What engine do you guys consider to be over powered and game breaking?

[GoSlash27]

They are if you're using the same engine for both the transfer and the landing. Its hard for me to see why I'd use LV-Ns to cart a less efficient engine to Duna. Or to Dres. Or Vall. Or Moho. Or pretty much anywhere but Eve or Laythe.

This would be drifting off- topic, but I'm sure if you tried designing a mission using each mode and comparing the size and cost you'd see why.

If you're interested in exploring this subject further, I'd be happy to join you in a dedicated thread.

Best,

-Slashy

[m4v]

Nukes are supposed to be the be all end all in vacuum. Split atoms, not wood.

too bad that the game doesn't model the inconveniences of using a NERVA, such as killing anyone that isn't behind its narrow shadow shield.

[tater]

True, if that was a mod I'd already have it installed. As Niven said, any sufficiently interesting drive is also a powerful weapon.

i should add that I put Verniers pointing forward on nukes so I can brake without pointing at whatever I am attempting rendezvous with (kill most velocity off axis, then finish with RCS).

Edited February 1, 2015 by tater

[SkyRender]

RAPIER engines mass the same as Turbojets, but possess the same air-breathing thrust/ISP output and remove the need for a rocket engine for final orbital insertion (admittedly not very well, but still). Even with a mod like FAR reducing their power, they remain very, very powerful little engines.

[Jouni]

They are if you're using the same engine for both the transfer and the landing. Its hard for me to see why I'd use LV-Ns to cart a less efficient engine to Duna. Or to Dres. Or Vall. Or Moho. Or pretty much anywhere but Eve or Laythe.

If the mission is to fly somewhere, land, plant a flag, and return home after a minute on the surface, then the LV-N is hard to beat. There are other mission profiles that are also possible, and for many of them, you have to design your ships in a different way.

[Frozen_Heart]

48-7S: Has high thrust and ISP while weighing not much. (Though a high thrust, low ISP probe sized engine would be nice for launchers)

KR-2L: Too light and absurd vacuum ISP for such a high thrust engine.

Turbojet: Its thrust at high speeds is stupid. Should function more like a Ramjet.

Engines I don't consider OP: All of these are much more high tech and futuristic engines and should be far superior to their counterparts. Their main method of balancing should be making them end of tech tree and very expensive.

I actually think that the LV-N needs a buff in atmo. The real NERVA didn't have a stupidly low atmo ISP.

Ion

Rapier

LV-N

[tater]

Rover from the mid 1950s became Nerva in the early 60s. It was tested fully functional in the mid 60s. That's the problem with the tech tree, stuff really came about almost simultaneously.

the trouble with many high Isp options is the inability of the game to deal with continuous thrust.

[numerobis]

Turbojets are only useful in two places, and they are a bit tricky to use, so their massive OP nature is quite mitigated. I wouldn't argue against nerfing them somewhat -- I rather liked Slashy's concept of making them have zero thrust until about Mach 1, to make them act like a ramjet / scramjet kind of thing.

It's the 48-7s that annoys me. Dozens of engines to choose from but the only two choices worth exploring in most cases are the 48-7s and the LV-N.

[EdFred]

Aye, but the important thing in a game like this is balance. You need a reason to use all the different engines in the game. If you're going to have something that's clearly superior across the board, it should be either more expensive or unlocked later in the tech tree.

Ideally, there should be no engine that's superior across the board.

Best,

-Slashy

I've pretty much only been futzing around in sandbox as of late. I didn't realize/forgot the -7S was available so early. It should really be the last engine available if they leave it as is, maybe its own node after everything else has been unlocked. OR make it only usable on Eve.

Edited February 1, 2015 by EdFred

[Brotoro]

I don't mind the LV-N...it is already nerfed quite a bit compared to a realistic nuclear engine. But I don't use it on landers and planes because of self-imposed rules -- such an engine would be spewing deadly amounts of gamma and neutron emissions when operating (and for a while afterwards) unless you keep it at a sufficient distance and behind some shielding.

[Technical Ben]

Can we look at the obvious though? It's the LV-N. I use it for literally everything outside the atmospheres of Kerbin, Eve, and Laythe. It's totally necessary, and I would by no means nerf the ISP because it makes the game playable, but it needs both bigger drawbacks and closer competitors so there is some decision making required besides "how many LV-Ns do I strap to this thing?". I think it should cost more and overheat nearly instantly after coming into contact with the thinnest atmosphere. I also would love a 2.5m variant that has slightly better ISP, but lower TWR and weighs close to 10t, making it great for big payload transfers but terrible for landers. Another larger electric engine could be great too, for instance a 1.25m VASMIR that runs on xenon for large solar sailers.

Heat management. That would make it less "answer to all problems". But remember, IRL it would be the interplanetary choice every time, except where size is a problem or when ion engines can do the job. It's one of the "suppose to be OP" situations along with the heavy lifter engines.

[m4v]

nuclear engines could have some spooling for make it impractical for lander use without getting into simulating the deadly radiation.

[Renegrade]

They are if you're using the same engine for both the transfer and the landing. Its hard for me to see why I'd use LV-Ns to cart a less efficient engine to Duna. Or to Dres. Or Vall. Or Moho. Or pretty much anywhere but Eve or Laythe.

Or Tylo.

However, I'd like to point out that a basic vacuum landing payload for hitting all the surface science in a biome in stock is something like 950kg. Bringing the entire 2250kg nuke down to the surface and back is putting a big bump in your mass cost.

For example, let's say you're going to Dres (ew). The current version of the charts say about 430 dv, which is just a touch higher than my own calculations. We'll assume you take the most efficient route to get to a 12km low-Dres-orbit. Nukes, ions, kraken drive, whatever. To land on Dres and return, you'll need 860 dv. TWR isn't likely going to be an issue with any of the engines I've selected below, so we'll only need one.

LV-N

- mass: 950 + 2250 = 3200kg (3.2t)

- Ve: 7856

48-7S

- mass: 950 + 100 = 1050kg (1.05t)

- Ve: 3437

LV-909

- mass: 950 + 500 = 1450kg (1.45t)

- Ve: 3829.8

(Ve is Isp * 9.82 in KSP due to a Squad derp)

The LV-N will require 370 kg of fuel, the 48-7S will require 299kg, and the LV-909 will require 365kg per landing. In all cases, the LV-N is worse. If you use two LV-Ns or more in your transfer stage and land both, the results become catastrophic for it (630kg of fuel for two, 891kg for three, 1151kg for four).

Note that the LV-N transfer stage would have to detach ONLY the science payload, engine(s), and the landing fuel. Taking the return-to-Kerbin fuel or any other mass will increase that 370kg requirement.

(I've left out tankage overhead for simplicity. The single-engine numbers are close enough that it wouldn't matter much)

Regarding the aerospike being underpowered, I think it should just be made possible to put a decoupler under it, like with nearly all the other engines in the game. On paper it's not that bad, it's like a more powerful 909 that's also efficient in atmosphere, but not being able to take a decoupler makes it just plain awkward to use.

I'd have to re-consider it under the new aero, but with the old aero, since mass=drag, the aerospike ends up being really terrible in air :/

[Brotoro]

nuclear engines could have some spooling for make it impractical for lander use without getting into simulating the deadly radiation.

That would certainly make them more interesting to use.

[Cripion101]

I don't know if this really counts 'cos it isn't stock, but the DT Vista Inertial Fusion engine from KSP interstellar is ridiculous.

[Pthigrivi]

Renegrade I may be misunderstanding you but isn't the dead-weight of the extra engines themselves more than what you're saving in fuel? Take the 909s which aren't likely to take the nerf the 7S will. 2 LV-Ns cost you 630kg in fuel, which isn't drastically more than the 909, and is certainly less than the 1t weight of the 909's themselves. If you use the 2 LVNs for both transfer and landing you don't need to bring the 909's along to begin with. Thats extra weight you have to push into orbit and through your transfer stage. Just leave your return fuel in orbit. It may cost more for the landing itself, but the total mission weight is still much less.

Am I not seeing this right?

[GoSlash27]

Renegrade I may be misunderstanding you but isn't the dead-weight of the extra engines themselves more than what you're saving in fuel? Take the 909s which aren't likely to take the nerf the 7S will. 2 LV-Ns cost you 630kg in fuel, which isn't drastically more than the 909, and is certainly less than the 1t weight of the 909's themselves. If you use the 2 LVNs for both transfer and landing you don't need to bring the 909's along to begin with. Thats extra weight you have to push into orbit and through your transfer stage. Just leave your return fuel in orbit. It may cost more for the landing itself, but the total mission weight is still much less.

Am I not seeing this right?

Let's look at a hypothetical trip to Eeloo. Nice and far, and Mun sized target.

Using a dedicated launcher and assuming a 1 tonne payload, the ant produces a package totaling 1.32 tonnes.

Landing that 1.32 tonnes will create a vehicle that weighs 1.77 tonnes.

So your LV-N would need to cart 1.77 tonnes to Eeloo and 1.1 tonnes home.

Getting home from Eeloo is 2700 m/sec and getting there is 3,700 (worst case). Splitting the difference on the payload yields a mission package of 9.9 tonnes.

If you use the LV-N to execute the entire mission, than it's a 1 tonne payload for 7,650 m/sec is a 10.8 tonne package.

Leaving 3 tonnes of fuel in orbit won't save you a tonne of fuel over the landing and launch at Eeloo, so it works out cheaper to have a daughter lander.

This loss is compounded as you visit more planets in a single trip, and of course the demand on launch vehicles increases linearly with mass.

Best,

-Slashy

[Kerbin Dallas Multipass]

I dislike the LV-N from a pure gameplay perspective. I almost never use it because I dislike how it affects gameplay.

-Sitting through 38 minute burns is just boring. Not my definition of playing a game.

-Always having spare dV do do practically any mid-course correction is not very challenging.

-Landers that are unlikely to run out of fuel and can basically hover cross country from biome to biome - well, slamming down the capsule with the last drop of fuel Armstrong style is more exciting. And one more reason to build a cool rover.

-Using the same engine for most everything beyond LKO is a bit boring as well.

nuclear engines could have some spooling for make it impractical for lander use without getting into simulating the deadly radiation.

I think this is a really good idea. Nuke engines should take significant amounts of time to spool up, then have a really powerful thrust and then again take long to spool down. A burn time of a few minutes would catapult your ship on any trajectory, just not very precisely. Would require to take along extra chemical rockets for the precision work. The LVN would do the "heavy lifting" in terms of interplanetary dV and would only be good at that and nothing else. I could imagine this would make the game better.

[Torquemadus]

I play on hardcore custom difficulty. I've been getting by with the engines from the tier 1 R&D building. I use the LV-909 for my crewed spacecraft and the 48-7S for my satellites. The LVT-30 and LVT-45 power my fly-back boosters, while the SRB-KD25k is used as a cheap first stage. In the early game, I used the much less capable BACC booster.

Two-stage to orbit launch vehicles are not fashionable in KSP at the moment, because the mighty Turbojet rules supreme. My designs throw away "trash bins full of boom" solids as a first stage, and recover their liquid fuelled second stage to the runway. The crewed upper stage also recovers to the runway, while the satellites can be recovered via parachute. They are more than adequate for the economic demands of the extreme difficulty settings I use. My launch costs consist of a bunch of cheap SRBs, and a small cost for the liquid propellants used by the other stages. No parts other than empty solids are expended. Nonetheless, the launch costs of my Kerboosters are pathetically inefficient compared to Turbojet SSTOs.

It's worth bearing in mind that proposed IRL SSTO concepts are only economical because of the low operational costs involved. The NASA Space Shuttle offered far higher payload to orbit costs than the expendables it should have replaced. This was mostly due to the cost of paying the standing army of workers needed to prepare the re-useable shuttle for each launch. The enormously high dry mass of the orbiter didn't help either. SSTOs try to reduce costs by offering rapid turn-around times with a small ground crew, allowing the launcher to spread costs over a very large launch manifest. IRL SSTOs need to have extremely low dry mass to have any hope of reaching orbit, which leaves very little margin for cargo. This means that they either need to launch very small satellites, or deliver small but very frequent service to orbiting stations to deliver parts and consumables for on-orbit assembly or other operations.

The economics of launches in KSP don't take ground crew costs into account. We don't have to pay our workers. An enormously large and expensive Turbojet SSTO will give excellent launch costs, because it doesn't really cost anything at all. Provided it recovers to the runway, it costs nothing. Only the propellants are paid for, and most of the propellants were sucked in through the air intakes! Since air-hogged turbojets can place payloads into orbit with only a little help from RCS, Turbojet SSTOs are almost a free lunch!

[Renegrade]

Renegrade I may be misunderstanding you but isn't the dead-weight of the extra engines themselves more than what you're saving in fuel? Take the 909s which aren't likely to take the nerf the 7S will. 2 LV-Ns cost you 630kg in fuel, which isn't drastically more than the 909, and is certainly less than the 1t weight of the 909's themselves. If you use the 2 LVNs for both transfer and landing you don't need to bring the 909's along to begin with. Thats extra weight you have to push into orbit and through your transfer stage. Just leave your return fuel in orbit. It may cost more for the landing itself, but the total mission weight is still much less.

Am I not seeing this right?

You are, more or less, but consider the following:

- You only need one drive to land on Dres, regardless of the type (unless you're bringing a LOT more non-drive mass with you)

- You probably want multiple LV-Ns for interplanetary (I like two to four myself when I use 'em)

- The 48-7S is already ahead even for a single landing, and the single 909 is ahead of the dual-LV-N overall.

- 48-7S nerfs can take different forms: Isp, or mass, which would affect this comparison, or thrust, which would not (unless it had less thrust than say an ant~)

- Multiple landings will eventually favor the 909 over the LV-N(x1) too

You could build an LV-N ship that had say, three LV-N drives, one of which was attached to the lander, the other two to some sort of framework ahead of the lander, undock the lander, and land/return as required and have a 1x LV-N lander with a 3x LV-N interplanetary..but that's still falling behind the 48 on each landing, and falling behind the 909 after .. well, many many landings.

Of course if you want to get into crazy designs, you can always just land the EAS seat thing, which will really throw the game towards the lighter drives...

Anyhow, here's the overview in a different layout. Remembering that I've assigned a value of 0.95t (950kg) to the basic craft (the lander can, a Science Jr and a goo pod).

LV-N (x1)

Payload: 0.95t

Engine Mass: 2.25t

Fuel required for 860 dv: 0.37t

Total: 0.95+2.25+0.37 = 3.57t

TWR: ~14.87 Dres local

LV-N (x2)

Payload: 0.95t

Engine Mass: 2.25t x2 (4.5t)

Fuel required for 860 dv: 0.63t

Total: 0.95+4.5+0.63 = 6.08t

TWR: ~17.47 Dres

LV-909

Payload: 0.95t

Engine Mass: 0.5t

Fuel required for 860 dv: 0.365t

Total: 0.95+0.5+0.365 = 1.815t

TWR: ~27.55 Dres

48-7S

Payload: 0.95t

Engine Mass: 0.1t

Fuel required for 860 dv: 0.299t

Total: 0.95+0.1+0.299 = 1.349t

TWR: ~19.68 Dres

For the record, the last time I landed on Dres, it only had one biome and I had no intention of returning, so I went 'meh' and just made a single-LV-N lander that had no orbital portion. Was less efficient, but was easier to click together in the VAB. Just because something isn't the most efficient course doesn't mean you can't take it anyways. For instance, I sometimes use Skippers for moderate size interplanetary missions just to cut down on burn time

[Jouni]

Let's look at a hypothetical trip to Eeloo. Nice and far, and Mun sized target.

Using a dedicated launcher and assuming a 1 tonne payload, the ant produces a package totaling 1.32 tonnes.

Dedicated lander engines are obviously a better choice for small payloads. The situation changes, if you consider multi-kerbal landers with 3-5 tonnes of payload, or large landers carrying rovers, surface bases, and other stuff.

[Brotoro]

The nuke shouldn't have a significant "spool-up" time, but it should have a significant "spool-down" time. The reason being because when you slam in the control rods (well...rotate the control drums in a Nerva design) you need to continue to run propellant through the reactor core as it cools down.

This wouldn't be a terrible burden, since you just need to watch the delta-V remaining on your maneuver node and cut off the engine early. But it would make it impractical to use the nuke as a lander engine.

Somebody should make a mod so I can play with the extra challenge. And liquid hydrogen tanks so I can feed the nuke with a proper propellant.

[EdFred]

I dislike the LV-N from a pure gameplay perspective. I almost never use it because I dislike how it affects gameplay.

-Sitting through 38 minute burns is just boring. Not my definition of playing a game.

You're doing it wrong.

There should never be a 38 minute burn.

[tater]

The nuke shouldn't have a significant "spool-up" time, but it should have a significant "spool-down" time. The reason being because when you slam in the control rods (well...rotate the control drums in a Nerva design) you need to continue to run propellant through the reactor core as it cools down.

This wouldn't be a terrible burden, since you just need to watch the delta-V remaining on your maneuver node and cut off the engine early. But it would make it impractical to use the nuke as a lander engine.

Somebody should make a mod so I can play with the extra challenge. And liquid hydrogen tanks so I can feed the nuke with a proper propellant.

This makes a lot of sense. I've never made a nuke lander, myself because I don't see them as lander engines.

[problemecium]

I have a nomination for the opposite - since the nerf, I've found the Kerbodyne LFB to be useless. Before, it was superior to the Mainsail in terms of Isp, but now I haven't found a reason to use one in place of a Mainsail + Jumbo-64 tank.

The Poodle has a similar problem. I actually saw a mod wherein it was converted to a Size 0 nuclear engine, and I thought that was a really neat idea, so something similar like giving it a higher Isp to make it like a big nuclear engine would be good in my opinion.