Author Topic: Chokepoint Balance  (Read 20263 times)

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,570
Chokepoint Balance
« on: March 18, 2013, 08:08:08 pm »
There seems to be a general feeling that chokepoints are a bit too "useful" in the current balance. Or possibly, it is that "distributed" defenses are underpowered at the moment.

If this is true, this would be messing with the balance of things like exos, capturables, defensive structures, threat, threat-fleet, and a whole host of other things. Complicating matters is that some aspects of those "other things" are pushing towards having a few chokepoints.

Now some good work has been done on this. The mini-forts and the wave time/size adjustments based on "in-points" have made multi-chokepoint and no-chokepoint style strategies finally viable (note, not necessarily balanced, but viable) . However, when using such a strat, things like exos and defending capturables still push one towards a more chokepoint style gameplay. The argument being that this "push" is too much, to the point of making it impractical not to if you want those other things in play. (Yea, I'm not expressing myself very well)

So, what do you guys think? Do you think single-chokepoint is OP? Or do you think single-chokepoint is fine balance wise, but multi-chokepoint is UP? What about no-chokepoint?

Also tying in this discussion is whether "defense in depth" is sufficiently rewarded, beyond merely having some stuff on the homeworld just in case the worst happens.


My position? I don't know yet. I do think that more options like the mini-fort may be worth pursuing, and maybe nerfing exo size when they are going after a capturable rather than the HW, or possibly even nerfing exo sizes overall. Or even applying some sort of "inpoint" tradeoff style thing to exos. However, I don't have enough hands on experience with >2 chokepoint or no chokepoint style strategies to really give an informed opinion on their current balance.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Chokepoint Balance
« Reply #1 on: March 18, 2013, 08:28:51 pm »
I think the main issue is that the AI's challenge on higher difficulty levels has grown to be balanced around the kind of defense that a player can put up in a small number (1 or 2) of chokepoints.  Therefore, if you don't do that, you're often at a disadvantage in terms of concentration-of-force.  Not always a decisive disadvantage, but it's there.

As a consequence, any change that would make distributed-defense more necessary (not just desirable, but necessary) is going to seem fairly punishing ;)
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Eternaly_Lost

  • Sr. Member
  • ****
  • Posts: 336
Re: Chokepoint Balance
« Reply #2 on: March 18, 2013, 08:35:58 pm »
Here how I view it.

You have structures that you can't replace if you lose them. (Factories, Fabs), You have a limited number of defense structures that are on a global scale rather then a local scale. The AI is limited to attacks along wormholes.

Therefore, One of the most Tactical sound actions is to make choke points. Before you factor in Exowaves that make most Choke points defenses cry if you don't have every turret and it brother waiting for it.

To really make Chokepoints less important, would likely take two very radical changes. Make it so that turret cap was per planet, rather then global. (Make them spread out completely naturally, maybe even tie the size of the cap to the Command Station type and level.) And structures that are 'destroyed' like Factories, Fabs and whatnot can be rebuild after a delay. Make the delay long enough to basically hurt if you lose them, but not so long that they are out of the game. Say a countdown timer of 30 minutes if the AIs blow up a Factory before it can come back online. Exact delay pending testing. You want to make it enough to really hurt if you lose it, but not so much that you again want to do the solid wall of guns to stop it from happening.

Keep the AI liking to focus power with less doors in, and you get player tend to leave many door opened with turrets all around. AIs will send weak waves everywhere, rather then at one door with every gun that the humans own pointing at it waiting for them.

Offline orzelek

  • Hero Member Mark III
  • *****
  • Posts: 1,096
Re: Chokepoint Balance
« Reply #3 on: March 18, 2013, 08:48:16 pm »
I would really like the turrets per planet solution - I proposed that long time ago.
It provokes quite significant arguments against especially from people that really like to build huge choke points.

We could try to find some kind of compromise - like have per planet turret caps (dependent on command station or not) and also special galaxy wide limited "control" structure that would allow you to increase that caps to create a choke points. Depending on how that control structure would work and turret limits we could allow for certain amount of configurable in depth vs choke point defense.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Chokepoint Balance
« Reply #4 on: March 18, 2013, 08:52:37 pm »
Yes, I've wanted to do per-planet turret caps for a long time, but multiple long forum discussions on the topic have resulted in fairly sharp controversy.  Contrary to appearances, I do try to resist the urge to simply force my vision of what I think would be cool on the player-base ;)
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Hearteater

  • Core Member
  • *****
  • Posts: 2,334
Re: Chokepoint Balance
« Reply #5 on: March 18, 2013, 08:53:05 pm »
One thing I think would be really nice, especially on high-connection count maps, would be a Worm Hole Destabilizer.  It would be built next to a Worm Hole to make it unusable to anyone, but it costs Energy.  A lot of energy.  Say 240k.  But it isn't a straight 240k because that would really benefit low-count maps too much.  Instead it costs 240k divided by the number of worm holes in that system.  So totally isolating a system would cost 240k, but blocking 3 of 8 worm holes would cost only 90k.  Just add some evil AI effect if you cut yourself off from the AI completely (like all waves, Exos, etc. use the current Counter Attack GP spawning logic).  This way at least you could shape your map to something more defensible, but you probably couldn't get anything down to a single choke point if it wasn't already close.

Offline Eternaly_Lost

  • Sr. Member
  • ****
  • Posts: 336
Re: Chokepoint Balance
« Reply #6 on: March 18, 2013, 08:57:03 pm »
Yes, I've wanted to do per-planet turret caps for a long time, but multiple long forum discussions on the topic have resulted in fairly sharp controversy.  Contrary to appearances, I do try to resist the urge to simply force my vision of what I think would be cool on the player-base ;)

The biggest problem I have with it, is irreplacable structures you can't move that are very useful. They tend to making you want to focus your defenses between them and the AI. Leading to Chokepoints.

Move away from that somehow, and I have no objections to a per planet turret cap. Or maybe have those special structures have enough health to give you time to fly in support when a wave beaks though, or they grant a bonus to that planet's cap that should make it basically a mini chokepoint.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Chokepoint Balance
« Reply #7 on: March 18, 2013, 09:04:20 pm »
This is just off-the-top-of-my-head, not sure if it would be a good idea at all, but I'm considering:

a) Convert all very-important capturables (AdvFact, ASC, Fabs... probably leave ZPGs and Ions and such as-is) to use this approach I mentioned in the other thread:
"Have the unit be invincible, but it shuts down for an hour after a human command station dies on the same planet.  So it's not fully gone, but you lose use of it for a time long enough (pending tuning) that it hurts bad enough that you genuinely care about not losing the planet."

b) Make most turrets per-planet-cap.

c) Whenever you lose a system with such a capturable, the AI gets a bit of "blood in the water" temporary boost to its aggressiveness.  Like throwing an extra wave at you or adding 10% or whatever to the next 30 minutes' worth of waves, etc.


The idea being that losing such a capturable could start a snowball that could kill you relatively quickly, but that if you survive that you can get back on your feet without permanent loss (and thus with less risk of stalemate).
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Diazo

  • Master Member Mark II
  • *****
  • Posts: 1,717
  • I love/hate Diff 10
Re: Chokepoint Balance
« Reply #8 on: March 18, 2013, 09:12:51 pm »
As the person who's pretty much started this conversation (or at least being very vocal about it recently), I suppose I need to provide some background on why I've been harping on this.

(See my next post for actual gameplay suggestions.)

I generally play very difficult games, my 'standard' challenging game setup is Random All AI Types, diff 9/9, lattice map and then hit "New Map Seed" and then "Start Game" without looking at the map or choosing your starting ship.

I do this so I have a random map location and a random bonus ship. If I allow myself to pick starting map location and starting bonus ship it is in a diff 10/10 game.

However, I get away with this because I do not enable any of the minor factions that help me and so in return I do not enable anything that helps the AI, in other words I play straightforward games, it is me vs. the AI and nothing else.

However, even in that setup the best games I play end up being the ones I unlock warp jammers and play a low aip planet hopping strategy so I am effectively "chokepointing" the AI by only exposing a single planet to waves and with the low AIP I do not require significant defenses to defeat AI waves. (Relative to what I see posted on the forums anyway.)

Whenever I do not unlock warp jammers the game is significantly harder as I now have significantly more systems to defend.

On a lattice type map, having more 'safe' systems not adjacent to an enemy warp gate then there are 'hot' systems the AI can send waves at is the exception, not the rule.

And because I play low AIP on lattice, gate raiding is not an option. 6 or 7 games ago now, my starting homeworld had 14 adjacent systems, that would be 70 AIP in gate raiding alone.

However, what pushed me over the edge was the fact that I read a couple great Fallen Spire AARs so I decided to finally play the Fallen Spire campaing. I enabled the Fallen Spire at 4/10 and dropped the difficulty down to 8 for both AIs, figuring that as it was my first Fallen Spire playthough I would need the leeway.

I then proceeded to get flattened, over and over and over. I'm not talking a mistake where I miss something and have to save-scum, I'm talking having to save-scum the same event multiple times because I simply did not have enough units and had to get both a lucky roll on the composition of the exo-wave and lucky on how the combat went to survive.

As I understand the minor faction difficulty, Fallen Spire 4/10 is supposed to be default, that is tough-but-fair. However, on a lattice map that exo-wave is anything but fair because the fact that it can come at me from multiple directions means that I either split my defenses on the border worlds which is a losing prospect against an exo-wave that is expecting to meet a chokepoint, or I allow the exo-wave through to its target (my homeworld) and turn my homeworld into a 'chokepoint'. Both of these options are bad.

This experience took me from "ah, lattice maps are a little tougher but I can take it" to "lattice maps are a full difficulty, if not a difficulty and a half harder" and made me focus on what makes the map with fewer connections so much easier, the chokepoint.

Having said that, this is a strategy game and acheiving the chokepoint is a desirable thing as it simplifies your defense and allows concentration of force. The problem is that because AI War is procedurally generated, the game designed can't place chokepoints in strategic locations, but force the player to have to leave them to win the game. Rather, you get map types where natural chokepoints form when the game generates its layout.

I just took a quick look and I would say anywhere from half to two-thirds (depending on your personal definition of chokepoint) of the map types generally give you maps that have a lot of chokepoints. A 'lot of chokepoints' in this case being you can expand your empire in such a way as to never have more then 3 systems exposed to the AI without gate raiding.

Maybe part of the issue is the 'lattice' map type. I just did three test maps using the same map seed for both lattice and realistic map types.

At least one starting option on each of the three the lattice map types had 11 warp point connections (one of the three had a 14 connections option), on the realistic maps the maximum warp point connections was 4 across all three map seeds.

I think at the end of the day it boils down to the fact that I want lots of warp point connections so I have multiple routes to get places and deciding which systems to take is important and also give me the leeway to go around a Mk IV system with a whole bunch of specials if I want to.

I don't want to be playing a low connections map where my choices are limited.

I also feel that because the game is harder as the number of warp point connections goes up, a lot of the recent changes to how AI reinforcement, waves, special forces, strategic reserve, etc. have been balanced by feedback from players and the average player plays a map with a lot fewer connections then I do, so my map choice is giving me a harder experience for the same balance point.

That's why I was happy to see the topic about the Counter-Attack posts and warp gates sending waves to more then just the adjacent planet. Because I expose so many systems to the AI due to my map choice of lattice, I'm effectively already playing this way. Implementing a mechanic like that where you could not (as easily) setup a single chokepoint would then swing the game balance back towards what I prefer, that is a lot of connections between systems, which brings with it a lot of systems exposed to AI attack.

(See my next post for actual suggestion.)

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,570
Re: Chokepoint Balance
« Reply #9 on: March 18, 2013, 09:14:02 pm »
Eternally_Lost brought up a good point:
Many of the capturables' durability has not kept up with the rest of the game, making them very fragile, thus reducing their usefullness too much when the AI has lots of ways to attack you.
Fixing that should solve many (though most certainly nowhere near all) of the complaints.

Personally, I am opposed to the per planet turret caps for all turrets. However, some turrets/defensive structures having that I could see. I would be in favor of adding more turrets/defensive structures that use the per planet cap mechanic.

The "invincible but forced powered down time if human loses planet" idea is also very nifty. The powered down time would have to be quite long to make any real impact though.
Not so sure with the "temporary boost" to aggressiveness. Though, I guess I could see it as a temporary boost in AIP or something.

I still think that exos need a good ol' whack with the nerf bat, or at least some way for it to get less intense when attacking non-HW targets, especially remote ones. Or some way to detect "chokepoints" and get stronger but longer to charge if it sees one (like waves) so the base exo strength can go down (making remote captures more viable) but keep the "chokepoint breaking utility" of them when the AI sees one and no such remote outposts.

Also, I'm surprised to see the "control structures" idea come back from a long time ago when we discussed this.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Chokepoint Balance
« Reply #10 on: March 18, 2013, 09:24:04 pm »
Eternally_Lost brought up a good point:
Many of the capturables' durability has not kept up with the rest of the game, making them very fragile, thus reducing their usefullness too much when the AI has lots of ways to attack you.
Fixing that should solve many (though most certainly nowhere near all) of the complaints.
Is there any realistic level of durability that would not die very quickly to sustained fire from a serious attack?  You can pile lots of FFs on them already if it's simply a problem with HP.

Once the AI is there, and once the FFs are down, the thing is going to die.  I don't see how different HP numbers could change that unless I went totally nuts with them :)
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Hearteater

  • Core Member
  • *****
  • Posts: 2,334
Re: Chokepoint Balance
« Reply #11 on: March 18, 2013, 09:29:40 pm »
Personally, if the Adv. Factory (or whatever) isn't going to die, then the AI can't really threaten it no matter where it is located.  Warp Jammer CS and no waves to worry about, drop Mini-Forts + Fort and your threat is so high no AIs will randomly wander in.  Until an Exo or insane threat fleet builds up, you're fine.  Then just walk back in and rebuild.  No problem.  It isn't like you lost anything except 5-10 minutes of your time.  If my AF is locked out for an hour, I'll just wait an hour to capture the system back.  If you account for the time until I've used up all my Mark IVs and could really use the AF again, its probably closer to 10 minutes of inconvenience.

I'd say have the AI blow them up, they leave remains you can rebuild them, but doing so costs AIP (once they finish being rebuilt).  At least that's permanent consequences.  No worse than a ZPG really.


Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,570
Re: Chokepoint Balance
« Reply #12 on: March 18, 2013, 09:30:46 pm »
Eternally_Lost brought up a good point:
Many of the capturables' durability has not kept up with the rest of the game, making them very fragile, thus reducing their usefullness too much when the AI has lots of ways to attack you.
Fixing that should solve many (though most certainly nowhere near all) of the complaints.
Is there any realistic level of durability that would not die very quickly to sustained fire from a serious attack?  You can pile lots of FFs on them already if it's simply a problem with HP.

Once the AI is there, and once the FFs are down, the thing is going to die.  I don't see how different HP numbers could change that unless I went totally nuts with them :)

While I would argue this is part of the "problem" (offensive power both per ship, relative and the sheer magnitude of the numbers, and how easy it is to get a large number of ships) have made it such that sane durability is almost impossible. When it comes to "the enemy is trying to take you out with full force" type situations, you either get absurd durability (like motherships) or they will go down quickly. The fact that there current balance allows almost no middle is itself a problem in my eyes.


However, more to the point (and less about my personal opinion of wide sweeping aspects to the game :P), I guess we are asking for enough durability such that a couple of bombers (like no more than a dozen) coming into a world with a capturable with little defenses doesn't spell death for the capturable before you have time to react.

Offline Diazo

  • Master Member Mark II
  • *****
  • Posts: 1,717
  • I love/hate Diff 10
Re: Chokepoint Balance
« Reply #13 on: March 18, 2013, 09:34:10 pm »
Okay, now for actual suggestions:

First, I prefer the galaxy-wide caps for units. It forces you think about how you place your limited units. This is also the biggest reason chokepoints are so overpowering because they allow you to focus where you place your units to such an extreme degree.

If you had a per-planet turret cap, can you say build time? On the higher difficulty levels this would turn every system into a fortresses as on the higher difficulty, you are eventually going to lose a system, so you fortify the system behind it to the same degree.

So yes, I'm arguing against chokepoints while saying chockpoints are a good thing.

So, to avoid the chokepoint syndrome, we need to give the AI ways to get around it. The mechanic in the counter-attack thread of allowing some attacks to warp in 2 or more hops out is a good starting place for this.

The AI needs more then just that though to actually give it more options to get around chokepoints.

What about everytime a wave happens, a chance for it to be a "deep jump" wave. Significantly weaker but it can spawn an extra hop (or more) deeper into player territory.

I'd suggest something to do with supply, but that is beachheads and those are quite hated.

Or a "long-range warp" where a CPA instead of releasing as normal retreats to the AI homeworld where they enter a structure that warps them to a player's system. The key here being that they would have to ignore the firepower check for warping and just go as soon as they got there so they would stream in over a period of minutes so you would actually build defenses rather then just stop offensive operations and use your fleet to defend it.

Or a ship hunter. Really low damage against turret type hulls, but damage bonuses against almost everything else in the game that warps in randomly?

I'm streaching things now because I'm trying to balance "chokepoints are desirable" with "one (or two) chokepoint defenses should fail" and coming up short on options.

But as long as maps vary as widely as they do, from snake with nothing but chokepoints, to lattice which have innumerable connections, I'm not sure this has an answer.

D.

Offline Faulty Logic

  • Hero Member Mark III
  • *****
  • Posts: 1,194
  • Bane of the AI
Re: Chokepoint Balance
« Reply #14 on: March 18, 2013, 09:54:06 pm »
I like chokepoints as they stand for most cases. They are luxuries in most games, an impossible dream in others, and occasionally an absolute necessity.

The only real issue is exos. Without a choke, they are virtually impossible to deal with. I don't know how to deal with this, but some kind of exo-strength scaling with number of ingress points would be a good start. Until then, if you play an open map with exos, you are electing to play a brutal game.
If warheads can't solve it, use more warheads.