AI SuperTerminal V 2.801

[Oewyn]

Couple bugs w/ them.

1) Waves start as soon as construction of an orbital begins, not when it completes

2) Can be shield-boosted (Like, I suspect, many any other non-turret that has shields)

3) (perhaps more of a suggestion than a bug) Probably a bit too difficult for the payoff (I had maxed out MK I-III Basic turrets, paralyzers and tractors in a shield ball, and only was able to get about 70 before I was completely overwhelmed.  Ideally, without all the setup, one should be able to get 30-40 (with a reasonable setup)before they have to destroy it.  Probably the frequency of waves should be a multiple of the 15 second tick, or there should be longer ticks with more AI progress/tick and/or a shallower difficulty increase per tick.

[Oewyn]

Okay, I take back a bit about what I said for #3, it looks like you can get 30-40 AI reduction before a mix of about 150 MK I/II basic  turrets succumb.  Even so, i think maybe halving the spawn rate and doubling the number of ships may be a bit better.

[x4000]

#1 and #2 are fixed for 2.900.  For #3, I'm not going to mess with that anytime soon because I really want more data first.  I'd rather having something like that be too weak (and thus not useful) than too powerful (and thus completely game-breaking).  Right now if it is a bit on the weaker side, that can be resolved later more easily.  On the other hand, I suspect that some people will figure out some clever ways to exploit these -- with golems, high level ships, something -- that will make them moderately-powered in normal circumstances and high-powered if you play your cards carefully.  So, this is something where I just want to wait and see for now.

Incidentally, regarding your turret ball under the force fields -- that's really shooting yourself in the foot, a bit.  The attack power of ships shooting out from under force fields is reduced so much that it's going to make it harder to defend.  Better to have the FF over your command station and maybe a dock, then have the turrets arrayed in front of it with the dock also churning out ships to fight in front of the FF.  I'd imagine that you would get more success from that alone.  Also, potentially "spawn camping" the enemy units as they come out of the superterminal might work well.  And, this might be one of the most useful places ever for lightning turrets -- just build a little forest of those around the superterminal, potentially, and that might make a huge difference.

It's potential exploits like that which make me worry that superterminals will turn out to be too powerful in the wrong hands even as they currently are.  I suppose we'll see.  I'm going to move this to strategy discussion, so that people can comment on #3 as they encounter superterminals.

[eRe4s3r]

If you put turrets under shields you lose 75% damage output too

I will report back here how my super terminal escapade goes in my current game i have one. But i have fortresses and grav turrets... which might be better performing

What kind of ships did come at you? What numbers?

[RCIX]

#1 and #2 are fixed for 2.900.  For #3, I'm not going to mess with that anytime soon because I really want more data first.  I'd rather having something like that be too weak (and thus not useful) than too powerful (and thus completely game-breaking).  Right now if it is a bit on the weaker side, that can be resolved later more easily.  On the other hand, I suspect that some people will figure out some clever ways to exploit these -- with golems, high level ships, something -- that will make them moderately-powered in normal circumstances and high-powered if you play your cards carefully.  So, this is something where I just want to wait and see for now.

Incidentally, regarding your turret ball under the force fields -- that's really shooting yourself in the foot, a bit.  The attack power of ships shooting out from under force fields is reduced so much that it's going to make it harder to defend.  Better to have the FF over your command station and maybe a dock, then have the turrets arrayed in front of it with the dock also churning out ships to fight in front of the FF.  I'd imagine that you would get more success from that alone.  Also, potentially "spawn camping" the enemy units as they come out of the superterminal might work well.  And, this might be one of the most useful places ever for lightning turrets -- just build a little forest of those around the superterminal, potentially, and that might make a huge difference.

It's potential exploits like that which make me worry that superterminals will turn out to be too powerful in the wrong hands even as they currently are.  I suppose we'll see.  I'm going to move this to strategy discussion, so that people can comment on #3 as they encounter superterminals.

If I'm correct the waves from it act like they're coming out of a wormhole? If so, then a fw hundred grenade launchers + 5 light starships to boost their attack to awesome levels should clean them up nicely...

[Oewyn]

#1 and #2 are fixed for 2.900.  For #3, I'm not going to mess with that anytime soon because I really want more data first.  I'd rather having something like that be too weak (and thus not useful) than too powerful (and thus completely game-breaking).  Right now if it is a bit on the weaker side, that can be resolved later more easily.  On the other hand, I suspect that some people will figure out some clever ways to exploit these -- with golems, high level ships, something -- that will make them moderately-powered in normal circumstances and high-powered if you play your cards carefully.  So, this is something where I just want to wait and see for now.

Incidentally, regarding your turret ball under the force fields -- that's really shooting yourself in the foot, a bit.  The attack power of ships shooting out from under force fields is reduced so much that it's going to make it harder to defend.  Better to have the FF over your command station and maybe a dock, then have the turrets arrayed in front of it with the dock also churning out ships to fight in front of the FF.  I'd imagine that you would get more success from that alone.  Also, potentially "spawn camping" the enemy units as they come out of the superterminal might work well.  And, this might be one of the most useful places ever for lightning turrets -- just build a little forest of those around the superterminal, potentially, and that might make a huge difference.

It's potential exploits like that which make me worry that superterminals will turn out to be too powerful in the wrong hands even as they currently are.  I suppose we'll see.  I'm going to move this to strategy discussion, so that people can comment on #3 as they encounter superterminals.

That turret ball is actually tractor turrets.

[x4000]

Oh.  Wow.  Well, that's good, then -- I worry about this being overpowered, so clearly the main strategies (even extreme variants of which, such as your huge mass of turrets and tractors) are not too over-effective.  I still do worry about some of the other possible exploits, though, so let's give it a bit of time.  But thanks for the clarification -- that certainly changes the immediate equation by quite a lot.

[Oewyn]

Yeah i'll see how well i do once I finish my planetary shield inhibitor and super fortress :-)  speed +10 FTW.


BTW any reason it's capped at +10?  I saw that the simulation runs at like +16k or so.

[x4000]

I capped it at 10 simply because on most modern hardware, when you are also rendering the video for the game you get almost no performance benefit beyond 10 (really, not much beyond 6 or 8 in most cases).  So rather than having users cranking this up to 100 pointlessly or something, and then having to hit it 100 more times to get back to zero, I just capped it at 10.  Given that most strategy games I'd played recently capped it at 4 (to my recollection), I felt pretty okay about it.

The simulation does run at +1600 speed, but honestly that probably isn't much different from +60 speed on modern hardware.

[Oewyn]

Realistically though, I doubt that it's worth the 5+ hours of game time i have put into building up the defense.  I probably could have built it faster if I had more than my home planet's worth of resources to build off of, but then again, so too would I have to defend more wormholes. 8 hours worth of AI progress is already 96 AI progress, so I'd have to last until the reduction is ~200 to even pay for itself.


And picked up a MK 2 radar jammer, that should help things, but prolong the waiting several more hours.

[x4000]

Yeah, I wouldn't really be one for waiting that insane sort of timespan particularly, unless you're just wanting to see how it goes.  It's really meant to have tons and tons of engineers stacked on it so that it takes way less time (but of course costs a pretty penny, so you'd have to have a huge resource income already, or you'd need to be background building the expensive stuff while doing something else -- kind of the same idea as Experimentals in SupCom).

[Oewyn]

Yeah pretty much the best I was able to do without scrapping my orbital to "pause" the ST and rebuilding was 120, even with radar jammer, shield inhibitor and superfortress.  In the end it's the lightning shuttles that did my turrets in.  They each get 1 volley before they die, so eventually they take out all of my turrets.  I was able to get to around 212 or so reduction after rebuilding, but even after that, it was after ~ 15 hours of game time.

Of course, it was all for curiosity's sake, and since i was running at 10x speed, it was more like a 2-3 hour experiment rather than a 15 hour ordeal.

One question I did have is, the ST wave sizes is based off of how much AI progress the ST has reduced, not the total reduction, correct?  Otherwise these are pointless after a few data center destructions as anything greater than say 40 AI reduction is impractical.

[x4000]

Interesting results, there -- that is good data, I'm still reluctant to change them just yet, though.  Just too risky for today's release.

Their reduction is based on the total reduction so far, not based on theirs alone, so if you find one super late in a game that is a ton less useful, yes.  This prevents compounding these too terribly much with other AIP reducers.

[Oewyn]

If you find one after destroying 1 or 2 data centers, then it will basically be impossible to use at all.

I would recommend having it's own internal timer, and the wave sizes are based off of that.  With the plethera of data centers in an average 80 planet game (15ish?) already reducing the AI progress by up to 300, being able to actually use a superterminal in the mid/end-game when you actually find one doesn't seem like such a bad idea.  It's probably too risky to push it to the 3.0 release, but something to keep in mind for the future.  Maybe some sort of compromise where the "tick" level starts at AI reduction/10 and increases normally afterwords.  This way you are penalized some for already having a high AI reduction, but not so much that the ST becomes useless after a few data centers.

[eRe4s3r]

Reduction is based on OVERALL reduction so far? You mean when i destroyed 7 datacenters i can not use the super terminal at all?