MH Engine Rebalances for 1.6

[klesh]

Quote

+++ Parts
* The following LFO engines were rebalanced : Wolfhound, Cheetah, Kodiak, Mastodon, Cub, Skiff, Bobcat.

 

Let's have a look and argue like nerds over this, shall we?  I got the pre-1.6 stats from here.  I read elsewhere that some have been moved around in the tech tree.

 

Wolfhound

Cost: from 1,680 to 3,000 +1,320

Mass: from 2.5t to 3.3t   +0.8t

ISP:  from 70asl/412vac to 70asl/380vac   -32vac

Thrust: from 63asl/375vac to 69asl/375vac +7asl

 

Cheetah

Cost: from 1,000 to 850 -150

Mass: Unchanged

ISP: from 150asl/345vac to 150asl/355vac +10vac

Thrust: from 54asl/125vac to 52asl/125vac -2asl

 

Kodiak

Cost: from 1,300 to 1,100 -200

Mass: Unchanged

ISP: from 265asl/305vac to 285asl/300vac +20asl/-5vac

Thrust: from 208asl/240vac to 247asl/260vac +39asl/+20vac

 

Mastadon

Cost: from 22,000 to 8,000 -14,000

Mass: Unchanged

ISP: from 280asl/290vac to 290asl/305vac +10asl/+15vac

Thrust: from 1,303asl/1,350vac to  1,283asl/1,350vac -20asl

 

Cub

Cost: from 1,000 to 800 -200

Mass: Unchanged

ISP: from 270asl/320vac to 280asl/310vac +10asl/-10vac

Thrust: from 33asl/40vac to 28asl/32vac -5asl/-8vac

 

Skiff

Cost: from 1,500 to 2,300 +800

Mass: from 2t to 1.6t -0.4t

ISP: Unchanged

Thrust: Unchanged

 

Bobcat

Cost: Unchanged

Mass: Unchanged

ISP:  Unchanged

Thrust: Unchanged

 

 

Things of note to me:

Big cost changes, especially in Mastadon and Wolfhound.

Slight Wolfhound vac ISP nerf.

Strange Mastadon asl nerf

 

Do these changes satisfy you?  What are your thoughts on them?

[GluttonyReaper]

Still updating everything, so haven't had a chance to go over everything yet, but the two most interesting changes for me are the Kodiak and the Skiff.

Skiff appears to have had a weight increase (I think you may have made a slight mistake there), which fixes the issue where it outclassed the LV-T engines in literally every department, and makes it an inferior 1.25m launch engine, which is neat. But unfortunately the buff on the Kodiak kinda negates this - it has a slightly lower vacuum ISP, but otherwise it's basically just a Reliant but better.

As you've mentioned, the Mastadon change is a little odd. It's now got slightly higher ASL ISP, but it's thrust is now even lower than the Mainsail. Cost is actually significantly lower than the Mainsail, though, so that might've been what they were going for - a cheaper, less thrust-y version, that most people will replace with the Mainsail once they unlock it. Cheetah engine is more or less the same (not sure why they chose 355 vacuum ISP, though. Terrier and Poodle are both 350), as is the Bobcat, which makes sense considering those were totally fine before.

Wolfhound has been nerfed as expected, so I'm guessing it'll play like a 'heavy' Poodle now, for really big crafts. It'll be interesting to see how that works.

[klesh]

11 hours ago, GluttonyReaper said:

Still updating everything, so haven't had a chance to go over everything yet, but the two most interesting changes for me are the Kodiak and the Skiff.

Skiff appears to have had a weight increase (I think you may have made a slight mistake there), which fixes the issue where it outclassed the LV-T engines in literally every department, and makes it an inferior 1.25m launch engine, which is neat. But unfortunately the buff on the Kodiak kinda negates this - it has a slightly lower vacuum ISP, but otherwise it's basically just a Reliant but better.

I'm seeing the Skiff as 1.6t ingame (v1.6), and the wiki currently shows the old stats, which shows it at 2t.  To me, that is 0.4t lighter, no?

 

 

It seems a fair amount of this balance pass had to do with the engines' position in the tech tree and cost vis a vis the stock engines.  I agree with your Mastodon take, which gives it a home in the career game. Indeed, the Wolfhound has been moved to the back of the tech tree, been made more expensive.  Late-game heavy vac engine for sure.  That makes sense to me as well.  The version of the AJ-10 that Apollo used, the one this Wolfhound is based on, was intended to be just that; a heavy lander engine for direct landing and ascent.

I'm actually very satisfied that the Poodle has a place again, and just in time for its glamour makeover.

[Snark]

Incidentally, they just posted a devblog with all of this info nicely summarized in a table. 

 

 

[klesh]

2 hours ago, Snark said:

Incidentally, they just posted a devblog with all of this info nicely summarized in a table. 

 

 

Ahh, very nice.  I tried using the table feature here on the forums, but it was a mess to try to work with.

[petlahk]

Coming from my forum post over there.

So. I do understand that maybe 412 ISP is a bit too high. I'm still sort of grumbly about it - as I said there because I have to re-do all my rockets Delta-V balances (AGAIN). Though. I wonder if it may have been better to just increase the mass, cost, and massively decrease the thrust. That way it's still a really good vacuum optimized engine, but not so great at lifting massive spaceplanes up into orbit... I'm inclined to say that I would have preferred they touch literally any part of that engine *except* the Delta-V.

Granted, Partly because I used it a bunch and don't wanna do that work again, but also because it was nice having an engine that sat so squarely between the rest of the engines and the NERV. 

Addendum: I'm probably just gonna have to make it work though. 'Tis the Kerbal way. Sorta forced to update because the update actually runs smoothly. Heh. I guess that does outweigh my grumbles.

Edited December 21, 2018 by petlahk

[Superfluous J]

15 minutes ago, klesh said:

Ahh, very nice.  I tried using the table feature here on the forums, but it was a mess to try to work with.

I find it's easier to make a table in html, then load that html in a web browser and copy/paste the table into the forums. You can even make them on other forums and paste them here, though I prefer the former.

If you don't know, you can do it all in a local text file and load that in your browser by just putting the local path into the location bar.

[Maxsimal]

32 minutes ago, klesh said:

Ahh, very nice.  I tried using the table feature here on the forums, but it was a mess to try to work with.

Cutting and pasting a table from a google sheet also works pretty well, FYI   But not from excel or word.

Edited December 21, 2018 by Maxsimal

[klesh]

html, google sheets... you guys are my kinda nerds.

[Tyko]

Here's the page I've been using for my Engine Tweaks for MH Mod. The new stats from 1.6 have been added. ( I hope I got them all correct)

https://docs.google.com/spreadsheets/d/1tzovHSBZm_4_xLrglqAKbmLzeGQozpZfQr2YAaOTs8k/edit#gid=0

 

[petlahk]

The terrier no longer has integrated fuel tanks...

[Aegolius13]

My quick thoughts on the new engines -- kind of a follow-up to my old MH Engine Round-Up linked below:

Cub: After nerfs to thrust and vacuum ISP, the Cub is drawing closer to the  Spark in performance, but is still generally a better option as long as you need the extra thrust.  The main disadvantage continues to be that the Cub only gimbals in one direction, so you really need at least four of them for effective attitude control.  I feel like radial engines are only used situationally anyway (e.g., in a lander or a ship that needs the back end free for docking), and for those uses this remains just fine.  Frankly, this engine was just too good in vaccum for something that was to be used as a vernier for the Kodiak or Reliant at launch.

Cheetah: The Cheetah used to be "two Terriers stapled together," was was generally an afterthought unless you needed something to fit a 1.875m stack.  Now, however, its ISP is second only to the Wolfhound in terms of chemical engines, and thrust has received a nice upgrade as well.  So the new Cheetah looks like a very attractive option for anything in that gap between Terrier and Poodle/Wolfhound sizes.  It does not appear worthwhile to cluster Cheetahs, however -- the Poodle's superior TWR vs a pair of Cheetahs should outweigh the minor gap in ISP.

Kodiak: The old Kodiak was practically indistinguishable from the Reliant, which made especially little sense given how far up the tech tree the new engine was found.  Now, there is a pretty clear differentiation: the Kodiak is a superior launch engine, with better ISP and thrust than the Reliant for the same cost.  The Reliant's only relative advantage is marginally better vacuum ISP, but that's not terribly important for what is still basically a launch engine.  In other words, the Reliant, somewhat like its sibling the Swivel, is now a bit of a "tweener" that is unlikely to be your best option for either a launch or a later stage.  

Skiff: The Skiff is/was another "tweener" engine, suited for a J-2-style sustainer role that makes more sense in real life than in KSP.  Prior to the update, this was more than made up for by its absurdly high TWR (tops in the game if I remember correctly).  Predictably, the Skiff's mass was brought up in the update to bring TWR closer in light with the other engines.  I worry that, between the unimportance of sustainer engines and this model's low thrust relative to its surface area on a rocket stack, it might not see a lot of use.  More than ever, I think it's best viewed as an upgrade to the Swivel, and kind of a "jack of all trades, master of none" option.  It could still be pretty good as, say, the core stage of a rocket in conjunction with some kind of radial booster.  

Bobcat: Of all the MH engines, the Bobcat seemed like the one that made the most sense from a design perspective, as it had a clear niche without being imbalanced.  Squad seemed to agree, as the changes made in 1.6 were minimal.  This remains a great launch engine for anything on a 1.875 meter stack, but works especially for radial boosters.  

Wolfhound: This seemed to be the most controversial engine (or probably part of any kind) in Making History, due to its absurdly high vacuum-ISP.  It not only spawned a million arguments about accidental stat swaps with the Skiff, but it was just way too got.  So not surprisingly, we see the ISP brought down to a (still very robust) 380, and a mass hit as well. I think these changes bring the engine in line with what we expected/wanted -- a low TWR, high ISP vacuum workhorse that bridges the gap between the Poodle and NERV.  

Mastodon: The community seemed to agree that the Mastodon was much too expensive in career mode, and generally upstaged by the more efficient and powerful Mainsail.  Squad responded by cutting the cost to a figure that puts it right between the Skipper and Mainsail, and by giving a nice boost to ISP.  As a result, the Mastodon now trails only the Vector and Mammoth in terms of launch efficiency.  So it's very well-suited as a core launch engine for anything a little too small to make sense for a Mainsail or Twin-Boar.   I still feel like the TwinBoar will make a better radial booster engine for most applications, due to its superb thrust and very reasonable cost.  Likewise, I'm not convinced that clustering Mastodons, a la the Saturn V, will beat other options like the Mammoth (or just MOAR TWIN-BOAR BOOSTERS).  But now it's nice that the game's analog to the legendary F-1 is now at least worth considering.  

Overall thoughts: I almost did a post on possible tweaks to rebalance these engines, it it came relatively close to what we see here.  I think Squad has done a pretty good job of making these engines closer to their real-world analogs in terms of use cases, but having them still make sense in the game and in balance in comparison to the vanilla KSP engine offerings.  As noted above, these changes have the potential to relegate some new engines to afterthought status, but in general I think this is a very positive step.  

 

My original (and now sadly obsolete) engine review:

 

Edited December 22, 2018 by Aegolius13

[fourfa]

Excellent review, thanks!

[Waxing_Kibbous]

I finally ended up buying the DLC because of this, I have been on the fence because I didn't want OP parts for the Kerbal system, but the fact that things were given a second look and tweaked cemented the decision for me. Now to play with them!

[theonegalen]

My only complaint about the rebalance is that the Mastodon is still no good for making Saturn V replicas, which was partially the stated purpose of Making History. Otherwise, I like the changes, and will integrate them into my own mods.

Edited December 23, 2018 by theonegalen

[GeneCash]

On 12/21/2018 at 2:17 PM, klesh said:

Ahh, very nice.  I tried using the table feature here on the forums, but it was a mess to try to work with.

FYI, I write a table in raw HTML in an editor, then copy/paste it from a browser. For me, it's a lot less painful.

[maccollo]

 

On 12/23/2018 at 4:08 PM, GeneCash said:

FYI, I write a table in raw HTML in an editor, then copy/paste it from a browser. For me, it's a lot less painful.

I'm not sure any engine would be good for a saturn V replica in stock. The problem is that, given the overall performance of stock parts, the delta V requirements for a mun round trip is to low, so a replica that looks good will be over-engineered for the task.
Here's a replica that can perform a mission to mun in an Apollo like mission plan. The proportions of the rocket is roughly correct, the initial TWR quite close to the real deal... but it's built for a 4X kerbol system where the ratio between max stage deltaV, and deltaV to orbit is a lot closer to real life than it is in stock.

This means you end up with fairly realistic rocket designs using stock parts. So a 3 stage booster to land 2 kerbals on the moon is actually quite optimal, and even though it is 4X scale there's enough margin to add science payloads to the lander.

 

[KerikBalm]

On 1/2/2019 at 11:16 PM, maccollo said:

I'm not sure any engine would be good for a saturn V replica in stock. The problem is that, given the overall performance of stock parts, the delta V requirements for a mun round trip is to low, so a replica that looks good will be over-engineered for the task.
Here's a replica that can perform a mission to mun in an Apollo like mission plan. The proportions of the rocket is roughly correct, the initial TWR quite close to the real deal... but it's built for a 4X kerbol system where the ratio between max stage deltaV, and deltaV to orbit is a lot closer to real life than it is in stock.

This means you end up with fairly realistic rocket designs using stock parts. So a 3 stage booster to land 2 kerbals on the moon is actually quite optimal, and even though it is 4X scale there's enough margin to add science payloads to the lander.

 

Indeed, I built a Saturn V replica for a 3x rescale, that did pretty well with the old parts.

I found the skiff a bit too weak: its stats were great on paper, but 5x Skiffs weren't going to do it for the 2nd stage, starting with a 5 meter tank. IMO, the skiff should have had its thrust doubled, and its mass tripled. That would give it proportionately similar stats to what it has now.

On 12/22/2018 at 3:46 AM, Aegolius13 said:

Skiff: The Skiff is/was another "tweener" engine, suited for a J-2-style sustainer role that makes more sense in real life than in KSP.  Prior to the update, this was more than made up for by its absurdly high TWR (tops in the game if I remember correctly).  Predictably, the Skiff's mass was brought up in the update to bring TWR closer in light with the other engines.  I worry that, between the unimportance of sustainer engines and this model's low thrust relative to its surface area on a rocket stack, it might not see a lot of use.  More than ever, I think it's best viewed as an upgrade to the Swivel, and kind of a "jack of all trades, master of none" option.  It could still be pretty good as, say, the core stage of a rocket in conjunction with some kind of radial booster. 

That is my concern as well. The skiff is too weak for clustered use on a larger diameter stack. I had to do the 8+1 clustering on my 3x scale Saturn V inspired Mun rocket.

I find similar issues with the Bobcat.

The cross section area is 2.25x as much, but the thrust is only (400/240.. comparing to a reliant) 1.67x as much. They can't lift stacks of equivalent height as a reliant can... making 1.875m rockets too short.

Edited February 5, 2019 by KerikBalm

[Tyko]

@KerikBalm I just rebuilt my Saturn V using the new MH balances. I found 5 Skiffs to work quite well in a 2.5x scaled rocket with a 5m Stage 1 & 2. I thought it was going to be underpowered as well, but it performed really well. If anything S2 has too much power.

Here's a screenshot showing my tank layout and DV budget. The ASL TWR of Stage 1 is actually 1.33, so I could stack more fuel on both S1 and S2 if I needed to. Happy to go over more details if you want.

 

 

I agree on the Bobcat - I was able to build an Atlas/Gemini for 2.5x by increasing the thrust to 450. So it didn't take much to make it workable, but 400 was definitely too low.

Edited February 7, 2019 by Tyko

[SnakyLeVrai]

My only complaint so far is how it broke the missions I've already launched... I don't mind when it comes to designing new ones but, for those I've launched, it's really annoying, as I calculated the exact dV I needed, with a small percentage for maneuver errors and whatnot.

I've had to write a MM patch to unnerf them util my current missions finish all their maneuvers. That's quite annoying. Couldn't @SQUAD make the thing they did with obsolete decouplers and tanks, so current missions don't get sabotaged by nerfing? I understand the need for nerfing but, come on... this feels like a punishment for trying to engineer things properly rather than overengineering all the time...