Spirecraft Penetrator balance... mostly

[Irxallis]

Spirecraft Penetrators.

Story

We (me + my wife) think they are a bit on the overpowered side. Would you be so kind to look at the screenshots below?

Situation:


Location of Masnu and the other homeworld:


The screenshot presents an AI homeworld which has been deepstriked. There are 16 Penetrators on the planet (8 are mine, 8 belong to her). All those Penetrators have been created using Pysite, the second most common asteroid (so we have plenty left, still).

After the first screenshot Masnu as a planet lost all the guard posts. Penetrators ignore shields and deal 7 mln damage, so nothing could stop those. We decided to retreat them to safety and 10 out of 16 Penetrators managed to live (with no fleet supporting them and planets on the way just neutered from guard posts except wormhole ones).

We scrapped them and rebuilt our fleet of Penetrators (as we control 30 planets and Pysite is very abundant), sending them to the second homeworld. Not a problem, same approach, this time 9 Penetrators managed to live (same case, no guard posts except wormhole ones).

Bear in mind we had one Mad Bomber AI, thus those poor Penetrators needed to pass ~80 Bomber Starships per system. And most of them just lived through this experience.

We didn't really have to worry how to kill the homeworlds. We positioned 8 Penetrators per planet (because we could overkill like that) and opened fire simultaneously. A bit anticlimatic ;-).

Why overpowered?

- They come at Pysite (#2 asteroid), therefore they can get very abundant.
- They have 80 mln hitpoints, therefore it is VERY hard to kill them and they can live to fight another day.
- They have perma-invisibility, so even tachyon guardians cannot really stop them.

What would be a good change to them in my opinion?

- Move tier1 from Pysite to Xampite (#3 asteroid). This will stop them from being a cheap 'I kill you' button, as tier1 being VERY powerful in its own rights would be less accessible. In my opinion this is the single most important change which would fix the problem, as this would make us conserve Penetrators much more carefully.

- Reduce their hitpoints greatly, from 80 mln to, say, 10 mln. That way in case of deep strikes you know they will not come back, unless your fleet is nearby (add it to a Xampite change and they need at least some support - unless you have too much Xampite which we always fight for ;-) ).

- Change their 'always-invisibility' to normal stealth. This will at least force you to kill all the Tachyon Guardians (which we did anyway due to ignorance - we'd have a win 2-3 h earlier if not for those and neutering).

- MAYBE reduce their cap from 8 to 4 to prevent homeworld insta-killing. For more players this will return anyway, though, so this is no solution.

What would be a bad change in my opinion?

- Make Guard Posts 5 immune to Penetrators. In our last game we were "blessed" by a Core Raid Engine under a Force Field tier 3 on the homeworld. To add insult to injury the core world leading to the homeworld had a gravity guardian on The wormhole and an AI got... gravity rippers. At Tier5 those are... Game over, even with cloaker starships - Raid Engine spammed waves of ~1.5k ships on us every minute or so and those cloaked transports barely moved.

- Greatly reduce their damage. Ok, 7k -> 5k is not a problem (requires 2 Penetrators / Guard post on Tier1 instead of 1 - consider this too, please), but 7k -> 2k would be a bit insane ;-).

- Make them vulnerable to gravity stuff. NOOOO!!! <insert famous Vader cry from Star Wars 3 (shudder & cower in fear)>.

- Make them work only if they have supply. The sheer thought is horrifying, but it would still be better than making guard posts T5 immune to Penetrators ;-).

Side note

Disclaimer: those suggestions below haven't been analysed with in-game data. I would strongly suggest not listen to them unless someone more experienced also agreed.

- Would you consider moving Ion Blaster 1 from Pysite (#2) to Reptite (#1)? As those stand now, I cannot find any reason to use them, between Martyrs mk2 (able to kill ~1k ships from the retaliatory attacks of ANY tier or as a final wormhole defense), Attritioners and Penetrators. If a Pysite (still a common asteroid) Ion Blaster would kill Tier2 ships it would be, for example, a good wave defense assistant under a forcefield. Then a Xampite one could fight Tier3 ships - but those are much more rare. And Ebonite... well, we had one Ebonite asteroid in our 30 planets. One Ion Blaster killing Tier4 ships is not overpowered in my opinion ;-).

- Would you consider increasing the number of Ebonite and higher class of asteroids a bit? I haven't really played with those as I simply never managed to get them at 30 planets in a 60-80 planets galaxy (1 Ebonite asteroid). I mean: increase the presence on different planets, not the concentration on particular planets. It would still leave them as rare bonus (unless you hold 50-75% of the galaxy) but would usually give you some goodies to play with.

[BobTheJanitor]

I have not toyed with these much as they require micromanagement and I tend to forget that they are there until I accidentally select them in a fleet blob and sent it all at a target, so they waste their mighty shot and 1/2 hour cooldown when I tell my fleet to attack a random vampire guardian or something. If they can indeed take out all of the guard posts and wipe out an AI homeworld by themselves, that is quite a bit. Being able to deepstrike with them and take out a homeworld is obviously over the line. I'm guessing you were just able to shrug off all the defenses along the way every time you got decloaked by tachyon guardians? It seems like the solution would be to turn them more into a glass cannon and drop their HP drastically so that they require actual fleet support to be able to make their shot.

Edit: Oh they're perma-cloaked even through tachyons? Guess I should try that read > comprehend > post concept I've heard of in the past. I had no idea. That is a bit crazy. If that's going to stay, maybe some kind of self-attrition per hostile wormhole, like transports?

[Sunshine!]

I haven't gotten to that stage of abuse with penetrators, but with the way Spire Civilian Leaders currently work, I'll send out some Spirecraft Scouts (like mk4 scouts, but self-attrition and don't require knowledge) to find all the Spire Civilian Leaders, and then I'll disperse Penetrators to knock out all those command stations so that I'm getting some negative amount of AIP every hour depending on how many I've been able to locate/destroy given the reload time (as well as incidental command station shieldposts/counterwave posts).

[BobTheJanitor]

Clearly I hadn't considered the level of abuse one could glean from what amounts to a one-button kill-anything-in-the-galaxy unit. Clearly I should go abuse these before they suffer the wrath of the nerf bat.

[Sunshine!]

It's one-hit-kill-anything-not-an-FF-or-Fortress.  Not really an issue because the ignore FFs.

[Red Spot]

I tried this with 6 or 7 planets and it is easy, just build the Pysite/mk1 Penetrators and send them to an AI homeplanet (no fog of war in that game). 1st AI homeplanet toast 1 hour in.

I do however think this is one of those features that works or doesnt, I dont think you can effectivly balance them differently while stil keeping them as they are and you just have to decide wether to follow that route or not.
Perhaps a flag can be added so the lower marks for instance cant enter core/home-planets? I rather not see them nerfed though, it would make them useless on higher difficulties, I think.



- Move tier1 from Pysite to Xampite (#3 asteroid). This will stop them from being a cheap 'I kill you' button, as tier1 being VERY powerful in its own rights would be less accessible. this is the single most important change which would fix the problem, as this would make us conserve Penetrators much more carefully.
Wont make any difference to me, just the max amount of Penetrators I will yield as I see roughly the same amount of both mineral/use. I use the Reptite(sp? lowest mark) for mk1 Martyrs and Siegetowers. Basicly skip using Pysite except for a cap of mk2 Siege towers, and use Xampite for mk2 Martyrs.
Any other ship I can build with Pysite is so situational and in most cases wasted resources I dont bother with them, except perhaps some shielbearers on a defensive planet to boost the total amount of FFs I can put in the field. From Xampite and up the ships actually get enough power to be usefull to have even if only for general purpose.
So the change would just mean I now build mk2 Martyrs using Pysite and mk1 Penetrators using Xampite resulting with the same fleet for the same cost in minerals.
- Reduce their hitpoints greatly, from 80 mln to, say, 10 mln. That way in case of deep strikes you know they will not come back, unless your fleet is nearby (add it to a Xampite change and they need at least some support - unless you have too much Xampite which we always fight for ;-) ).
You never see one return home, could at this stage just as well make it a suicide unit.
- Change their 'always-invisibility' to normal stealth. This will at least force you to kill all the Tachyon Guardians (which we did anyway due to ignorance - we'd have a win 2-3 h earlier if not for those and neutering).
Damages their usefulness greatly. Now I can concentrate on front A while I neuter a gate on front B using these guys. Make it so I have to take out the tachion guardians (on every gate on higer difficulties, and even near posts), I could just as well get a cloaker and send in some bomber starships or a transport, saving me even more Pysite for other nasties.
- MAYBE reduce their cap from 8 to 4 to prevent homeworld insta-killing. For more players this will return anyway, though, so this is no solution.
I'd sooner say, double their cap, and make them require a twice as large cooldown period. This will mean you need a lot more Pysite to get the same effect out of them. (Now you can take out 16targets/hour for 8 Pysite, with my suggestion you would need 16 Pysite to take down 16targets/hour, doubling the cost on the limited resource for the same effect.

[Irxallis]

Thank you very much for your comments, much appreciated. I didn't think of the Spire Civillians before; need to check it out ;-).

I wouldn't like those to be nerfed to infinity and beyond as well, though right we feel those are a bit too much on the strong side. I used them as a 'last resort weapon', 'Raid Engine killer weapon' and I think I will pick a Golemite our next game - who knows, maybe I'll use them as a Golem Killing Engine (especially with some Reptite Rams with a cloaker starship) - AI War makes me search for interesting combinations all the time.

Right now they are definitely broken, especially with the 'giving Penetrators resets their countdown timer' bug which I described here (NOT going to abuse it, it would be too easy):
http://www.arcengames.com/mantisbt/view.php?id=2798

I think my problem with Penetrators might be the perceived lack of the focus of the unit. Is it supposed to be a long range deep strike suicide unit ('Spire version of a Raid Starship') or a unit able to kill even very durable targets but requiring fleet support, able to pull back?

In both cases I think mark1 should have the damage reduced from 7m to 5m. That way you need 2 Penetrators t1 to kill a home guard station and this 5m damage is extremely high anyway.

After performing the analysis of Red Spot's analysis of my analysis (sorry, I had to ;-) ) we have two suggestions how to change them, regardless of the function of the unit in such a way to allow them perform their functions (and allowing some cheesy tactics) and in the same time make it much harder to auto-win.

- move T1 Penetrators to Xampite. Red Spot is correct that there still are many Xampite asteroids, but you don't have access to Pysite asteroids for Penetrators anymore. This naturally reduces the access to those, kind of forcing at least some spreading out.

- Reduce the damage of T1 Penetrators to 5m. This makes us require 2 Penetrators per a Core Guard Post instead of one not influencing any other 'valuable kill' there is I can think of.

Assuming no other change to Penetrators and... say, 13 core guard posts per AI planet it would require 26 shots to kill all the guard posts to 'open' the AI homeworld (not counting things like Core Neinzul Fortress with 21m HP). This means we'd need 52 (or more) Xampite or higher to execute a 'game over' scenario even without performing any other change to Penetrators in case of ultra-deep strikes. In other words, those won't be the 'I win' button anymore, still allowing things like 'kill that miserable Core Raid Engine'.

Some additional comments:

You never see one return home, could at this stage just as well make it a suicide unit.

I can see Penetrators return home with fleet support (deepstrike with killing key targets) and I would support making the cap larger and lenghtening the recovery time from 30 min to 45 min - 1h. After we thought about it we think it is not the cap which should hold us back from using the unit, but it is the amount of Xampite and opportunity cost - it would make double full cap of Attritioners (8 Xampite) something to really consider, not auto-grab.

About Tachyon guardian vulnerability - Damages their usefulness greatly. Now I can concentrate on front A while I neuter a gate on front B using these guys.

I don't understand. If this is only one wormhole away, even with 10-20 mln HP it shouldn't be a problem for a Penetrator to kill a gate. Furthermore, giving them this vulnerability might warrant not decrease their HP that much. And, of course, we could consider opportunity cost of early Penetrators killing Tachyon Guardians and returning home, losing one shot out of their 30 (45? 60?) minutes. It would damage deep raids, but I cannot see how it influences close range raids - maybe this is because I haven't played on difficulty higher than 7 and I might lack the data.

Thank you very much for your insight,
~Irx.

[Kjara]

Another option would be to possibly add attrition based on world level to the various types (as with the spire scouts/transports).  That way the lower mark ones could be used on near targets or defensively, while the higher marks would still have the ability to make pretty deep strikes if you chose your path well.  Ideally the rate would be slow enough that it can survive about 2-3 planets worth of attrition, so that you can hit a raid engine 2 steps out, but deep raids would be reserved for the higher level versions(and if used well, the higher level versions should be able to strike and return).  Could counter this out with regen on friendly planets(slow, but fast enough that they will be full before they have recharged and can fire again), to encourage reusing them after a strike rather than just scrapping and building new ones.

[Red Spot]

Something like how transports work could do I guess (5hops/20% ?), arrtrition like how the spire scouts work would become painfull I think.
You dont get any (much) of the higher class minerals, so it would just become one of those 'out of reach' units, where the lower marks would be frustrating to use in the long run. So far I have only seen Ebony/Ebonite(sp?), and only few of those.

[BobTheJanitor]

Yeah I built one spire scout and could not figure out a way to make it useful. Attritioning units are usually frustrating except in cases where they have a ton of HP like golems. Or the spire transport ship, which is the best thing for deep scouting with liberal use of micro and pausing. The quick attrition on that doesn't matter when you only have to unpause long enough for it to go through a wormhole, drop a scout, and go to the next.

Anyway, if the penetrators attritioned per wormhole, that would be OK, although remember you can't quite make it the same as transports if you want to get it back. It has to have enough HP to make the journey out and back while also taking fire on the way back since it will be uncloaked. So if it should survive 5 hops out, you'd want it to lose like 8-9% per wormhole. That'll take 40-45% hp per leg of the journey and leave it with 10-20% hp to take fire on the way back. I don't know if it's possible to combine attrition per-wormhole with the higher mark rating attrition thing though. But maybe you could have higher marks take less damage per jump along with having more hp and firing bigger shots.

[Red Spot]

Why I suggested 5 hops (20%), you can go 5 hops, 3 hops and get the ship back, or 4 and have to conquer a planet before you make it attack as it cant make the full yourney back home.
With a high enough HP is doesnt matter much it has only 20% health (for instance) when it passes through the last planet (20% of a lot is still a lot .

[chemical_art]

Rather then trying to work out the nitty grittynumbers (quantitative), I decided to think about what exactly the penetrator is all about (qualitative). Ultimately, its meant to be a lone wolf ship, according to the description. A ghost that moves into enemy territory, unleashes a terrorizing single strike, then retreats. Unfortunately, there is nothing to prevent it from being "lone wolf" and not "fleet support / stealthy artillery golem".


A lone wolf unit would be unique in the hands of the human, but how do you make one that is not also useful to your own fleets? That I do not know, but if a mechanism could be used to discourage tag-teaming them with fleet support then that alone would balance them greatly. I don't know if, for example. you could somehow give them penalties of any sort for being around friendly craft and/or planets.

[BobTheJanitor]

Well they could have some sort of debuff area of effect that only effects friendly ships. An anti-munitions-booster effect, or some sort of gravity effect that slows friendly ships. You'd just have to come up with some sort of good fluff explanation for it. Although then you have the problem of finding a world to store them on while you're not using them. But I don't know that pairing them up with fleet support is really the problem here. After all, the original example was sending a ton of them to an AI Homeworld and singlehandedly destroying it. No fleets involved in that picture.

[chemical_art]

Well they could have some sort of debuff area of effect that only effects friendly ships. An anti-munitions-booster effect, or some sort of gravity effect that slows friendly ships. You'd just have to come up with some sort of good fluff explanation for it. Although then you have the problem of finding a world to store them on while you're not using them. But I don't know that pairing them up with fleet support is really the problem here. After all, the original example was sending a ton of them to an AI Homeworld and singlehandedly destroying it. No fleets involved in that picture.

I thought the proposed changes of lowering their firepower would have done that though? A command grade bonus of "zero" would mostly solve this as well, and on reflection I am really surprised it was not done already.

EDIT: .1 command grade would work as well. A mk I ship wouldn't destroy many command stations in one shot, but the higher mk could. Even a MK V can't one shot the AI homeworld with .1 as well.

[BobTheJanitor]

Oh, I must have missed that suggestion somewhere along the way. Oops. That would work OK, although it does also affect the other things that have command grade armor that you might want to kill. Which, looking at the list, is mostly command shield generator guard posts, warp gates, data centers, troop accelerators, and hunter-killers (I don't even know if those spawn any more, or if their role was removed when carriers were added). Warp gates are weak enough that it wouldn't matter though. The com-shield posts could be kind of annoying to be unable to destroy, and being unable to use penetrators to kill a data center would be a big pain. I guess they could get a different armor type if it came to that. Troop accels might also be something you'd want to deep strike against before alerting the world that they're on, so that's another down side to this. In a game as complex as this one, any simple solution probably isn't. 

I was really looking forward to being able to abuse penetrators in the next game I tried, and then of course I randomly rolled a One-Way Doormaster AI which took the wind right out of my sails. How about making these immune to black hole machines, eh?