Exploit List - Restricted Tactics for Imperia

[Catwalk]

I'm flattered Ugh, need to resist the temptation to start on Master of Orion as well. I'm deeply engrossed with a MoM mod project at dragonsword.com which is turning out well, aided by some excellent editors that have been created recently. But yeah, I'll probably give in to temptation and try my hand at an Imperium sooner or later. Would I be allowed to permit the use of kyrub's patch for it? :P

I do think it's clearly an exploit (with significant consequences in Small and Medium games) since it's something that takes advantage of a flawed AI as well as faulty game mechanics. It's worth noting that in MoO II you cannot repel enemy ships with unarmed ships, and you can colonize planets with enemy scouts in orbit. You have a good point that disallowing this would also disallow a perfectly valid strategy, so my proposal isn't the best. I might just go ahead and copy your concept if I ever get around to hosting an Imperium, probably with Alkari instead.

[RefSteel]

But yeah, I'll probably give in to temptation and try my hand at an Imperium sooner or later. Would I be allowed to permit the use of kyrub's patch for it? :P

If kyrub has a version of the patch he that's stable enough to use for the games, I don't see any reason why we can't try an Imperium with it. Does anyone else have an opinion?

I might just go ahead and copy your concept if I ever get around to hosting an Imperium, probably with Alkari instead.

There are other alternatives too, that make avoiding it a natural consequence of a larger variant/scenario. In fact, I'm thinking about one now that might be interesting to try soon....

[kyrub]

First, I would be honoured to see the patch used in this way.

Sadly, there's no currently usable version.
Embryonic version is reported to be stable, but it has the "eternal AI alliances" side effect.
Larval version has massive stability issues.

Any voting is premature.

[ignatius]

I used the embryonic_18 in all my MoO games I did since it came out - including rbo27 and the current rbo29 (which I hope to finish in time for tomorrow) and never encountered a problem. The game feels much saner and less erratic with it - and you get a lot of interface improvements (research leds, auto-save, etc.). The AI attacks are less frequent but much more dangerous when they happen which I consider a good thing. Also, I didn't notice any "eternal AI alliances" problems.

So regarding the patch, I would would definitely say: go for it!

as for the scout issue, cat, you really seem to like it the hard way, not just in MoM! ;-) I side with DWiggles on the issue - it's a matter of game balance: The AI get's 3 (4 iirc with the embryonic patch) free colships pretty much from the start and the scout behaviour is clearly meant to counter this. Also, if both ships are unarmed, you have to resolve the matter _somehow_, so either the AI or the player has to retreat (unless a NAP is in place) - deciding one way or another is an explicit design decision and not an exploit. Anyway, after a few decades, both things are a non-issue.

A ban would only force the player to bother with single 2-laser LR-mediums for the same purpose - I don't see how this would make the game more fun to play.

[PlzBreakMyUmMap]

I used the embryonic_18 in all my MoO games I did since it came out - including rbo27 and the current rbo29 (which I hope to finish in time for tomorrow) and never encountered a problem. The game feels much saner and less erratic with it - and you get a lot of interface improvements (research leds, auto-save, etc.). The AI attacks are less frequent but much more dangerous when they happen which I consider a good thing. Also, I didn't notice any "eternal AI alliances" problems.

So regarding the patch, I would would definitely say: go for it!

as for the scout issue, cat, you really seem to like it the hard way, not just in MoM! ;-) I side with DWiggles on the issue - it's a matter of game balance: The AI get's 3 (4 iirc with the embryonic patch) free colships pretty much from the start and the scout behaviour is clearly meant to counter this. Also, if both ships are unarmed, you have to resolve the matter _somehow_, so either the AI or the player has to retreat (unless a NAP is in place) - deciding one way or another is an explicit design decision and not an exploit. Anyway, after a few decades, both things are a non-issue.

A ban would only force the player to bother with single 2-laser LR-mediums for the same purpose - I don't see how this would make the game more fun to play.

The last part is absolutely right. Why not just make the patch detect who is 'arriving' and who is 'defending'. Then make the AI sit for 50 turns where the arriving ship will flee if it decided to fight. It only slows down the beginning a bit and then only if the player wants it to.

[RefSteel]

So after reading the Imperium 41 reports (as well as remembering some of the late-30s ones) I'm tempted to promote the Empty Threats guideline to a hard-and-fast "Never threaten to attack" rule. The problem is that you can get a lot (techs, cash, guaranteed auto-retreats even from your unarmed scouts even by massive megafleets) for no investment, and even get these things repreatedly; the only "balancing" effect is the small possibility that the AI will instead declare war in response, but this is too random to be really balanced. Getting good results from a threat is a measure of luck rather than skill, and its impact on the game can be very strong, especially if intentionally abused.

Note this was intentionally left on the table and an absolutely legitimate tactic in Imperium 41, which specifically dictated that all exploit rules be ignored. I'm bringing it up for purposes of future games only. For that matter, I'm not planning to just put a new rule up there as a unilateral action; does anyone have an opinion, for or against?

[toddestan]

So after reading the Imperium 41 reports (as well as remembering some of the late-30s ones) I'm tempted to promote the Empty Threats guideline to a hard-and-fast "Never threaten to attack" rule. The problem is that you can get a lot (techs, cash, guaranteed auto-retreats even from your unarmed scouts even by massive megafleets) for no investment, and even get these things repreatedly; the only "balancing" effect is the small possibility that the AI will instead declare war in response, but this is too random to be really balanced. Getting good results from a threat is a measure of luck rather than skill, and its impact on the game can be very strong, especially if intentionally abused.

Note this was intentionally left on the table and an absolutely legitimate tactic in Imperium 41, which specifically dictated that all exploit rules be ignored. I'm bringing it up for purposes of future games only. For that matter, I'm not planning to just put a new rule up there as a unilateral action; does anyone have an opinion, for or against?

In my opinion, the "Threaten to Attack" functionality is more or less broken. It's just too easy to use it to get free techs, money, and gauranteed auto-retreats from opponents far stronger than the player. While I would expect that one could bully around smaller empires, it just seems that the AI is all too eager to appease the player when the player realistically cannot back up a threat in any meaningful manner.

I guess the question I have though, does this balance out other aspects of the game? I find that the AI's "cold war" tactics can be rather obnoxious and do get old after a while, where they attack my fleets/invade my worlds/bomb my planets while not at war, but if I try to do that to them they pretty much will declare war right away. Threatening to attack is about the only way I have to say "hey, knock that off!". Take that away, and about the only thing I have left is either to retaliate (which will pretty much gaurentee a hot war), or suck it up and take it. However, with that said, I still feel it's broken in the sense that if they feel strong enough to attempt one of their blatant land grabs, I would expect that they would also ignore any threats to attack.

So in conclusion, I would support promoting the rule to "never threaten to attack" (specific imperiums, of course, may choose to suspend this rule). It's already something I rarely use myself, because I already consider it kind of broken.

A compromise: You can only threaten to attack when relations are good (say amiable or better, but this could be adjusted). This rule could work because if you have good relations, an opponent would be more likely to back off in order to maintain the good relationship and any trade agreements. Also, because threatening causes a relations hit (correct?) you can't just repeatedly threaten the opponents to get free stuff. Thoughts?

[RefSteel]

I guess the question I have though, does this balance out other aspects of the game? I find that the AI's "cold war" tactics can be rather obnoxious and do get old after a while, where they attack my fleets/invade my worlds/bomb my planets while not at war, but if I try to do that to them they pretty much will declare war right away. Threatening to attack is about the only way I have to say "hey, knock that off!". Take that away, and about the only thing I have left is either to retaliate (which will pretty much gaurentee a hot war), or suck it up and take it.

Well, there's always stopping them. I don't need to threaten the enemy if my fleet (and bases as my planets and techs mature) can destroy their probing missions and cold war attacks in the first place ... and if I don't have the fleet to do so, even with potential missile base support, why would they feel threatened by anything I say? If I have to defend all my holdings, I need to figure defensive fleetbuilding into my plans instead of running a flat-out farmer's gambit.

So in conclusion, I would support promoting the rule to "never threaten to attack" (specific imperiums, of course, may choose to suspend this rule). It's already something I rarely use myself, because I already consider it kind of broken.

Yeah, I'm kind of leaning this way as well (not just since Imperium 41; I was thinking about it when I introduced a scoring penalty for threats in Imperium 37, and that one remains my favorite Imperium of all time).

A compromise: You can only threaten to attack when relations are good (say amiable or better, but this could be adjusted). This rule could work because if you have good relations, an opponent would be more likely to back off in order to maintain the good relationship and any trade agreements. Also, because threatening causes a relations hit (correct?) you can't just repeatedly threaten the opponents to get free stuff. Thoughts?

If relations are Relaxed or better though, you can generally get a NAP. While the NAP is in force, the AI will never send a fleet to any of your worlds, so you'll only run into problems (when they end the NAP or) when you both have a fleet heading to colonize/explore/attack the same unclaimed/enemy world and you get there first. I could see a change to close that loop-hole though: "Threats may be issued only to a race with whom we have a Non-Aggression Pact, and which nevertheless has a fleet that appears poised to attack one of our worlds."

Any other opinions?

[Ianus]

I've been thinking about this for several days trying to quantify my position. My first instinct is to strongly dislike the idea of losing the option to use threats as I have always used them as a counter to the AI's ridiculous early game antics, especially given the ridiculous strength of Impossible fleets in the early game. Also I have never bought how broken everyone says diplomacy is. But then last night I was doing some play testing and observed the following:
In the early game ~2375-ish I was carrying out my usual land grab when I observed a substantial Alkari fleet incoming to a roughly half-built colony. I threatened the AI, who responded by comparing my threats to the braying of a Rigellian jackass (I love this game!), withdrawimg his ambassador and pressing the attack. I rushed a single base, but it wasn't even close. The AI (Agressive something, Expansionist maybe) gained orbital superiority and sent invaders. Three turns later, with a large fleet in orbit and marines incoming the Alkari ambassador returned. I threatened him again having literally no other option, no fleet, no way to save my planet, and the AI agreed to withdraw his ships in the face of my completely empty threats. I've always observed that fleet action (even unarmed scout blockades) seem to bolster the player's ability to make successful threats and reasoned that even empty threats don't appear so due to early game shows of strength, but in this case there was absolutely no way they should have backed down.
So yeah I'd say that Threats are broken, and as much as if pains me I have to agree that they should be placed on the restricted list.

[RefSteel]

Mmf. Good points; the rule needs to be rewritten for clarity and to close a couple loop-holes. That said:

The rule is agnostic about the reason you chose to retreat. As written, if you expended any ammunition at all (even one bombing pass of ten) you can't return directly to the same star from which you retreated.

Bouncing back and forth between two enemy worlds is legal according to the rule as written now, as is retreating a severely damaged ship and returning it "good as new." These are against the spirit of the rule though, and I'll change it to exclude them unless others think there's some compelling reason not to

The new rule would be, "Ships that retreat after firing expendable weapons (e.g. bombs, missiles, spores, etc.) or after taking damage that has not been fully healed by auto-repair may not be redirected to the same star, nor to a rival's colony."

Even if folks approve, it obviously would not apply retroactively. In any case, what do you think of the change?