Forts in times of distributed defenses

[keith.lamothe]

Well, I probably wouldn't make any such time-is-costly change completely non-optional.  But opting out would be more like toggling on Lazy AI rather than setting Auto-AIP to zero.  I don't get a lot of balance feedback from cases with the former, but people expect the latter to be fully balanced, because it's the norm

[Fleet Unity]

Well, I probably wouldn't make any such time-is-costly change completely non-optional.  But opting out would be more like toggling on Lazy AI rather than setting Auto-AIP to zero.  I don't get a lot of balance feedback from cases with the former, but people expect the latter to be fully balanced, because it's the norm

If you do change it please let it be a choice as yes I like to play with auto progress set to 0 but I like to go at my own pace and not having the game say "Hurry Up" I do use the spire leaders however it takes a while to free them so I do get +10 every hour so theirs that.

Edit: I also use mining golems and the neinzul missile silos as well as dark spire ect.

[Kahuna]

Another idea that comes to mind is to make AIP-over-time into AIP-Floor-over-time, so it doesn't hurt as long as you keep up the pace.

Dooo iiit!
EDIT: No sarcasm intended. I would actually use this option.

[Peter Ebbesen]

Turning AIP-over-time into AIP-floor-over time sounds neat, but only if the top panel was adjusted such that both values were clearly visible at all times so you didn't have to inspect the AIP tooltip to see the AIP floor, but could at all times see the floor and would at all times be gently reminded that it was increasing.

[Hearteater]

I don't really feel +AIP (or AIP floor) over time is a penalty that is felt by players quickly enough for this purpose. I think AIW has enough "you've made a mistake, wait an hour and you'll see how bad it was" penalties.

I also dislike AIP floor just because it complicates what should be something simple (AIP), and uses a technical math term to describe itself. Not to mention how weird it makes Super Terminals work (and hey, ST + Hacking Pool is another weird interaction). If AIP floor could go away in favor of something that accomplishes the same effect I'd much prefer that. Change the functionality of -AIP structures could probably allow AIP floor to get removed.

Random brainstorm: "Keep" AI floor, but let's rename it Alert Level. Instead of using it as a floor, when AIP < Alert Level, the AI is on alert because something funky is happening that shouldn't. Let it sound out something mean from the homeworld every 5 minutes or so. How about an "Investigator" which is a mini-Hive golem combined with a Beachhead, that picks a small escort and goes after random player worlds.

A little rambling, but I blame it on lack of lunch.

[Toranth]

I used to play with 1-AIP-per-30-minutes as the base AIP over time.
Then, for some reason, I ended up with 0-AIP over time (probably because I'd been using Spire Civilian Leaders) and noticed that the game felt exactly the same.

Most of my games run 20-25 hours.  That's 40-50 AIP added on over the course of the game.  My usual gameplay style is step 1: explore, step 2: kill all the AIP reducers, step 3: kill all the CSGs, step 4: wildly expand to grab fabs and hack stuff in the last hour before attacking the AI HW, step 5: attack both AI HWs.
So that extra AIP from time basically has no effect except for the first hour or two (when it is too small to matter) or the last hour or two (when it means a minor increase in size for a few waves, which are likely to get warhead-ed anyway).

No, what I find to be the far more important time factor is the CPAs.  They launch at regular intervals, and they are the usual source of death for me.  If you really want to apply a time time-pressure factor, just make the per-CPA size increase rate higher.

[Kahuna]

Super Waves

Aka salvage on steroids?

A little rambling, but I blame it on lack of lunch.

Lol

I like the idea. Except the "Hive Golem" part of it.

what I find to be the far more important time factor is the CPAs.  They launch at regular intervals, and they are the usual source of death for me.  If you really want to apply a time time-pressure factor, just make the per-CPA size increase rate higher.

That's why I keep telling people who have "Netflix time" to increase the difficulty level.

[TechSY730]

Excellent points about the auto AIP (I play with +1/(30 minutes)), shark plots, and CPA sizes. I forgot about those in my previous analysis. Since only auto AIP and CPAs are enabled by default, and only CPAs are not disablable, I'm not sure if this is enough of a time cost to penalize excessive use of "patching up economy by waiting".

Matter converters now have a limit per planet. 10, IIRC. So you still do have an absolute energy cap at any stage in the game. Now, maybe 10 is too much (yea, it can be limiting on more "involved" scenarios like Fallen Spire or ultra low AIP clears, but what about "normal" play?), but at least there is something in place to keep energy vs metal (read, time) tradeoffs from going too out of hand.

I am not a fan of the shark plot (Don't get me wrong, I don't think it should be removed, but I just don't play with it). The AI getting a one time "boost" for taking out something of the humans just feels too artificial for me. As the shark plot (either A or B) as they are now, I would not want them in the base game.

Maybe if AI planet retaking can be implemented, maybe a shark like thing can be added then to the base game (and turn off-able?). Like, when the AI takes out your planet and (at some point) retakes it, it gets a surge in defensive "points" for that planet as well as an offensive "burst". The rationale is that the AI is "spending" in a "burst" to re-secure their planet, both by ensuring that the new planet is well defended, and the surrounding planets are weakened some to reduce the chance that the human can fight back while the planet is weaker. Sort of mimicking what humans do when they take a new planet. Yea, this would be in many ways the same, but it would "feel" more natural. Also, the offensive burst could come while the planet is still in the process of retaken, or even before that (like the AI is planning on retaking the planet), to make the effects a bit more immediate. It would also help out cleaning up the planet they would want to retake if there is still lots of human stuff on it.

If something like that happened, I would not mind it in the base game, even if opt-outable. But, as I stated before, with the shark plot (either A or B) as they are now, I would not want them in the base game.


I have no opinion on changing auto-AIP over time to auto-AIP floor over time. I don't think the AIP-floor needs changing; the concept is intuitive enough IMO. The current tooltip paragraph describing AIP floor is a bit more verbose than it needs to be IMO, which makes it seem less intuitive than it really is.

Also, the current tooltip for AIP floor still says the "suggested" is +1 per 5 minutes, which is way too high of a rate, and isn't even the default.

[Hearteater]

I don't like the AI retaking planets, or defensive buffs. All those do is slow down the game. AI retook a planet? I just take it back. Assuming I even care. It isn't like the AI gets a benefit from having the planet. It only matters for the obstacle it is to me. Say I take a CSG planet, but it isn't on the path between me and a home world. Do I even care that it gets retaken after I leave?  No, because it literally has no impact.

[Nodor]

I might be thinking about this backwards, but with the distributed defenses, the waves got a LOT meaner.  Reprisal waves on Difficulty 8+ are not a joke.  You need your chokepoints to have backups because when those reprisals arrive your fleet isn't going to be "readily available". 

Making more defenses distributed is going to require increasing the CPA and reprisal waves to even greater numbers so that they actually threaten players like they are supposed to.

I would prefer to return to a world with smaller waves, and per galaxy turret caps (even if it's just sniper/sprider).   But given this discussion, that doesn't seem to be one of the options.

[Hearteater]

I have trouble telling who has played with per-galaxy turrets and who hasn't. Obviously if you haven't it is hard to imagine that world. But going back to the time when we had per-galaxy caps, the only  reason I can recall anyone arguing for per-planet caps was because on some map types it is impossible to choke point. This made it difficult to do high difficult games on those maps. But that arguement was made from a high-level challenge perspective.

Now it seems the argument is per-planet is either "needed" or wanted. It clearly isn't needed as previous versions can attest. Reprisal waves don't change that. The "want" appears to be play styles that involve largely taking nearly the entire map as normal, which runs counter to the entire premise and design of AIW. Sure, its fine as a special challenge, but if the game supports and allows it as a standard play style, that's how people will play. And then they will direct the game balance around large scale conquest which isn't AIW.

But maybe the direction of AIW is changing.

[keith.lamothe]

The switch to per-planet turret caps happened fairly late in the 7.x cycle.  It was largely an evolution of the core turret controllers from the previous expansion, which led a number of players to find that distributed-defense was actually more fun/viable than expected.

It certainly isn't intended that you be able to take the whole map or anything like it (outside Fallen Spire or very low difficulties).  To the extent that that is possible (which has varied significantly through various rule changes), it's an imbalance.

On the other hand, I think it is valuable to be able to translate extra territory into a stronger overall defensive position.  I also think it's valuable to not _have_ to play ultra-low-AIP and/or exo-grade-chokepoints on the high challenge levels (9+, say).

It took quite a bit to get away from "chokepoints are optimal, all day, every day".  Interestingly enough it appears we have done so   But it's not surprising that doing so involved pushing hard enough that the pendulum went too far in the other direction.

[Nodor]

The switch to per-planet turret caps happened fairly late in the 7.x cycle.  It was largely an evolution of the core turret controllers from the previous expansion, which led a number of players to find that distributed-defense was actually more fun/viable than expected.

It certainly isn't intended that you be able to take the whole map or anything like it (outside Fallen Spire or very low difficulties).  To the extent that that is possible (which has varied significantly through various rule changes), it's an imbalance.

On the other hand, I think it is valuable to be able to translate extra territory into a stronger overall defensive position.  I also think it's valuable to not _have_ to play ultra-low-AIP and/or exo-grade-chokepoints on the high challenge levels (9+, say).

It took quite a bit to get away from "chokepoints are optimal, all day, every day".  Interestingly enough it appears we have done so   But it's not surprising that doing so involved pushing hard enough that the pendulum went too far in the other direction.

I like having the option for (some) distributed defenses.  I can trade 20 AIP progress from smacking warp gates on 4 planets for an extra planet with shiny toys.

I do not want to swing the pendulum towards more distributed defenses at this time.

During normal playthroughs, I don't tend to take "everything" but during multi-player games there is usually a coherent consistent chain between most of the owned planets. (Coordinating transport movement through AI worlds tends to involve hiccups when you have 4 people drinking and playing AI War. )

[Peter Ebbesen]

I have trouble telling who has played with per-galaxy turrets and who hasn't. Obviously if you haven't it is hard to imagine that world. But going back to the time when we had per-galaxy caps, the only  reason I can recall anyone arguing for per-planet caps was because on some map types it is impossible to choke point. This made it difficult to do high difficult games on those maps. But that arguement was made from a high-level challenge perspective.

Not quite.

Per-galaxy capped turrets strongly supported making a few chokepoints or even a singular chokepoint as the approach to defence.

At the same time it seemed thoroughly ridiculous that the turrets you spent expensive knowledge on unlocking were galaxy-capped, while those you got merely for occupying Turret controllers (or hacking them) were not only much stronger, but per-planet capped.

Thus distributed defence was possible essentially at the whim of the RNG, with players (unsurprisingly), building mark V turrets they'd unlocked for distributed defences if they were so lucky to easily get their hands on any, and not if not.

All taken together, this led made chokepointing and "tank everything on the homeworld" approaches superior defences to most other approaches.

This led to the suggestion to switch the caps, such that those unlocked by knowledge under player control were per-planet, while the controller turrets were made per-galaxy instead. (And the size of the caps would switch too, such that it were controller turret caps that were twice the size of the unlockable caps.)

This would lead to distributed defences being available as a deliberate strategy by the player based on unlocking turrets with knowledge rather than something intermittently possible depending on map layout and controller positions.

The suggestion came up several times, and in the end, most of those discussing it were in favour, and Keith started the magic.

This is the major thread where we discussed it prior to implementation. It was implemented in 7.032 (Peaceful Interlude).

Later on, mark IV and mark V turrets were made per-planet too (7.041) as further balancing.

It has been clear throughout the process that rebalancing of energy costs may be needed depending on how this played out in practice.

The balancing of the defences and their costs may be out of skew with desired defensive strength at the moment, but the reasons for per-planet caps on the unlockable turrets remain as valid now as they were then.

[Peter Ebbesen]

Now it seems the argument is per-planet is either "needed" or wanted. It clearly isn't needed as previous versions can attest. Reprisal waves don't change that. The "want" appears to be play styles that involve largely taking nearly the entire map as normal, which runs counter to the entire premise and design of AIW. Sure, its fine as a special challenge, but if the game supports and allows it as a standard play style, that's how people will play. And then they will direct the game balance around large scale conquest which isn't AIW.

But maybe the direction of AIW is changing.

Wouldn't be the first time.

As an example, something that amused me greatly when I picked up the game earlier this year (with per-galaxy caps, thank you) was reading on the Wiki how at one time (2010) reaching 400-600 AIP in the endgame was considered entirely normal by whomever wrote that article, with gameplay advice being to seize some 20-30 planets on an 80 planet map, but this advice had been stroked out to state instead that in the 2013 state of the game 400 AIP was very high and might not be survivable if you hit it before hitting the AI homeworlds, ending with the sensible advice to check the forums to find out what level of AIP was considered normal at any given time rather than trusting the advice on the wiki on this particular point.

What became clear to me reading that was that AI War did not have one fixed goal with respects to how it should be played, but had evolved significantly over time and was likely to continue evolving, with overarching goals gently keeping things under some sort of control.

When I then examined all the setup options, it became clear that whatever today's "normal" AIP and "recommended approach" was, it was likely to be tomorrow's old story, with players widely encouraged to use the wealth of difficulty settings, AI plots, etc. to get the level of challenge they preferred while using their preferred playing style.

Want to play minimum-AIP? There are settings there to make that a fun game. Want to feel time pressure? Ditto. Want to play conquest oriented? No problem. Want some weird mix where champions dominate? Sure. And so on and so forth.

I have no idea how AI War will play a year from now when compared to today. But I bet it will be interesting to find out.