Engine balancing issues in ARM?

[Rakaydos]

The first thing I thought when I started testing the new engines was "wow, it looks like Squad has taken another step towards moving the game away from the veterans and towards rookies." And if you take into account their business model, this should come as no surprise. I suspect this trend will continue. Squad is not making any money off of us grandfathered in players. In order for the veterans to still get a challenge out of the game, the veterans will have to either use mods or modify configs themselves.

Which is why my suggesstion, in strting the other topic, is to rebalance the engines using the SLS parts (and the equally OP rockomax probe engine) as the baseline. "If everything is equally broken, nothing is broken."

[Guest]

And what is wrong with that?

SLS is not the "White Tanooki Suit" of KSP. It will not fly itself into Orbit unless you use MechJeb. In the game mode they will likely be extremely expensive to use to make up for their slightly better stats.

There is not anything wrong with it. One of my favorite old saying's that I use almost on a daily basis is "do what you gotta do". Also, I should have said "veterans will likely have to use modifications" instead of "will have to". Hopefully, everything will balance out in the end.

I do agree with what Tiberion said, single stage to Duna is not good. Why not just turn on infinite fuel? To expand on that thought, the ability to come here to these forums and compare game play experiences is what keeps many players around for so long. When all of the sudden a new part is introduced that makes what used to be a tough achievement not so tough anymore, then it takes away the excitement of "hey, look what I did!".

[Kasuha]

There's one thing that makes Mainsails still useful. You can mount them below the new engine for even further increase of thrust (at a price of slight ISP decrease). It does not stand in the way of flames and it is very Kerbal

(source: Reddit)

Apart of that I'm ok with the game having some parts which are not used much. There are already many of these.

Edited April 3, 2014 by Kasuha

[rkman]

I've taken the time to graph it and propose some solutions:

I object to the proposition of making the thrust of the LFB-KRx2 and the KR-L2 virtually identical to the mainsail.

[rkman]

overpowered rockets doesn't help in any of that.

It does help in that with overpowered parts it is more likely that a randomly thrown-together rocket will be able to make orbit and then still have plenty fuel to go places. TWR and delta-v become almost non-issues.

But i also think it is the wrong way to go about it. TWR and delta-v are the nuts and bolts of rocketry, it seems odd to hide that aspect of rocketry in a rocket game.

If Squad wants to target non-geeks i'd suggest it would be much more effective to include rocket design tools such twr- and delta-v indicators, and a delta-v map.

Non-geeks are usually less technically inclined to begin with, and are definitely not OCD enough to just keep doing trial-and-error to design a rocket that will get them to the Mun.

I may not be the most hardcore armchair rocket scientist in the world but i am getting weary of designing blind and having to rely on whatever rocket design instincts that i may have developed after a few years of playing ksp.

Edited April 3, 2014 by rkman

[SpaceGremlin]

Right now the proposed nerfs are rather extreme. I could get behind increasing the KR-2L mass up to 9 tons though (TWR 27.8).

[Z-Man]

Kasuha: Using cubic octagonal struts, you can also cluster 24 mainsails around the S3-KS, maybe more if you disable snap. For when you need that little bit of extra thrust. No debug menu part clipping required, though of course there is plenty of clipping.

TWR*I_sp is not a sensible engine metric. The most sensible short one is I_sp*(1-TWR_desired/TWR), where TWR_desired is the desired total TWR of the vehicle. So about 2 for a launch engine and however low you want to go for space. That formula assumes you are building an infinite asparagus with perfect fuel tanks and shed engines as soon as you no longer need their thrust. Just a bit of nitpicking.

Personally, I think it is fine if designated huge stage 1 launch engines are strictly (but slightly) better than everything smaller. After all, their weight disqualifies them for most applications once in space. The new parts just highlight how shoddy the Mainsail is at its job The real launch engine efficiency comparison would be a 48-7S cluster, and they're not much better than that.

If I were calling the shots, I'd have given them a slightly lower vacuum I_sp, though, just to make absolutely clear they are for lifter stages only.

Edited April 6, 2014 by Z-Man
Oops, corrected nonsense formula. Nobody even noticed it claimed lower TWR is better before :(

[ArmchairGravy]

I believe it was the video interview they had with Manley before ARM launched. Point is. The main part of the game where they can reasonably say it is feature complete for version 1.

Nerfing parts while most still use sandbox and use it to do stuff instead of game mode is absurd. There is not enough "game" in KSP to warrant nerfs. Just people who want a reason to use the lifter they spent a week designing instead of SLS parts.

There are plenty of "realism" "uber-hard mode!" etc... mods and I am sure there will be mods to nerf SLS into oblivion. Install those.

+1. Exactly my thoughts on the whingeing over SLS being OP.

[Varses]

The new parts just highlight how shoddy the Mainsail is at its job The real launch engine efficiency comparison would be a 48-7S cluster, and they're not much better than that.

That point of comparison is kinda flawed because the 48-7s is considered pretty OP as well. Comparing overpowered engines to other overpowered engines misses the point. Again, I direct you to stupid_chris' graph near the beginning of this post. It is unequivocally clear that the new parts are far and above the power curve set by all previous parts.

I'm not arguing against the idea of bigger, stronger engines. KWRocketry, for example, has been doing 3.75m engines for a long time without them being stupidly overpowered. And that's because they have reasonable masses and isp ratios.

And it's not a matter of nostalgia, of wanting to "use the lifter they spent a week designing instead of SLS parts". It's the fact that now, there is pretty much no reason at all to use some engines because of the SLS parts. The Mainsail has become almost completely irrelevant because the SLS engines are the same or better in every way. The KR-2L has 66% more thrust than the Mainsail yet only 8.3% more mass. It also has an enormously higher ATM isp which makes it a completely superior replacement for the Mainsail since both engines occupy the 'heavy launch, first stage' niche and the KR-2L is better in every way.

By contrast, when the Mainsail was introduced it did not make LVT-30s and LVT-45s irrelevant because they still had clear advantages over the Mainsail. Yes, their TWR was worse but they had overall better isp in both atmosphere and vacuum. The Mainsail was also significantly heavier--it weighed 4x more and delivered 7x the thrust. There was a clear tradeoff.

Edited April 3, 2014 by Varses

[vger]

If Squad wants to target non-geeks i'd suggest it would be much more effective to include rocket design tools such twr- and delta-v indicators, and a delta-v map.

Non-geeks are usually less technically inclined to begin with, and are definitely not OCD enough to just keep doing trial-and-error to design a rocket that will get them to the Mun.

I may not be the most hardcore armchair rocket scientist in the world but i am getting weary of designing blind and having to rely on whatever rocket design instincts that i may have developed after a few years of playing ksp.

MechJeb has a deltaV calculator...

[mdosogne]

I don't think they should be that expensive. They should just unlock so late in the game that rookies won't have them outside the sandbox.

If we want the tech tree to mimic what happened in the real world, there could be three generations of rocket engines. (I'm considering only the engines used in launch stages now.)

1. Space race: Budgets are big, payloads are even bigger, and efficiency isn't that important. First there is the LV-T30, then the LV-T45, then the Skipper, and finally the Mainsail. The Mainsail is the ultimate engine: powerful, hungry, and expensive.
   
2. Efficiency first: Payloads become smaller, as the goal is no longer to beat the enemy, but to do something useful in space. We could see cheaper and more efficient replacements for the LV-T30, the LV-T45, and the Skipper.
   
3. Near future: Big payloads become interesting again for some obscure reason. Things are more efficient now than during the space race, with the SLS engines being bigger, cheaper, and more efficient than the Mainsail.
   

I agree: there should be different generations of rockets. This can be 'balanced' by requiring loads of research to unlock a new generation. I don't think the SLS should become cheaper, however...

[Z-Man]

That point of comparison is kinda flawed because the 48-7s is considered pretty OP as well. Comparing overpowered engines to other overpowered engines misses the point. Again, I direct you to stupid_chris' graph near the beginning of this post. It is unequivocally clear that the new parts are far and above the power curve set by all previous parts.

If I read the excel file correctly, the curve is an auto-fit power law without significance. The graph and formula used give too much weight to TWR. Twice the TWR does not make a twice as useful engine.

Yes, the 48-7s is a bit overpowered in the lifter/lander department. Still, it's a valid comparison point for new engines simply by the fact that it has been that way for a couple of updates. Either it's intended to be this way or the devs intend to make a global rebalance pass sometime in the future, probably only after the economic aspects of the game are in place.

[vger]

I agree: there should be different generations of rockets. This can be 'balanced' by requiring loads of research to unlock a new generation. I don't think the SLS should become cheaper, however...

It could conceivably be treated as an end-game rocket.The point in career mode where you're essentially a master of space anyhow and you get a few 'super parts.'

Unfortunately, as far as prices go, reality is somewhat counter-intuitive to how 'arms races' work in game mechanics. Typically in games, you gain access to more expensive gear, at a rate relatively parallel to the increasing payouts from your missions. In reality, tech tends to become gradually better AND cheaper. Just look at computers.

[Varses]

I don't disagree. I'm just saying that it's quite clear they are overpowered. Whether that's something that needs to be changed soon or in the far future is a different discussion but I think it's pretty clear they need to be changed at some point.

[a2soup]

The graph and formula used give too much weight to TWR. Twice the TWR does not make a twice as useful engine.

Don't underestimate the importance of TWR for launch engines. High TWR has two effects:

1. The rocket can clear the "boost phase" through the low atmosphere quicker and gravity turn more aggressively without losing vertical velocity. These effects significantly reduce the amount of ∆v spent fighting gravity, although they are mostly relevant only in atmospheric launches.

2. The engine contributes less mass to the rocket, increasing the possible payload fraction. All else being equal (i.e. fuel mass, total thrust, Isp), a higher TWR engine will increase the possible payload of the rocket. This effect is relevant in both atmospheric and vacuum launches, which is a big reason why the 48-7S is the go-to lander engine these days even though the LV-909 has a better ∆v.

Edited April 3, 2014 by a2soup

[Z-Man]

True to some extent, still: Twice the TWR does not make a twice as useful engine. The best lander engine depends on where you want to land. For Tylo, sure, it's the 48-7S. For Mun, no. Purely by stats, it's the LV-N, actually. I think. Don't have my spreadsheet here. But if the total TWR you aim for goes below 1 or so, the 48-7S stops being attractive and the LV family takes over. Do the proper math and you'll see.

[a2soup]

This math depends on the mass of your lander. You need a pretty heavy lander for the LV-N to be better than the 48-7S on the Mun. I know this just from designing landers using Engineer Redux.

Someone posted these (somewhat out-of-date) graphs in another thread: http://imgur.com/a/iNqmQ#0. The inclusion of SRBs is a bit confusing, but the basic idea comes through-- for lower total mass, lower engine mass trumps higher ∆v.

[m4v]

True to some extent, still: Twice the TWR does not make a twice as useful engine. The best lander engine depends on where you want to land. For Tylo, sure, it's the 48-7S. For Mun, no. Purely by stats, it's the LV-N, actually. I think. Don't have my spreadsheet here. But if the total TWR you aim for goes below 1 or so, the 48-7S stops being attractive and the LV family takes over. Do the proper math and you'll see.

well, maybe one 48-7S stops being attractive, but then you can have two, or three, as many as you need for reach that TWR.

[Z-Man]

a2soup: Yeah, of course for lower mass landers a lower mass engine would win, largely independently of small I_sp and TWR differences.

m4v: No. For lower required total TWR, clusters of 48-7S also become less attractive. That's the point. And the whole trouble of the 48-7S was that clusters of it were better than anything else for launches from Kerbin. Well, not any more.

[ArmchairGravy]

One thing being left out of the conversation is part count. I'm not going to use a cluster of LV-45's and tanks and struts when I can use one Skipper and one tank. Same with the SLS. I got the same performance out of 30 SLS parts and struts as I would have gotten out of my 70-part onion booster set with its tanks, engines, fuel lines, and struts. I'm running an older PC with not a lot of RAM, so reduced part count is a very welcome thing.

[stupid_chris]

I object to the proposition of making the thrust of the LFB-KRx2 and the KR-L2 virtually identical to the mainsail.

I'll say it again: the trust means nothing. It's a matter of TWR, not thrust. An engine of 1000000kN of thrust but of a mass of 10000t is not going to get you anywhere further than the mainsail.

[rkman]

trust means nothing.

A larger payload requires more fuel and more per-stage thrust to get sufficient per-stage twr to get into orbit. Aside from outliers such as the LV-N, engine mass is usually small relative to fuel mass. Relative engine mass (engine twr) is not the only thing that matters.

Per-engine thrust would not matter much if we could fit an arbitrary number of engines under a fuel tank, but we can't. A certain amount of thrust needs to be fitted in a certain amount of space, only so many engines fit there, so per-engine thrust is important.

[Slouch]

There's one thing that makes Mainsails still useful. You can mount them below the new engine for even further increase of thrust (at a price of slight ISP decrease). It does not stand in the way of flames and it is very Kerbal

(source: Reddit)

http://i.imgur.com/EYJjreg.png

Apart of that I'm ok with the game having some parts which are not used much. There are already many of these.

^^This

I think the new engines are fine, maybe you could put them later in the tech tree or have them cost more in career. But just like other updates that have introduced new parts this update has just changed how the mainsail will/can be used. The mainsail might not be used with the iconic orange tank anymore, as a heavy lifter engine, but it still could be useful with smaller tanks in a rockets higher stage. You might just need lots of power, very quickly, such as an emergency redirection or just for the lulz, the mainsail is this in spades, one great big umph of thrust.

The mainsail's not dead, it's just moved job.

[Red Iron Crown]

It is amazing to me how many people don't seem to understand stupid-chris' excellent graph, and assume that by balancing them he means the thrust of the new engines should be reduced.

Let me see if I can clarify:

• In KSP, TWR is generally balanced against Isp for gameplay reasons.
  
• The new engines, particularly the KR-2L, break that balance. This is mathematically factual, not opinion.
  
• To balance them, a thrust reduction, Isp reduction, or mass increase is required (or some combination of those). Thrust need not be reduced if that is considered undesirable.
  
• There is no reason an engine with very high TWR cannot be balanced against the old engines by lowering its Isp to an appropriate level.
  
• If the new engines were balanced, they would still reduce the part count on large lifters. Larger fuel tanks and engines with greater thrust mean fewer parts whether they are balanced well or not.
  
• Because they are unbalanced, the new engines relegate almost all the old engines to niche-use status for lifters. Basically, there is no logical reason to use one of the old 2.5m lifter engines; once your payload is large enough to justify 2.5m parts the new engines simply outclass the old in every way.
  
• Other engines in KSP have been nerfed compared to their real world counterparts (i.e. the LV-N's ridiculously low TWR and atmo Isp compared to the real NERVA) for gameplay reasons, there is no reason that SLS parts should be any different.
  

[Z-Man]

It is amazing to me how many people don't seem to understand stupid-chris' excellent graph,

I don't object the use of graphs in general or the general principle that high TWR should go with low I_sp, at least for engines of similar mass and size.

What I object to is the use of a power law for it. That is simply factually wrong.