[1.1] RLA Stockalike 13.4 [25 Apr]

[Yemo]

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

*snip

Great!

Though I actually liked the probe cores as a separate download.

[Sandworm]

Test Version 1 Download

Thanks! I've been waiting for this for a while. Will test tonight.

[Starbuckminsterfullerton]

Emissive looks great, I vote xenonparticle, although if it were me I would use a % of the newish stock purple/cyan one. I can't make custom fx though, so what you have is better.

[TicTacToe!]

Don't know if this has already been noticed, but all your probe cores I assume are meant to be named according to the rule "probe type"-G, I notice this is missing in the case of the OKTO core.

[hoojiwana]

Great!

Though I actually liked the probe cores as a separate download.

Yeah that was the original plan, but it never went anywhere after 10 months now, so I thought since I wasn't working on "Probe Parts" I would give them niches and include them here. The folder layout and texturing is set up so that people can easily delete the ones they don't want though.

Thanks! I've been waiting for this for a while. Will test tonight.

Cheers, looking forward to your feedback.

Emissive looks great, I vote xenonparticle, although if it were me I would use a % of the newish stock purple/cyan one. I can't make custom fx though, so what you have is better.

The smoother xenonparticle is what I was leaning towards as well, though it is very faint looking with that colour. Next version might have a custom FX included for better feedback, though that won't be out for a day or two to let you guys and gals try things out.

Don't know if this has already been noticed, but all your probe cores I assume are meant to be named according to the rule "probe type"-G, I notice this is missing in the case of the OKTO core.

This is exactly why this test version is here! Thanks a lot.

[sp1989]

Dont forget the trench smoke.

[Sandworm]

The new probe cores look great.

The tiny monoprop engines are great for fine tuning landings.

And the whole art package fits very well with the other stockalikes. There are at least seven parts mods in this pic and RLA fits right in.

Edited July 23, 2015 by Sandworm

[Ruedii]

Thanks for the test release.

Are you going to eventually convert the textures to DDS?

DDS is usually preferable to MBM on KSP now.

[Initar]

Hurray for the new RLA! Thanks Hoojiwana!

Are you going to eventually convert the textures to DDS?

I second this. So many great mods, so little memory...

Tested it a bit, the SRBs seem a little OP. They burn in a few second at full power!

I'll keep testing. Can't wait for the 0.625 LFO tanks and the spinnaker!

Edit: Also, isn't the small LVN supposed to be 0.625? 'cause its apparently not:

If its intentional disregard this comment ^^

Edited July 23, 2015 by Initar

[hoojiwana]

More Important: This is incompatible with any and all previous versions of RLA Stockalike and will break saves.

Less Important: Folder changes mean you must delete any previous installs, you will need to delete PartDatabase.cfg in your KSP directory as well. If you didn't have any installs on KSP 1.0.x then ignore this.

Big Changes

Added all the other parts previewed so far, removed the Xenon modes on the Arcjet and Resistojet since they were very hard to balance effectively, and did lots of fiddling around.

What needs testing:

Everything really, but special attention should be paid to:

• SRB balance (should be much better now)
  
• Probe core balance
  
• Electric engine balance (big changes here)
  
• 0.625m NTR balance
  
• RCS heating
  
• 0.625m LFO engine heat

• Added 5 revamped 0.625m LFO tanks
  
• Added revamped 0.625m high thrust engine
  
• Added revamped 0.625m vacuum engine
  
• Added new <0.625m LFO engine
  
• Added new tiny radial reaction wheel
  
• 
  
• Changed RCS folder to Control
  
• Fixed 0.625m nuclear engine scale (thanks Initar)
  
• Split old "Propulsion" category into Engine and FuelTank
  
• Set IKOTET thermal mass to the default
  
• Fixed OKTO gold probe flavour name (thanks TicTacToe!)
  
• Added balancing info to probe descriptions
  
• Named more engines (STEADLER MP engines were named in TV1)
  
• 
  
• Lowered TET probe SAS to level 0 (this is basic stability assist only)
  
• Increased 0.625m nuclear engine mass from 0.25 to 0.32
  
• Increased 0.625m nuclear engine heat production from 230 to 275
  
• Removed Arcjet Xenon+EC mode (leaving only Monopropellant+EC)
  
• Increased Arcjet mass from 0.2 to 0.23
  
• Decreased Arcjet thrust from 1.5 to 1.4
  
• Increased Arcjet vacuum ISP from 850 to 1050
  
• Removed Resistojet Xenon+EC mode (leaving only Monopropellant+EC)
  
• Decreased Resistojet mass from 0.25 to 0.21
  
• Increased Resistojet thrust from 3 to 3.5
  
• Increased Resistojet vacuum ISP from 650 to 875
  
• Decreased Electrostatic Ion mass from 0.25 to 0.2
  
• Decreased Electrostatic Ion vacuum ISP from 6300 to 5700
  
• Decreased 0.625m short SRB thrust from 52 to 40
  
• Decreased 0.625m long SRB thrust from 130 to 110
  
• Decreased 0.625m upperstage solid motor from 20 to 8

• Not all textures converted to DDS
  
• Tech tree placements not updated
  
• No engine has ModuleSurfaceFX added
  
• Decoupler drag is incorrect
  
• 0.625m nuclear engine FX is incorrect
  
• Some descriptions missing
  
• 0.625m nosecone heat config is incorrect

Dont forget the trench smoke.

It's on the to do list (also known as Known Issues). That's what ModuleSurfaceFX is. Thanks for the reminder though!

The new probe cores look great.

The tiny monoprop engines are great for fine tuning landings.

And the whole art package fits very well with the other stockalikes. There are at least seven parts mods in this pic and RLA fits right in.

While that probecore is shiny and gold, it isn't one of mine. Thanks for the other feedback though! The little monoprop engines are intended for how you're using them, someone suggested the radial version to use exactly that way.

Are you going to eventually convert the textures to DDS?

DDS is usually preferable to MBM on KSP now.

Yep, the new parts I'm adding (or touched up textures) have so far all been DDS, and the FX folder is mostly converted as well. The particle the monopropellant engines use however always turns purple when I convert it due to compression, so I'll have to look into fiddling with that. All the other left over MBMs will be converted soon enough, I won't do them in one go with a batch method because the converters I tried all severely degraded texture quality.

Tested it a bit, the SRBs seem a little OP. They burn in a few second at full power!

Edit: Also, isn't the small LVN supposed to be 0.625? 'cause its apparently not.

I barely looked at the new SRB numbers, so thanks for going over them. I've decreased the thrust on all of them but one by varying levels, they all burn a bit longer now. Still needs some tweaking so give them another try. The LV-Nc (now called Mighty!) was incorrectly scaled since I didn't tell the game to not scale it up. Fixed now!

Edited August 18, 2015 by hoojiwana

[DMagic]

Yep, the new parts I'm adding (or touched up textures) have so far all been DDS, and the FX folder is mostly converted as well. The particle the monopropellant engines use however always turns purple when I convert it due to compression, so I'll have to look into fiddling with that. All the other left over MBMs will be converted soon enough, I won't do them in one go with a batch method because the converters I tried all severely degraded texture quality.

What are using to convert textures? The only reliable method I've found is using Nvidia's texture tools through Photoshop. Everything else seems to ruin small textures, give compression artifacts in some cases, and for some reason breaks the small agency flag textures.

[hoojiwana]

What are using to convert textures? The only reliable method I've found is using Nvidia's texture tools through Photoshop. Everything else seems to ruin small textures, give compression artifacts in some cases, and for some reason breaks the small agency flag textures.

Yep that's what I'm using too. I experimented a lot when I was setting up AB Launchers for KSP 1.0 and noticed every gradient turned into bands of colour, shadows were all purple and all detail was annihilated when I used anything that converted textures after they had been through Unity. The flags are probably due to mipmap settings, though I still leave mine as PNGs anyway since they're small enough.

[Initar]

Alright, the SRBs seem tamed ^^ , still a little bit much for me, but so are the stock ones. Ideally another point of view on the matter would be usefull.

Also just checking, changelog says "Added revamped 0.625m vacuum engine", which I suppose was the LV-T5 (if I remember correctly) but I can't find it anywhere... And I'm pretty sure the new Aphid is not an equivalent Or maybe it's the aerospike?

Edited July 25, 2015 by Initar

[FiiZzioN]

I can't wait for when this is fully released. Playing with the test version and having a blast just the same as I did with 0.25. Keep it up the awesome work man! When it's fully released my mind will blow!

[TicTacToe!]

Echoing Initar's statement, after some testing I think the SRB's are in line with stock.

Upon testing though, I'm a bit confused on the Arcjet thruster, it doesn't seem to have a niche.

That's just a sample craft with an ion thruster. If you look at a arcjet design that I tweaked to be the same weight:

It's worse on all counts, including TWR. Exactly the same process for the resistojet:

Yes it's worse in dV terms, but it does have a niche, that is it's TWR, beating the Ion for those "clutch" manoeuvres you may need to perform passing by Moho or spontaneous mission critical nodes.

That's all I've found weird though. I've tested probe cores and they all seem to be different enough without flat out hammering them into certain roles, which I believe is a good thing. At the end of the day, if you want your probe to be circular, you should be allowed to have it that way.

The Spinnakers on this rocket got a bit hot during ascent with this rocket, but I'm honestly not sure that's a bad thing at all... (it only got to about 3/5 of the heat bar). And that purpose there is the boundary really for what they should be expected to lift.

That's all I've tested, cannot comment on anything else yet. Would be great to get some other opinions.

EDIT: Hey hooj, when everything's done will we have to use something along the lines of Crossfeed Enabler to let us use inline engines with radial tanks? I've got it installed at the moment, but it doesn't seem to be working.

EDIT AGAIN: I've just realised my analysis earlier only takes into account mass, so I suppose in terms of space the arcjet may be better. Also, I don't play career so if the arcjet is intended as a worse, entry-level, lower cost ion engine, then there's a niche.

Edited July 26, 2015 by TicTacToe!

[Kerbas_ad_astra]

Hey hoojiwana,

One thing that I really like about this mod is that it fills in some gaps at the tiny end of the spectrum. Some of your parts are small enough that they fit into the 0.35m "size 00" category along with some Sounding Rockets and SSR Microsat parts. Here's a patch you can include so that these parts can get sorted into that category when Filter Extension is installed:

@PART[RLA_tiny_vac,RLA_mmrtg,RLA_mp_tiny_stack,RLA_tiny_torque_radial]:NEEDS[FilterExtensions]
{
    @bulkheadProfiles = size00,srf
}

EDIT: I'd also suggest adding "size1" and "size2" to your MAXO frames, since they're stackable and they do fit in that width.

[hoojiwana]

Alright, the SRBs seem tamed ^^ , still a little bit much for me, but so are the stock ones. Ideally another point of view on the matter would be usefull.

Also just checking, changelog says "Added revamped 0.625m vacuum engine", which I suppose was the LV-T5 (if I remember correctly) but I can't find it anywhere... And I'm pretty sure the new Aphid is not an equivalent Or maybe it's the aerospike?

Agreed that more thoughts on the SRBs are needed. I deliberately made the boosters very strong so they're actually useful as radial TWR boosters on bigger rockets (even up to 2.5m if you use a lot of them). And the upperstage ones aren't terribly common to see people using since they're hard to use.

And yes the Caravel (little aerospike) is the new 0.625m vacuum engine. The idea being that aerospikes are space saving since you can truncate the nozzle, making it fit under the smallest landing legs. Also ties nicely into Rockomax developing the Cutter (linear aerospike), some progression like that in the tech tree is pretty neat I think. So all the main 0.625m LFO engines are now all Rockomax, just like 2.5m, going to give the Caravel and Spinnaker numbers like the 48-7S so they all appear together in the editor list.

I can't wait for when this is fully released. Playing with the test version and having a blast just the same as I did with 0.25. Keep it up the awesome work man! When it's fully released my mind will blow!

Thanks! A fair few tweaks to things like particle FX and balance and it'll be there.

Echoing Initar's statement, after some testing I think the SRB's are in line with stock.

Cheers.

Upon testing though, I'm a bit confused on the Arcjet thruster, it doesn't seem to have a niche.

http://i.imgur.com/lv2zbZV.png

That's just a sample craft with an ion thruster. If you look at a arcjet design that I tweaked to be the same weight:

http://i.imgur.com/JTyWF5v.png

It's worse on all counts, including TWR. Exactly the same process for the resistojet:

http://i.imgur.com/PNH2zKY.png

Yes it's worse in dV terms, but it does have a niche, that is it's TWR, beating the Ion for those "clutch" manoeuvres you may need to perform passing by Moho or spontaneous mission critical nodes.

Lots of juicy feedback! One thing I noticed is that you tweaked the monoprop tank to remove fuel from it, which has the side effect of increasing it's dry mass ratio, which has a negative impact on dV (above and beyond that of simply lowering the fuel). If you do a direct comparison of Resistojet to Arcjet using full tanks the Arcjet should always come out ahead in terms of dV. The two may still need some tweaks but I don't think they'll be major ones.

The design behind the four 0.625m high efficiency engines (LV-Nc, Arcjet, Resistojet and Ion) puts them all in different niches. The LV-Nc gets the benefit of the best dry mass ratio tanks and has the highest thrust, the Resistojet is the higher thrust monoprop option, the Arcjet is the higher efficiency monoprop option, and the Ion is the super efficient Xenon option (and leads into the fun things Near Future adds, and is designed around that). Directly comparing Xenon tanks to anything else is also a bad idea since they have such poor dry mass ratios (nearly half their wet mass is tank mass).

That's all I've found weird though. I've tested probe cores and they all seem to be different enough without flat out hammering them into certain roles, which I believe is a good thing. At the end of the day, if you want your probe to be circular, you should be allowed to have it that way.

You've no idea how glad I am to hear that!

The Spinnakers on this rocket got a bit hot during ascent with this rocket, but I'm honestly not sure that's a bad thing at all... (it only got to about 3/5 of the heat bar). And that purpose there is the boundary really for what they should be expected to lift.

http://i.imgur.com/LyM0Yv8.png

Yeah they are supposed to get hot, they have a fair bit more thrust than they used to, and are supposed to be good for both 0.625m first stages and be good enough clustered for bigger rockets.

EDIT: Hey hooj, when everything's done will we have to use something along the lines of Crossfeed Enabler to let us use inline engines with radial tanks? I've got it installed at the moment, but it doesn't seem to be working.

Yes, I have no intention of changing the way the monoprop engines draw their propellant because the alternative (draining through decouplers) is still worse. Tried to make an aircraft that used decouplers at all in 1.0? Fuel lines don't work on radial tanks either so you can't get around the problem that way.

Hey hoojiwana,

One thing that I really like about this mod is that it fills in some gaps at the tiny end of the spectrum. Some of your parts are small enough that they fit into the 0.35m "size 00" category along with some Sounding Rockets and SSR Microsat parts. Here's a patch you can include so that these parts can get sorted into that category when Filter Extension is installed:

EDIT: I'd also suggest adding "size1" and "size2" to your MAXO frames, since they're stackable and they do fit in that width.

Thanks for the config! And I'll do that for the structural frames as well.

[Astrofox]

Back in February I made this thing (to the right of the QBE) to test how the reaction wheel module works. The idea was to have a radial reaction wheel (based on Keplers) that only worked on one axis and you could place them how you liked to get torque in various axis. Unfortunately the reaction wheel module applies torque based on the entire vessels axes rather than the parts axes. If I made it a regular radial reaction wheel would you guys and gals use it? Given it's small size it would have a fairly low torque so may not be that useful.

http://i.imgur.com/J4ura7d.jpg

I would use those (sorry for such a late reply) for QBEs and other craft, and low, accurate torque is what I need for my Hullcam VDS space telescopes...

[TicTacToe!]

The design behind the four 0.625m high efficiency engines (LV-Nc, Arcjet, Resistojet and Ion) puts them all in different niches. The LV-Nc gets the benefit of the best dry mass ratio tanks and has the highest thrust, the Resistojet is the higher thrust monoprop option, the Arcjet is the higher efficiency monoprop option, and the Ion is the super efficient Xenon option (and leads into the fun things Near Future adds, and is designed around that). Directly comparing Xenon tanks to anything else is also a bad idea since they have such poor dry mass ratios (nearly half their wet mass is tank mass).

Wow, thanks for sharing the knowledge about the dry mass ratios! I knew they were different but not to such an extent, upon research! Haha, well... that ties that problem up in a neat little bow!

[Initar]

Hey hooj, when everything's done will we have to use something along the lines of Crossfeed Enabler to let us use inline engines with radial tanks? I've got it installed at the moment, but it doesn't seem to be working.

If you want Croosfeed enabler support, you just have to make a MM patch containing these lines for every tank you want to add this module to:

@PART[RLA_tiny_mptank_rad]:NEEDS[CrossFeedEnabler]
{
	%MODULE[ModuleCrossFeed]   {}
}

Hope it helps!

@Hoojiwana: aw ok I get the reasonning behind this choice. I'll miss the little guy though ^^ Thanks for the precisions!

Edited July 27, 2015 by Initar
ninja'd

[Overlocker]

I don't know if its a bug or a feature, but is anoying to have the smaller monoprop engines without fuel pumping around all the ship, like normal rcs thrusters. Also if its a feature still the smaller radial ones look strange when they are attached to a fuel line that is BIGGER than the engine itself.

[TicTacToe!]

I don't know if its a bug or a feature, but is anoying to have the smaller monoprop engines without fuel pumping around all the ship, like normal rcs thrusters. Also if its a feature still the smaller radial ones look strange when they are attached to a fuel line that is BIGGER than the engine itself.

Take a look at the above couple of posts. If I understand what you're getting at correctly, in previous versions we had Crossfeed Enabler support, we don't at the moment due to still being in test versions. It'll come.

[Streetwind]

Okay, back from vacation.

Ran the electric engines through my Near Future spreadsheet. I know you don't want to align to that balance paradigm out of the box - but, at the very least it can give a mathematical comparison of the engines among each other, and against the stock Dawn engine.

(Disclaimer: I didn't bother tuning the formulas for stock power production equipment performance. OpTWR will be lower ingame than shown here, and ER/EE are artificial ratings anyway. The ratio of differences between them should remain identical though, and that is what ultimately counts here.)

What immediately stands out that - as TicTacToe's gut feeling said - the Umbra is grossly underpowered by Engine Rating (ER). By pure stats alone, the Eclipse looks weak as well, but it has that high TWR that at least makes it feel noticably different from the ion thrusters. Still, both of the monoprop engines suffer from the fact that they basically lost most of their Isp while still having the same power draw, and gained little in return.

This is perhaps best illustrated by comparing the electric monoprop thrusters to other options in the same form factor. Like your "Mighty" mini-nuke.

The electric engine consumes a little over one third of a gigantor array in power, so let's add some 0.11 ton to for example the Eclipse. You end up with 0.32t mass, where the Mighty masses 0.32t. They're equal. And for that, the Eclipse offers 3.5 kN thrust at 875s Isp, while the Mighty does 6.0 kN at 750s. Objectively, the Mighty is simply better - and that's without factoring in that it costs a mere third of the Eclipse+solars, and doesn't lose output as you move away from the sun like the Eclipse does (because the solar panel loses output, requiring you to throttle down the resistojet). Or the fact that the higher TWR probably weighs up the lower Isp, making the Mighty probably equal or better in absolute dV on a vessel that has the same overall TWR. There is only one thing in favor of the Eclipse right now, and that's the fact that there are no 0.625m pure LF tanks to go with the Mighty. Other than that, there's little reason to go for the Eclipse.

And in a direct comparison, the Eclipse is better again than the Umbra. A 16% dV advantage doesn't really weigh up a 60% loss of thrust, especially if some of that thrust can be traded for extra dV and still give you higher TWR. It's also more expensive.

Both of these things are practical manifestations of the ephemeral "Engine Rating" stat in the table above.

Of course, this math is just there to offer guidelines, and isn't the rule to end all rules. The monoprop eletric engines stray a bit in the territory of hybrid electric/chemical drives, and the formulas are not really suitable for chemical engines. Not to mention that stock KSP has grossly overbuffed the Dawn. If you simply went and tried to buff the Umbra to an ER of ~3k, like the Dawn has, you'll end up giving it 7.3 kN thrust. Which ends up walking all over the mini-nuke. And well, technically the Dawn already walks all over the LV-N, and most other stock engines... it just doesn't matter because you'd have to cluster so many Dawn's and solar panels to reach a LV-N's thrust that you start killing your framerate with part count, so only very few people do it. So stock KSP is kind of saved from screwing up their engine balance here. Unfortunately, you don't have that luxury, because you introduce a 0.625m nuke that tries to stay close to the LV-N's style. As a result, a trio or quad of Dawns utterly destroys the Mighty in all possible categories except price, and your mod even delivers the convenient 0.625m adapter to do it!

So ultimately your problem is that your electric engines are trying to straddle the line between the LV-N (by proxy of the Mighty) and the Dawn, which is an endeavour doomed to fail because the Dawn is just better, period. At the same time, you can't really make them objectively a lot worse than the Dawn, because they literally sit in the same tech node as it does. So you need to give it niches. The Sunrise does okay, it lost a lot of thrust but gained a lot of Isp. Not enough to bring it up to par, but enough to feel different. The Eclipse has its own niche with high thrust, but unfortunately the Mighty beats it handily at that for a third of the price, so the Elipse needs a buff. The Umbra though... it has nothing. It needs a complete revamp.

My recommendation:

- Sunrise: 8.2 Ec/s power draw, 1.0 kN thrust, 5900s Isp. Small buff, underlines the "higher Isp for less thrust" niche. Also, since this is the thematically most similar engine to the Dawn, it has the most similar performance rating to the Dawn. (Engine rating: 2617, up from 2507)

- Eclispde: 4.7 Ec/s power draw, 5.1 kN thrust, 875s Isp. Much higher thrust, and new niche: an engine for situations where there's less power available. (Engine Rating: 2003, up from 1291). Yes, it now has more TWR than the Mighty... but it is also a lot more expensive. And still massively weaker than the Dawn, objectively. Buffing it more would create problems for the Mighty and the other monoprop engines, though. See, this is why Near Future nerfs the Dawn

- Umbra: 7.3 Ec/s power draw, 2.4 kN thrust, 2100s Isp. Buffed all across the board, it's now a jack-of-all-trades engine that doesn't particularly excel at anything... honestly, it's either this or axing it, because I don't see a unique niche for it anywhere =/ Even this will probably make people pass it over 19 out of 20 times in favor of the Dawn. (Engine Rating: 2003, up from 572).

Finally, swap the prices for the Umbra and Eclipse, so that the former has the lower cost.

This is a math balancing approach. It's best if you tested how it feels ingame, and if feels off, then tweak it with your gut feeling. It's really hard to not render the low-thrust chemical engines irrelevant with these because the Dawn is just so absurdly good, and these need to at least go a little ways towards meeting it, else they don't make sense in the same node. At least the price point will ensure that for example the Ant will always have its niche. Career mode is nice like that.

(Free bonus: ER 2000 is actually the tier 1 balance target for Near Future. The arcjet and resistojet wouldn't need any compatibility patch with these stats.)

EDIT: You could also move the Mighty out of harm's way a little by doing, say, 12 kN at 0.65t weight.

EDIT2: Potentially you could also make the Umbra the engine with the smaller power draw. I just thought it felt a little odd because the word "arcjet" literally makes you think "a lot of electric power"

Edited July 27, 2015 by Streetwind

[TicTacToe!]

snip

And that ladies and gentlemen, is how you consult on balance.

Great to see soemone as experienced as yourself in these matters weighing in, I knew I was on to something! Though a lot more eloquent and thorough with your explanation.

[hoojiwana]

I would use those (sorry for such a late reply) for QBEs and other craft, and low, accurate torque is what I need for my Hullcam VDS space telescopes...

Go ahead and test them out, they're set at 0.5 torque right now, I do wonder if that's the right amount.

@Hoojiwana: aw ok I get the reasonning behind this choice. I'll miss the little guy though ^^ Thanks for the precisions!

Yeah the old design is gone for good. I didn't think it looked too great and what I wanted to change it to wasn't suitable for use as a landing engine, so the aerospike showed up instead.

I don't know if its a bug or a feature, but is anoying to have the smaller monoprop engines without fuel pumping around all the ship, like normal rcs thrusters. Also if its a feature still the smaller radial ones look strange when they are attached to a fuel line that is BIGGER than the engine itself.

Take a look at the above couple of posts. If I understand what you're getting at correctly, in previous versions we had Crossfeed Enabler support, we don't at the moment due to still being in test versions. It'll come.

Of all the engines, Mist and Fog are probably the only ones that would be fine using the "draw from everywhere" method that RCS uses, I'll change it back to that for the next version.

What immediately stands out that - as TicTacToe's gut feeling said - the Umbra is grossly underpowered by Engine Rating (ER). By pure stats alone, the Eclipse looks weak as well, but it has that high TWR that at least makes it feel noticably different from the ion thrusters. Still, both of the monoprop engines suffer from the fact that they basically lost most of their Isp while still having the same power draw, and gained little in return.

And in a direct comparison, the Eclipse is better again than the Umbra. A 16% dV advantage doesn't really weigh up a 60% loss of thrust, especially if some of that thrust can be traded for extra dV and still give you higher TWR. It's also more expensive.

Of course, this math is just there to offer guidelines, and isn't the rule to end all rules. The monoprop eletric engines stray a bit in the territory of hybrid electric/chemical drives, and the formulas are not really suitable for chemical engines. Not to mention that stock KSP has grossly overbuffed the Dawn. If you simply went and tried to buff the Umbra to an ER of ~3k, like the Dawn has, you'll end up giving it 7.3 kN thrust. Which ends up walking all over the mini-nuke.

So ultimately your problem is that your electric engines are trying to straddle the line between the LV-N (by proxy of the Mighty) and the Dawn, which is an endeavour doomed to fail because the Dawn is just better, period. At the same time, you can't really make them objectively a lot worse than the Dawn, because they literally sit in the same tech node as it does. So you need to give it niches. The Sunrise does okay, it lost a lot of thrust but gained a lot of Isp. Not enough to bring it up to par, but enough to feel different. The Eclipse has its own niche with high thrust, but unfortunately the Mighty beats it handily at that for a third of the price, so the Elipse needs a buff. The Umbra though... it has nothing. It needs a complete revamp.

My recommendation:

- Sunrise: 8.2 Ec/s power draw, 1.0 kN thrust, 5900s Isp. Small buff, underlines the "higher Isp for less thrust" niche.

- Eclipse: 4.7 Ec/s power draw, 5.1 kN thrust, 875s Isp. Much higher thrust, and new niche: an engine for situations where there's less power available. (Engine Rating: 2003, up from 1291). Yes, it now has more TWR than the Mighty... but it is also a lot more expensive. And still massively weaker than the Dawn, objectively. Buffing it more would create problems for the Mighty and the other monoprop engines, though. See, this is why Near Future nerfs the Dawn

- Umbra: 7.3 Ec/s power draw, 2.4 kN thrust, 2100s Isp. Buffed all across the board, it's now a jack-of-all-trades engine that doesn't particularly excel at anything... honestly, it's either this or axing it, because I don't see a unique niche for it anywhere.

EDIT: You could also move the Mighty out of harm's way a little by doing, say, 12 kN at 0.65t weight

EDIT2: Potentially you could also make the Umbra the engine with the smaller power draw. I just thought it felt a little odd because the word "arcjet" literally makes you think "a lot of electric power"

Here's the post I was waiting for!

Firstly, death to the Dawn. The buff Squad made to that has been such an enormous headache that I kinda of gave up on balancing the electric engines before now, and is the reason why the Eclipse and Umbra lost their Xenon modes, they just could not compete on any level. People didn't use it because it's a pain in the ass to fly anything with low TWR, and people always build huge spacecraft and try to use engines totally unsuited for things of that size. Increasing the thrust and decreasing the power draw just made the Ant the least useful one, it just moved the problem elsewhere since people still build bigger and bigger craft.

Anyway.

The Umbra has always been in a bit of a weird spot, between the Dawn and the Eclipse and so kind of iffy in its purpose. Right now it's the high ISP option for monoprop vessels but it may not be enough as your analysis shows. Removing it seems like the better option since buffing it to such high levels isn't something I'm keen on doing. It's ISP is already higher than a comparative hydrazine one (in real life), and of course its thrust is but it has to be to be competitive. Maybe an absurdly good hydrogen arcjet could hit those numbers but there isn't really a hydrogen analogue in game (LF doesn't really count).

I'll look at trying the low-power approach for the Eclipse as well. And as for the Mighty, because of the lack of pure LF tanks (and lack of tank fuel switch) that thing is going to be a nightmare to set up right. Need more thoughts on that subject, and on heating. Would people try to use the Mighty for smaller probes? Or would it be used on larger probes and using 1.25m and bigger LF tanks? What about clustering it? If people want to use it primarily for bigger vehicles leaving the other engines for small probes then that's great, it can have more thrust+mass for that.

In other news I may also cut the Puffin (0.625m solid upper stage) since it can't really compete with the other options in that size with the exception of cost. The 1.25m solid upper is okay since it ends up with marginally higher dV than similar mass and thrust of tank+Terrier. (And similar mass monoprop+Nimbus has a bit more dV than both, but less thrust and costs more.)

Edited July 27, 2015 by hoojiwana