices are not any important part of gameplay, and I guess they are just to prevent player from spaming space.

Edited by vv3k70r
Link to post
Share on other sites
24 minutes ago, vv3k70r said:

No. Most expensive parts of the ship is not the hull.

Methinks there might be miscommunication.  "Pod" != hull. "Pod" == hull + avionics + lights + life support + ....... 

Just like "parachute" != "canopy" but "parachute" == canopy + rigging + deployment mechanism + ....

Link to post
Share on other sites
1 hour ago, micha said:

"Pod" != hull. "Pod" == hull + avionics + lights + life support + ....... 

Just like "parachute" != "canopy" but "parachute" == canopy + rigging + deployment mechanism + ....

There is no miscomunication. Pod cant fail as a whole - subsystems are separate, and if one fail You have redundancy. There is nothing like "pod fail".

If parachute fail You will notice for sure.

Certification is diferent for system (parachute) that can fail as a whole because of its subsystems and only redundance is to have a second system (second parachute). And diferent where subsystem do not cause catastrophic failure (critical for execution) that cannot be substitute by this system itself.

You do not take second pod just in case, but You can take second parachute.

Parachute is certificate to work in very specific condition (pressure, air speed, temperature, chemical composition of gas) and if You do not deliver them (as SpeceX on early atempts) it is Your problem. But if conditions are right and after trigering it is not doing the job You can have claim.

Pod have no certification as whole - You asemble it (or order to asemble by Your documentation) and You cover it. Parachute is deliver as working piece and someone else cover it (even if one of Your company division). There is a split resposnibility for subsystems in pod and there is no split in parachute. Same way works execution of orders - there are people that cover whole range of subsystems and they give go or nogo for it after they get go from all comanded subsystems (responsible personel). In case of electrical power, hydraulic, pumps, coms, lifesup You can have not 100% confirmation and it still can get go. Parachut is 0 or 1 it works or not. So responsibility for covering 97% power supply is diferent then covering not 100% of deceleration.

 

Please refer to supply chains how it work. It explain a lot. What is critical function, redundancy and responsibility in such chain. Guy that put his signature on parachute hang on it. Literaly. In most parachute asembly works people that using them, it is very popular in military organisation (french example). After use they check them, they pack them back, and they are reusing. So guy that put signature on parachute is responsible for his colege, and his coleges are responsible for him. This signature is not cheap.

Link to post
Share on other sites
2 hours ago, vv3k70r said:

There is no miscomunication. Pod cant fail as a whole - subsystems are separate, and if one fail You have redundancy. There is nothing like "pod fail".

If parachute fail You will notice for sure.

You're talking IRL. I believe most other people here are talking KSP (where a Pod is a single item, not a collection of (redundant) subsystems). Of course, I reserve the right to be utterly wrong on this :)

The OP was (I believe) saying that it doesn't make sense that a fully assembled Pod consisting of chassis plus all integrated subsystems and components costs the same as a parachute system. And I'd go out on a limb here and say the same is true IRL; that is, that a fully assembled command capsule (eg, Dragon) costs WAY more than the parachute subsystem integrated into it (IRL, the parachute system is just a pod subsystem; in KSP they are separate items).

Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...