Rebellion criticism: Pirate nonsense, weaksauce AI and more

Pirates will not be changed. Not ever, though many have suggested things like yours. Including myself.

The AI is something else. Some say the AI is too strong, others (like you and I) say its too weak. Sometimes it does what you are talking about, never retreating. Other times it just sits there and when you attack, it runs.

I also don't see it as likely StarDock will change the pirate model since they've gone with it in four iterations of the game now. One can hope, however, and I think the issue is still worth debating if for nothing more than bringing some attention to the issue.

Have they stated officially somewhere they are happy with the pirate model and won't change it?

Actually the pirates have been changed very drastically in their difficulty. They were laughably weak in Entrenchment as Starbases made them easy XP, in multiplayer people wanted to get the pirate raids on them (in the rare match where they were enabled). So in Diplomacy they made the pirates actually a threat, with upgradable special abilities and stats, along with antistructure capability. The result is everyone but the most skilled players thought they were too strong, and they really did influence the game too much. They've sense been nerfed but aren't as weak as in Entrenchment.

I realize your post is more about the structure of the pirates, but my point is like the AI as Ryat mentioned, it's truly impossible to make everyone happy with the pirates. Either they're too hard for new players and discourage them from the game, or the pros find it way too easy and just XP farm them. You may think pirates are naturally commerce raiders and don't want to concentrate to fight tough targets, others will think commerce raiding is totally useless and want them to only attack tough defenses, otherwise why would they pay them. I think they wasted too much time as it is trying to improve the pirates, because only a minority is ever going to like the pirates a given way.

They haven't, but keep in mind the Pirates were with Sins since the first launch, and were thus created by Ironclad games. Stardock does most of the patch work, and it seems they don't like to change stuff Ironclad original did unless it is truly game breaking.

Not quite true. The AI can retreat, there's even a value you can mod to make them more or less likely to. The problem is how the AI evaluates how strong its forces and that of the enemy's are. I don't believe they take any account for capitalship and titan levels or Starbase upgrades, so they overestimate weak ones and underestimate strong ones all the time. Nor does it take research improvements into account. You try treating every titan like it's level 1 and see how well you do. Obviously it would be nice if this was improved, but it's no easy task, and even if that basic system was fixed, you get a mess of how do you treat an Eradica differently from a Coronata etc.

The small random map is notoriously bad. I wouldn't judge the game based on that. Also keep in mind normal is the "Second Easiest" AI (out of 6) for a reason.

Brilliant, informative post, Goa, thank you. I'll just address your points in order:

1) Perhaps you're right and pirates are hard to balance - I appreciate your detailed historical rundown - but I really don't believe it to be so. My point is simply that, as you say, I dislike the structure of the pirates. We have an exaggeratingly overfortified fortress in the center of the map spewing out waves of extremely large - if unshielded and with terrible AI, but that's another matter - fleets. That just isn't what a 'pirate' is! If they'd called them Antareans, like in Master of Orion 2, I'd actually have been fine with it (these Antareans worked practically the same as SOASE pirates: a very strong homeworld sending extremely powerful fleets). The problem is that in SOASE they're called pirates and I can hire and control this super force for 250 credits, less than the cost of a single ship. If anybody is truly happy with this implementation I can only say I'm amazed - I find than virtually anything else would have worked better and 'pirates' in SOASE are about as far off the mark as it gets. When, as an interstellar empire, defeating a bunch of space pirates takes a considerable investment into both resources and time (investment into defensive structures + possibly having to leave your capital ship(s) and fleet in the gravity well for the extremely lengthy duration it will take to destroy so many ships) it should serve to alarm to the fact that something is terribly wrong.

2) I've used StarDock to simply mean 'the developers' which as you point out isn't accurate. Personally, I think pirate implementation is 'game breaking'. The idea is great and the gameplay element of being able to launch covert attacks on allies an excellent one, however, so it's a shame the implementation is off.

3) The example I gave was about more than an incorrect calculation of forces: The AI simply should always retreat capital ships that are almost destroyed. It is simply an insight into how simplistic the AI still is which I think is very disappointing for a - dare I say it - mostly single player game that offers skirmish mode exclusively. You can turn up the difficulty, but then it'll just be two capital ships sitting there bleeding to death rather than one.

Good posts. The pirate situation is indeed tricky. I for one thought it would be cool to put bounty on specific targets, but this is beyond the reach of sins I think. Fortunately we have the option of turning them off or modding them.

I totally agree with the AI, it does seem like it could be improved without much difficulty. Retreating single ships (of most of the types) and sending ships to repair bays would be a big help.

Indeed, mercenaries is a much better term for them.

I was just thinking the same thing.  I don't mind what they do and the way that they do it in the current game iteration, but pirate is probably not the best name for it.

Also, my problem with the AI is that it seems to run away too quickly, not that it refuses to retreat, so I think this might be specific to certain game settings  So much so that I tend to plan my assaults with a smaller fleet coming in from one direction, suckering the AI into moving that way, and then hit them from behind with the main battle fleet, forcing them to run through it in order to get to their retreat warp point.  If I can take out their anti-matter reserves as they run by, the number of stragglers that I can pick off (which sometimes includes a titan) makes it all worthwhile.  Or at least really fun.

Hey, you make your own game universe and you can call them whatever you want. I think there is a canon reason why they are called pirates and not mercenaries, perhaps they started out as pirates but quickly grew as before the Vasari came the TEC only had a small military solely dedicated to anti-piracy.

Again I'm not saying it's good, but I try not to get upset over bad names, and again, I think they could do all of your suggestions and I would still usually play with pirates off. I just think it's not a feature worth improving. They're better taking their loses and fixing somethings that are both more important and more likely to be appreciated.

Anything wrong with a feature that is optional is not game breaking, provided said features don't literally crash the game. At least not by my definition or apparently that of Stardock.

Maybe I'm behind the times, but I've noticed this behavior in most RTS games. I think one factor is that Sins combat is extremely slow compared to most RTS. You don't have a chance to retreat individual units in most RTS games. Even say experimental in Supreme Commander are rarely worth retreating, perhaps because of the low possibility of success and the lack of a repair structure like Sins.

I should also point out that's a harder thing to determine than it seems. I'm not an MP player but I do it sometimes. I know the mechanics of the game as good as anyone, but I get beat almost every time. You know why? Because those guys have an intuitive sense of timing, when to build X or retreat cap Y given fleet Z that I don't. Against late game bomber swarms, you have to keep your capitalships at the edge of the gravity well if you want any chance of escape, because the bombers will kill your cap in two or three passes. In some cases you maybe better letting your cap do as much damage as possible than trying to save it.

This, of course, is the difference between tactics and strategy, which is minimized in the "cage match" environment of multiplayer.  In another thread, you said that any strategy that was worthwhile should be successful in multiplayer as well.  But some strategies need time and room to develop, and multiplayer short-circuits this by forcing a focus on the short game and tactical management, as opposed to the long game. 

Instead of a 5v5 multiplayer game, try 1v1 with 8 AI players.  The successful strategy will be significantly different.

I don't really see how this is relevant to the OP's issues but...

Not necessarily minimized. It just inherently different given the human players are competent and can kill you effectively if you don't counter what they're doing. When you choose to tech or eco up or anything else that takes "room to develop", your choice has a risk that your military might not be "good enough" to counter any military action your opponent might do to stop you. Against the AI that risk is essentially zero if you know what you're doing, so you don't really have the weaknesses present in those strategies. Against the pros, who you're right prefer to kill you quickly at any chance they get, that weakness will be exploited immediately.

I do not like much tactical management, but the truth is even in Sins multiplayer is much less tactical than many RTS games (I just got Wargames: European Escalation, and I think that more than anything deserves the title "Real Time Tactics Game", even in single player). While the frontline players have less scope to develop, large team games actually increase the strategic scope of the game as the eco players have to choose between things like boom and feed, titan rush or fleet up and double team. And if you're frontline and manage to mostly defeat the player next to you, you have many more options to pursue, develop, team up etc.

So I'd argue that multiplayer really isn't much less strategic, it is just going to have a much bigger military component than AI games, but that opens up as many new strategies as it closes other ones IMO.

"More strategies" (in terms of quantity) is not equivalent to More actual strategic gameplay.  To examine an extreme, if every single strategy is viable you aren't really making strategic decisions at all.  the essence of a strategic decision is assessing your resources and situation and devising a tactics which is superior to your other options.  

This is not to say there should be "one true superior strategy", but if everything is equally viable your decisions are ultimately meaningless leaving you with only the illusion of choice where all choices are in reality essentially the same.

True tactical gameplay without mindless repetition can only really come from the need to anticipate your opponent and act accordingly- rather then just spam whatever strategy suits your mood regardless of the enemy's actions and expect to be completely competitive.

And frankly no AI is smart enough to create the depth and difficulty of that task an experienced player opponent can provide.  Frankly, even well made AI tends toward predictability after you've played against them enough times.

Even settign aside the topic of "learning the AI", when you play against a human opponent they will be trying to read you just as you're trying to read them- a situation given to evolving tactics I've yet to see any AI reliably replicate.

anyway this has become a bit of a rant, but my point comes down to this:  true strategic decisions need some choices to be better then others, and for those decisions to not become rote repetition the right choice needs to depend heavily on the specific situation and the opponent's motives.  And frankly at the end of the day, in the long run AI will become predictable and fail to read you as an experienced player opponent might, which does inevitably result in some loss of strategic depth.

This is not to universally condemn single-player strategy games, but Sins IMO borrows too much from the RTS genre to have the true strategic depth lay is comp-stomps. 

That said, I understand and respect some people do prefer the pacing of single player games- most notably some people just hate being rushed in general.  But I feel these people tend to vastly underestimate the amount of strategic decisionmaking and counterplay evident in settings where rushes are present.  They assume rushing is "the only strategy" because it happens to hard-counter their attempts to attempt a zero-military all-in long term strategy.  it isn't- there exist late game strategies in this sort of setting as well- it's only that you reliably choose to utterly disregard your defenses while executing said late game builds.

Ok done ranting now for real- over the years I've really begun to tire of the hate the very presence of rush-strategies in the proverbial metagame generates among the players who prefer a slower burn.

No more ramblings, please, you're derailing the thread. This thread is about identifying SOASE's/Rebellion's weaknesses and coming up with solutions to them.

You needed the full quote - the line ended "that I can hire". It's not just about the name, but about the fact that they play out and are as strong as Antareans while still being 'pirates', meaning I can hire them. Either you let there be a neutral faction that occasionally lashes out at empires, MOO2 Antareans style, or you make the 'pirates', a faction I can interact with and control, not ridiculously powerful. You don't create a balance where 'pirates' are so insanely powerful they can directly threaten an empire in open confrontation. This is the crux of the issue. And a game like SOASE that undoubtedly tries to take itself seriously needs to address it.

Same deal - the details don't matter, what I'm addressing is the AI is making too many baby-level mistakes even in this 4th iteration 'polished' version of SOASE. I don't think asking for issues such as capital ships just sitting there letting themselves slowly but surely get blown up to be cleared is asking too much.

Well that's just the thing, to me the pirates really aren't very powerful. They're an annoyance, not a threat. Heck, even their pirate bases are just a fun way to get your titan to level up 3 or 4 times. Yet to others they are an existential threat.

Hey, better AI is always better, I won't argue with you there. I just disagree with your claim that Sins' AI is much worse than other RTS AI. I'll admit, Rebellion's AI is the most... haphazard AI the series has ever had. Sometimes it plays extremely well, constantly raiding you from unexpected directions and making you fight from the get go. Other times it seems quite sluggish. But every game has their own AI weaknesses once you play enough to know them. Even such major series like Total War, with much better Dev teams, persistently make bad AIs, their only saving grace is that they made the AI moddable, so the community could improve it, which is something Sins should have really done in the first place.

[quote who="GoaFan77" reply="14" id="3300796"]
Quoting Apheirox, reply 13You don't create a balance where 'pirates' are so insanely powerful they can directly threaten an empire in open confrontation. This is the crux of the issue. And a game like SOASE that undoubtedly tries to take itself seriously needs to address it.

Well that's just the thing, to me the pirates really aren't very powerful. They're an annoyance, not a threat. Heck, even their pirate bases are just a fun way to get your titan to level up 3 or 4 times. Yet to others they are an existential threat.

True. Some mods I've seen try to correct the issue by having an overwhelming fleet but minus any siege vessels, so the pirates destroy everything in the gravity well but cannot kill any player alone.

I do think improving the AI would be worthwhile, a large part because sins has a large SP base.

The pirates are never a threat. A single scout sends them on a wild goose chase indefinitely. They also die to repair pads. This doesn't change the fact they possess an overwhelmingly large number of extremely heavily armored, high HP ships that will require an unfairly large amount of attention to take down. Pirates. Posseing ships so advanced even the ancient Vasari civilization can only dream. And being powerful enough that if you don't make special preparations, they'll wipe out an entire planet.

Disabling the pirates are unfortunately not a real option in my eyes with factions and technologies that base themselves precisely around there being pirates. Even if disabling them produces a better game it nonetheless means there's an important component of the game you can't make use of.

I never said Sins' AI is particularly worse than other RTS AI though I certainly do consider it on the lower end. I think, however, that given how succesful a game SOASE is and the fact we're up to the 4th commercial version of the game now it's disappointing to see that the same the low-level, basic mistakes still persist.

Not really true, the lack of shields, or rather shield mitigation, makes pirate ships much weaker than equivalent frigates. No amount of HP and armor truly makes up for that.

Truce Amongst Rogues is the main good tech based around them, and it works fine with pirates disabled or not. Few other techs that deal with them are worth getting even when they are on. Likewise, do you feel there is something wrong with long range jump and wormhole techs not being needed on every map?

It's true the pirates have no mitigation - which is a design that also really doesn't make sense to me: Mitigation exists to prevent gameplay being about focus fire to create more cinematic battles. Why shouldn't this include battles against pirates? Instead we have this cheese of ships with an enormous amount of HP hidden behind better armor than the main factions can field until the very late game - it's retarded. I also don't accept your point that pirate ships are weak and easily dealt with - mitigation or no, fact is they send a lot of ships. You can't deal with a pirate attack without making a considerable effort one way or the other, whether it's cheesing them around with scouts, repair pad spam or something else. Pirates require far too much attention for the possibly as low as 250 credits price your opponent paid to send them at you. 250 credits to field a force that will eradicate a whole planetary system if you don't spend a good deal of both time and money is simply silly.

As for the point about wormholes, long range & wormhole are techs all factions have, that's the big difference. The more important reason for why I dislike disabling pirates, however, is that it takes away the possibility of stabbing your friends through in the back which I think is an ingenious feature more RTS games should adopt. This is the part of 'pirates' that works really well and where SOASE gets well-deserved kudos. If only the rest was as well done it'd be truly great (hence my motivation for posting this thread).

Anyway, I'm going to abandon this discussion. I've made my point - I think the implementation of pirates is hilariously (not really) bad and would greatly appreciate seeing it changed. So, this has been the Nth thread about pirates which most likely will be ignored - but at least the effort was made. So, thank you for your input, Goa, you've made some good points.

I instead have another topic to debate: Fleet upgrade costs, or more precisely: the capital ship supply part of it. Is it really fair these techs are so expensive? It seems to be an established fact that one gets more bang for the buck using regular frigates and cruisers than capitals. The latter can of course level up, but is it really fair and sensibly balanced that you have to pay not only the 3000 credits + extras for the capital ship itself, but also the extremely steep cost for the research (IIRC, around halfway through the tree the research costs more than a whole ship by itself)?

Oh I totally agree with this. Well okay not the eradicate a planetary system part, but I'm not a huge fan of the periodic raid system (we can directly hire the pirates whenever we want now, why do we have this random raid feature still?). I've done by best to mod it so that the force pirates send against you is mostly determined by current bounty (by default it goes up to at least a tier 3 raid just with time, even if bounty stays at zero).

Fair enough.

If I were making the game from scratch yes I'd make them cheaper. Titans helped by requiring more command points, but as it is in serious MP games (and even most of my single player games) you rarely need more than the first 4 upgrades. Sins already strongly encourages having a few, high level capitalships than lots of low level ones. Only in those long 8 hour multistar games might you have enough level 10 capitalships to make it worthwhile to build more, but at that point cost probably isn't much of an issue anymore.

My only point against your argument is the bold, but at that point you get more than 1 capitalship point per research, so if the tech cost 4000 credits and unlocked two capitalships worth, it's only 2000 credits per capitalship.