Author Topic: Upper Level Bombers a Requirement?  (Read 10053 times)

Offline darke

  • Hero Member
  • *****
  • Posts: 534
Re: Upper Level Bombers a Requirement?
« Reply #15 on: July 14, 2009, 08:31:52 pm »
It's a shame that more people don't use EMP missiles, which temporarily disable the force fields (thus allowing players to clear out the stuff underneath them).

I've tried this, since I'm used to trying to hammer higher level planets then I should be. :) IIRC the issue was that it's such a short downtime (30 seconds?), and that it seems to affect your ships in the system, as well as the enemy's, that it's not worth it unless the forcefield is adjacent to your entrance wormhole (or wormholes).

The electric missiles work well to clear out ships under the shield, however their range isn't large enough to clear out the entire forcefield, it usually only gets half to two-thirds due to general damage on them (I mentioned this in another thread a while back, IIRC, you didn't want to increase the range/decrease the damage since you wanted them to be anti-starship in specific, but you tossed more health at them so they could actually survive a few seconds longer :) ).

From memory the main problem with using this tactic earlier in the game or against softer targets was the cost of the missile solo was huge (100k? Was I dreaming or did I miss this change in the patch notes? :) ), but now it seems to be only 25k so that's somewhat better.

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #16 on: July 14, 2009, 09:13:21 pm »
I have never even SEEN an EMP missile.

In fact, I haven't seen the majority of things in the game, as I seem to get my small ships up to Mk III first and then research star ships, so I almost never need anything else before I win.

Well, that's your prerogative, I guess.  I tend to experiment around more in RTS games, even if I ultimately settle into a pretty narrow band of units I use.  The point of the rotating bonus ships is, of course, to get players to experiment around more and actually play with a larger selection of the ships, but that doesn't work for things like Starships, defensive ships, or missiles.  That's something I probably need to think on more -- how to get players to experiment more. 

I suppose one way to get them to experiment with missiles would be to seed missile silos around the maps (like starship constructors, but less frequent), so that players have a chance to experiment with them.  And for the case of nukes and EMPs, I think they are too overpriced at the moment to be attractive for players to experiment with.  Nukes already come with such a high AI Progress increase (50, I think) that they won't be overused even if I make them super cheap.  I'll have to look at that and see about rebalancing them some.

I like the Starship bonus idea. I usually try to build all five of them as soon as possible.

Awesome. :)
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 x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #17 on: July 14, 2009, 09:14:56 pm »
What ships do you need to build to get access to these EMP missiles?

Missile Silos -- those are under the CONST tab of your command station.  They are quite pricey, but really helpful in specific situations.  I'm going to probably reduce their costs in the next prerelease, though.  If they become too powerful I can always nerf them (this isn't pvp, so who is it really hurting), but right now they are so expensive that only a very few players are using them at all.
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 x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #18 on: July 14, 2009, 09:24:19 pm »
This is what I am seeing, too.  You can get most everything done with a strong offensive military - which is a downside to a degree because some of the other technologies are kindof nifty.

Yeah... that's a challenge.  I'll have to think on what to do with that.  Maybe the AI needs to be stronger on the offensive so that you need more of these technologies in order to defend yourself.  Of course, there again, that law of unintended consequences will likely rear its head.  This one needs some thought/discussion.

At the very least, I think I've succeeded in making MORE of the units used in average play compared to a lot of RTS games, but this problem of how to get players to use all the units over the life of their experience with the game is pretty endemic to this genre, I think.  I'm pretty experimental by nature, but in AoEIII and even SupCom I tended to try stuff once, and then just stick with what worked to the exclusion of all else.  In AoEIII I got very good at building just Musketeers as French (and very occasionally their strong melee horseman), and beating the pants off the AI as a boomer with two or three human allies in a 3x3 or 4x4 game.  In SupCom, I did a lot more experimentation and used a lot more of the ships, but I just played as Aeon and by the end was using ONLY tech 2 and tech 3 ships, just skipping tech 1 all together.  Out of the other RTS games I've played aside from my own, I think the original Empire Earth and then SupCom did the best job of getting me to experiment around.  I think I've mostly duplicated that with the offensive military stuff, but not so much with the defensive tech.

Maybe... maybe there should be some sort of "balanced unlocking" rule.  Such as, you can only unlock so many techs in one category before you have to unlock some techs in another category to progress onward.  The tech "tree" in Rise of Legends was really limited (4 techs each in 4 categories), but it was cool how it made you only able to get 1 or 2 ahead in any category before having to research something from the others.  Perhaps something like this would work here...  this probably bears more thought.
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 CautiousChaos

  • Full Member
  • ***
  • Posts: 100
Re: Upper Level Bombers a Requirement?
« Reply #19 on: July 14, 2009, 11:10:45 pm »
Well however this pans out, I'm enjoying the game immensely.  I appreciate your willingness to give this some added thought.  Perhaps one of your latest suggestions, one about making the AI a bit more aggressive to give you more incentive to use defensive units, could be the best with the least impact to balancing.  I'm playing Moderate at Level 5 and have the highest score over the AI's.  If I am attacking a planet I will occasionally see them sending everything within the planet out at me.  Pretty exciting stuff, but not terribly tough to beat.  Same with the waves.  Of course, my AI progress is pretty low right now because I am just starting out, so that may be a factor.  But a bump up in Offense on the side of the AI I would welcome. 

-cc

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #20 on: July 14, 2009, 11:34:25 pm »
Well however this pans out, I'm enjoying the game immensely.  I appreciate your willingness to give this some added thought.  Perhaps one of your latest suggestions, one about making the AI a bit more aggressive to give you more incentive to use defensive units, could be the best with the least impact to balancing.  I'm playing Moderate at Level 5 and have the highest score over the AI's.  If I am attacking a planet I will occasionally see them sending everything within the planet out at me.  Pretty exciting stuff, but not terribly tough to beat.  Same with the waves.  Of course, my AI progress is pretty low right now because I am just starting out, so that may be a factor.  But a bump up in Offense on the side of the AI I would welcome. 

Thanks!  Glad you are enjoying the game so much, at any rate.  As for the AI aggressiveness, just turning up the AI difficulty from 5 will accomplish that to a good degree, and there is also stuff like 2x waves, etc.  Those are just short-term things you can use if you want to up the aggressiveness before whatever longer-term fix is put in place.  At the moment I want to be careful about what sort of changes I make in that regard, to avoid making games unplayable on the higher difficulties, etc.  Right now I'm seeing a lot of people losing/stalemating games on difficulty 7 depending on which AI type they are facing, so I think the difficulty scale is where it needs to be in general. 

The main thing is that I would prefer to have the majority of games end in either win or loss, rather than stalemate (the player just giving up).  Right now, when players play against tougher AIs on harder difficulties, stalemate happens vastly more than outright losses.  I think this is basically because the AI isn't harming the players enough, but the AI also becomes so entrenched that the players have limited chances of victory in a reasonable amount of time.

Perhaps just some simple change to convert some of that defensive energy into offensive power would be a good idea.  On difficulty 5 and up, perhaps if I have the the AI players skim ships off of large-defense planets and put those into attacks.  For instance, maybe once any given planet hits 3000 ships or so, the AI takes 1000 of those ships and sends them against the human players.  That would add an interesting dynamic where players have to keep tabs on how large nearby planets are getting, to make sure they aren't at risk of getting steamrolled in the future.  By the same token, that would keep the enemy planets from ever getting more than 2k - 3k ships on them at once, which would make the offensive efforts easier for the players if they survive the additions to the AI offense.  At the very least this could be an interesting AI modifier, but it might be a sensible basic game mechanic update, too.  I'll be interested to hear what everyone thinks about this -- for now it is going on the "maybe" list.
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 darke

  • Hero Member
  • *****
  • Posts: 534
Re: Upper Level Bombers a Requirement?
« Reply #21 on: July 14, 2009, 11:37:06 pm »
This is what I am seeing, too.  You can get most everything done with a strong offensive military - which is a downside to a degree because some of the other technologies are kindof nifty.

Yeah... that's a challenge.  I'll have to think on what to do with that.  Maybe the AI needs to be stronger on the offensive so that you need more of these technologies in order to defend yourself.  Of course, there again, that law of unintended consequences will likely rear its head.  This one needs some thought/discussion.

The main issue is that unusual for the genre, to gain knowledge you need offense. And since knowledge is the most important resource, you are going to emphasise what's necessary to gain more knowledge: offensive ships.

Plus one of the more previously irritating problems was how wimpy the defensive turrets were. For wormholes with few attacks they were ok since you'd repair them quickly enough after each assault, but for any real assault (8x special forces base on an adjacent world...) they'd just die, and die, and die, and... even forcefields aren't all that helpful. I found I was better off tossing a couple of space docks on the world, tossing forcefields down around them, and cranking out ships, which is... offensive technology benefited. :) (Granted they have more health now, but I haven't had a chance to test that yet.)

In short, for minimally-attacked worlds, there's little need to upgrade from the basic turrets, unless you run out; for heavily attacked worlds, there's no reason to build defensive techniques since your offensive ships work much better.

In that above situation, a fortress or two would be good... except they're hideously expensive in knowledge. Four full worlds of knowledge to get the ability to build expensive buildings that can't be repaired. For that I can get two upgrades to Tech III ships which throws in Tech IV as a bonus if you've got an Advanced Factory; which reminds me I think they really actually should be knowledge upgrades, though probably cheaper then Tech III since there is your original "cost" of getting an advanced factory and defending it is high, maybe 1000 or something each so you're at least forced to think, and/or attack a couple of worlds.

In short:

1) Defense should be cheaper then offense (which they are, slightly, but definitely not in research).
2) Defense should be more robust then offense (which they aren't; SRT I health 23000, shields 600; fighters/bombers/cruisers I health 29000/20000/24000, shields 100/1500/150. Really I don't mind SRTs costing more then a fighter, or missile turrets costing almost as much as a bomber since they do more damage, but they really need at least 3x as much health/shields as they do now to be viable in "real" defense).
3) Defense should be less micromangement then defense (which they are in low contested systems, but a pack of offensive ships is also low-micro in a low contested system...)

Offense of course should be increased; this ties nicely into the problem of seriously overpopulated worlds.

Each world should have a threshold, (for the moment I'll pull numbers out of thin air :) ), we'll assume that 50+/-20% percent of total number of ships the player has for a Tech I world (of course higher Tech worlds, would have higher thresholds, and there should always be a minimal "max" value). The threshold is checked once every $timedelta, (5+/-2 minutes for variance so every world doesn't proc at the same time), then grabs a small force, say up to 6 ships from every wormhole, or command point and sends them towards the player's wormhole.

You'll get a lighter attack, then say a timed wave, since all the ships will arrive in small groups over time since every point will be a different distance from the nearest player owned wormhole, and but you'll get a much more sustained assault so it'll be slower to repair stuff. This will also help with the ridiculous number of ships that regularly build up on nearby worlds, but there should also be a substantial force available there.

As to the only-offensive-knowledge problem there is another semi-solution, involving creating a new kind of research called something like "defensive optimisation". Basically as you kill ships with your defensive turrets/ships on worlds you have an active command center on (to avoid gaining this sort of stuff from parasited turrets and the like, you really don't want to give them any more bonuses :) ), you get 0.01 point of "defenseop" per tech level (or whatever, needs balancing depending upon how many ships people kill in a usual game and the like. The "defenseop" then discounts on all defensive technologies, it subtracts from there first at one point per knowledge point, before draining knowledge (this will also help with the very "round" values of knowledge you usually have).

This will also benefit the player being risky and camping beside, say, a Tech IV place, or somewhere with a ton of special forces bases, or just not taking out gates so they get more warps, and thus more assaults.

At the very least, I think I've succeeded in making MORE of the units used in average play compared to a lot of RTS games, but this problem of how to get players to use all the units over the life of their experience with the game is pretty endemic to this genre, I think.  I'm pretty experimental by nature, but in AoEIII and even SupCom I tended to try stuff once, and then just stick with what worked to the exclusion of all else.  In AoEIII I got very good at building just Musketeers as French (and very occasionally their strong melee horseman), and beating the pants off the AI as a boomer with two or three human allies in a 3x3 or 4x4 game.  In SupCom, I did a lot more experimentation and used a lot more of the ships, but I just played as Aeon and by the end was using ONLY tech 2 and tech 3 ships, just skipping tech 1 all together.  Out of the other RTS games I've played aside from my own, I think the original Empire Earth and then SupCom did the best job of getting me to experiment around.  I think I've mostly duplicated that with the offensive military stuff, but not so much with the defensive tech.

Maybe... maybe there should be some sort of "balanced unlocking" rule.  Such as, you can only unlock so many techs in one category before you have to unlock some techs in another category to progress onward.  The tech "tree" in Rise of Legends was really limited (4 techs each in 4 categories), but it was cool how it made you only able to get 1 or 2 ahead in any category before having to research something from the others.  Perhaps something like this would work here...  this probably bears more thought.

Already got way too much verbiage, I might get to thinking about this when my fingers stop hurting from all the typing. :)

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #22 on: July 14, 2009, 11:53:17 pm »
The main issue is that unusual for the genre, to gain knowledge you need offense. And since knowledge is the most important resource, you are going to emphasise what's necessary to gain more knowledge: offensive ships.

This is a very interesting point.  I had not really thought about this before.

Plus one of the more previously irritating problems was how wimpy the defensive turrets were. For wormholes with few attacks they were ok since you'd repair them quickly enough after each assault, but for any real assault (8x special forces base on an adjacent world...) they'd just die, and die, and die, and... even forcefields aren't all that helpful. I found I was better off tossing a couple of space docks on the world, tossing forcefields down around them, and cranking out ships, which is... offensive technology benefited. :) (Granted they have more health now, but I haven't had a chance to test that yet.)

You know, it seems like maybe turrets should have regen, plus also a lot more health.  Combined with more ongoing attacks, as you also suggested, that could be a really interesting change.

1) Defense should be cheaper then offense (which they are, slightly, but definitely not in research).
2) Defense should be more robust then offense (which they aren't; SRT I health 23000, shields 600; fighters/bombers/cruisers I health 29000/20000/24000, shields 100/1500/150. Really I don't mind SRTs costing more then a fighter, or missile turrets costing almost as much as a bomber since they do more damage, but they really need at least 3x as much health/shields as they do now to be viable in "real" defense).
3) Defense should be less micromangement then defense (which they are in low contested systems, but a pack of offensive ships is also low-micro in a low contested system...)

The only problem is that if defense is too cheap and easy, then players are in no danger and the excitement factor goes down.  Everyone becomes an auto-turtle with the ability to go out and attack too easily (since the offensive ships are now no longer needed at all on the home front, or hardly so).  That is a pretty huge balance shift.

Offense of course should be increased; this ties nicely into the problem of seriously overpopulated worlds.

Each world should have a threshold, (for the moment I'll pull numbers out of thin air :) ), we'll assume that 50+/-20% percent of total number of ships the player has for a Tech I world (of course higher Tech worlds, would have higher thresholds, and there should always be a minimal "max" value). The threshold is checked once every $timedelta, (5+/-2 minutes for variance so every world doesn't proc at the same time), then grabs a small force, say up to 6 ships from every wormhole, or command point and sends them towards the player's wormhole.

Yes, something like this could increase the "trickle attacks" like the special forces.  I had a suggestion in the post right above yours (which probably came in while you were writing) that basically suggested something similar except in bigger batches. I'm not sure which would really be better, right now the AI tends not to do super-huge waves against the player planets, so I would think that having 1000+ ships suddenly arrive would be more interesting.  I would make some sort of warning for when that even occurs, maybe even a countdown timer for before they get released like with the waves, to give the humans time to prepare.  I think that could add an interesting angle to the defensive side.

You'll get a lighter attack, then say a timed wave, since all the ships will arrive in small groups over time since every point will be a different distance from the nearest player owned wormhole, and but you'll get a much more sustained assault so it'll be slower to repair stuff. This will also help with the ridiculous number of ships that regularly build up on nearby worlds, but there should also be a substantial force available there.

Yeah.  That will also help with performance on some planets for players with below-spec machines.  Having ships get attritioned over time (or even in moderate-sized chunks like 1000 ships at a time) is going to use less CPU then going to a planet with 5k ships on it will.  So that's another plus.  If the AI is attacking the players with more gusto, then the players will by turn have fewer ships to mount the average offensive, so the shift in the location of ships won't really have any effect on the overall balance of the planets, I wouldn't think.  It's just the AI setting the place and time of some of the battles, rather than the player always having that initiative.

As to the only-offensive-knowledge problem there is another semi-solution, involving creating a new kind of research called something like "defensive optimisation". Basically as you kill ships with your defensive turrets/ships on worlds you have an active command center on (to avoid gaining this sort of stuff from parasited turrets and the like, you really don't want to give them any more bonuses :) ), you get 0.01 point of "defenseop" per tech level (or whatever, needs balancing depending upon how many ships people kill in a usual game and the like. The "defenseop" then discounts on all defensive technologies, it subtracts from there first at one point per knowledge point, before draining knowledge (this will also help with the very "round" values of knowledge you usually have).

This will also benefit the player being risky and camping beside, say, a Tech IV place, or somewhere with a ton of special forces bases, or just not taking out gates so they get more warps, and thus more assaults.

Yeah, I like aspects that, but it seems a bit opaque and meta-gaming-y to me.  In many respects it seems like it would be simpler to just have "defensive knowledge" points that you can gain through some sort of defensive actions (maybe 1 point per kill on your planets), and which can only be used on defensive technologies.  These could be used in concert with regular knowledge, basically as a replacement.  So if the Fortress costs 8,000 knowledge, and you have 6000 defensive points stored up, you unlocking the fortress will spend all 6000 of your defensive points as well as 2000 of your knowledge.  That would be pretty cool, because it would add incentive to defense without unbalancing it too vastly much.

Of course, then this will lead to potential more exploits with trying to lead the enemy ships to your planets, etc.  But I suppose those can be tackled with further balancing after that initial addition of the defensive points or whatever. :/
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 darke

  • Hero Member
  • *****
  • Posts: 534
Re: Upper Level Bombers a Requirement?
« Reply #23 on: July 15, 2009, 12:39:17 am »
The main issue is that unusual for the genre, to gain knowledge you need offense. And since knowledge is the most important resource, you are going to emphasise what's necessary to gain more knowledge: offensive ships.

This is a very interesting point.  I had not really thought about this before.

Random thought: Capture data centers rather then kill them allows them to slowly generate "defensive knowledge" over time. Either that or a new form of capture-able item would work, which would probably be a better idea since most data centers are near the end of the game, not the beginning, so near the end-game it's going to be a non-decision to take them out. They'd have the same effect as the advanced factories then, they're immobile so you have to setup defenses in the world; and probably the computer won't like that so much, so they'll be a small-but-non-zero penalty for this (like the homeworder's human planets, but not quite as harsh :) ).

Plus one of the more previously irritating problems was how wimpy the defensive turrets were. For wormholes with few attacks they were ok since you'd repair them quickly enough after each assault, but for any real assault (8x special forces base on an adjacent world...) they'd just die, and die, and die, and... even forcefields aren't all that helpful. I found I was better off tossing a couple of space docks on the world, tossing forcefields down around them, and cranking out ships, which is... offensive technology benefited. :) (Granted they have more health now, but I haven't had a chance to test that yet.)

You know, it seems like maybe turrets should have regen, plus also a lot more health.  Combined with more ongoing attacks, as you also suggested, that could be a really interesting change.

I wouldn't really want to give them regen, really this sort of thing is what engineers are for. Plus this is one way to force more of the Tech I and II engineers off the battlefield, where the Tech III ones are supposed to shine anyway. :)

1) Defense should be cheaper then offense (which they are, slightly, but definitely not in research).
2) Defense should be more robust then offense (which they aren't; SRT I health 23000, shields 600; fighters/bombers/cruisers I health 29000/20000/24000, shields 100/1500/150. Really I don't mind SRTs costing more then a fighter, or missile turrets costing almost as much as a bomber since they do more damage, but they really need at least 3x as much health/shields as they do now to be viable in "real" defense).
3) Defense should be less micromangement then defense (which they are in low contested systems, but a pack of offensive ships is also low-micro in a low contested system...)

The only problem is that if defense is too cheap and easy, then players are in no danger and the excitement factor goes down.  Everyone becomes an auto-turtle with the ability to go out and attack too easily (since the offensive ships are now no longer needed at all on the home front, or hardly so).  That is a pretty huge balance shift.

Right, which is why you need more enemies pounding at the door to compensate. :)

Offense of course should be increased; this ties nicely into the problem of seriously overpopulated worlds.

Each world should have a threshold, (for the moment I'll pull numbers out of thin air :) ), we'll assume that 50+/-20% percent of total number of ships the player has for a Tech I world (of course higher Tech worlds, would have higher thresholds, and there should always be a minimal "max" value). The threshold is checked once every $timedelta, (5+/-2 minutes for variance so every world doesn't proc at the same time), then grabs a small force, say up to 6 ships from every wormhole, or command point and sends them towards the player's wormhole.

Yes, something like this could increase the "trickle attacks" like the special forces.  I had a suggestion in the post right above yours (which probably came in while you were writing) that basically suggested something similar except in bigger batches. I'm not sure which would really be better, right now the AI tends not to do super-huge waves against the player planets, so I would think that having 1000+ ships suddenly arrive would be more interesting.  I would make some sort of warning for when that even occurs, maybe even a countdown timer for before they get released like with the waves, to give the humans time to prepare.  I think that could add an interesting angle to the defensive side.

Given this used to be my bread and butter type of defense due to taking out the command center, my reaction to that has been that they haven't actually had to face a huge number of ships before. Nor have they taken a huge number of ships on an assault against a lower-class world. :)

With the current way the AI works batching it's assault forces up before streaming them through after the death of the command post, more then about 300 ships of any appropriate tech level is about the LD50 for the game (appropriate tech level == the planet is Tech I if you've got Tech I ships, Tech II planet if you have Tech II ships, Tech III planet is you've got Tech III ships, Tech IV planet if you've got your entire attack force sitting at the opposite side of the wormhole).

The main issue is that the size of the force and the tech level of it is a significant force multiplier. Which is why parasites and tossing a few engineers into the pack of your ships are so valuable in an assault or a defense, since they keep the number of troops up, so you minimise the chance of hitting the tipping point where suddenly your troops go from there, to not-there in a few seconds. :)

Granted I've never setup a serious set of static defenses for most, but the last game where I had serious defenses on an 8x special forces AI world, it's 150-ish ships every five minutes or so could really not be defended against in the long term at all despite using every mine I had, mine-layers, all my force fields to protect my turrets, and a couple of hundred ships.

You'll get a lighter attack, then say a timed wave, since all the ships will arrive in small groups over time since every point will be a different distance from the nearest player owned wormhole, and but you'll get a much more sustained assault so it'll be slower to repair stuff. This will also help with the ridiculous number of ships that regularly build up on nearby worlds, but there should also be a substantial force available there.

Yeah.  That will also help with performance on some planets for players with below-spec machines.  Having ships get attritioned over time (or even in moderate-sized chunks like 1000 ships at a time) is going to use less CPU then going to a planet with 5k ships on it will.  So that's another plus.  If the AI is attacking the players with more gusto, then the players will by turn have fewer ships to mount the average offensive, so the shift in the location of ships won't really have any effect on the overall balance of the planets, I wouldn't think.  It's just the AI setting the place and time of some of the battles, rather than the player always having that initiative.

Yup.

As to the only-offensive-knowledge problem there is another semi-solution, involving creating a new kind of research called something like "defensive optimisation". Basically as you kill ships with your defensive turrets/ships on worlds you have an active command center on (to avoid gaining this sort of stuff from parasited turrets and the like, you really don't want to give them any more bonuses :) ), you get 0.01 point of "defenseop" per tech level (or whatever, needs balancing depending upon how many ships people kill in a usual game and the like. The "defenseop" then discounts on all defensive technologies, it subtracts from there first at one point per knowledge point, before draining knowledge (this will also help with the very "round" values of knowledge you usually have).

This will also benefit the player being risky and camping beside, say, a Tech IV place, or somewhere with a ton of special forces bases, or just not taking out gates so they get more warps, and thus more assaults.

Yeah, I like aspects that, but it seems a bit opaque and meta-gaming-y to me.  In many respects it seems like it would be simpler to just have "defensive knowledge" points that you can gain through some sort of defensive actions (maybe 1 point per kill on your planets), and which can only be used on defensive technologies.  These could be used in concert with regular knowledge, basically as a replacement.  So if the Fortress costs 8,000 knowledge, and you have 6000 defensive points stored up, you unlocking the fortress will spend all 6000 of your defensive points as well as 2000 of your knowledge.  That would be pretty cool, because it would add incentive to defense without unbalancing it too vastly much.

Of course, then this will lead to potential more exploits with trying to lead the enemy ships to your planets, etc.  But I suppose those can be tackled with further balancing after that initial addition of the defensive points or whatever. :/

The whole research system is pretty meta-gamey in comparison to everything else (you acquire "knowlege points", then pick a research that you then instantly know, rather then picking a research item to study, then the research points go towards that research item, which is less-meta-game, :) also ignoring the whole "knowledge raid" aspect that seems rather quirky too :) ), so I don't think having more meta-ness is much of a problem.

And you pretty much restated what I said, just much clearer. :) I'd probably have something like 10 defense knowledge equal to one regular knowledge in that case, but that would be in conjunction with lowering the knowledge costs of the more expensive defensive technologies at the same time, but your option is probably better off to start with anyway since it's simpler and more obvious. :)

One thing I forgot to mention in the last set of verbiage is that we now no-longer have "System Defense Boats", ships that can't leave the system but are good for defense. Teleporting things used to be that spot (and especially teleporting engineers which were perfect for this support role), but they were a bit on the hard to use side. If there was a ship type that was mobile, but could only move between non-contested worlds (that is friendly worlds with command centers on them), it would be the "can't attack an enemy world easy, but is useful for defense" that the original teleporting ships were good for, but without their inconveneince of having to build them on the same planet.


Offline Admiral

  • Hero Member
  • *****
  • Posts: 547
Re: Upper Level Bombers a Requirement?
« Reply #24 on: July 15, 2009, 01:26:38 pm »
You know, it seems like maybe turrets should have regen, plus also a lot more health.  Combined with more ongoing attacks, as you also suggested, that could be a really interesting change.

I find that attack turrets, in general, are nearly useless for defense (in my style of play/opinion, of course). Then again, I rarely get more than one or two turrets researched, and the enemy rarely uses Mk I enemies against me. Mostly, I put them there just because I can, but usually I load up on tractor beams, mines and snipers (which I can put far away) and ignore the rest. They just don't hurt the enemy (although I don't know about the higher level ones) enough to rely upon them for defense.

I like that they distract the offense into killing them instead of my ships, though. :)

Cheers!

Offline Admiral

  • Hero Member
  • *****
  • Posts: 547
Re: Upper Level Bombers a Requirement?
« Reply #25 on: July 15, 2009, 01:30:12 pm »
I would make some sort of warning for when that even occurs, maybe even a countdown timer for before they get released like with the waves, to give the humans time to prepare.  I think that could add an interesting angle to the defensive side.

What's with all the warnings? I mean, why isn't the enemy allowed to do a silent, strong attack? If you want warning, shouldn't you post scouts in every system and then watch them moving?

I think the AI needs more unpredictability (and variable sizes which can occasionally have a chance of some minor success) in its offense, not more warning and advanced notice. :)

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #26 on: July 15, 2009, 02:15:39 pm »
Random thought: Capture data centers rather then kill them allows them to slowly generate "defensive knowledge" over time. Either that or a new form of capture-able item would work, which would probably be a better idea since most data centers are near the end of the game, not the beginning, so near the end-game it's going to be a non-decision to take them out. They'd have the same effect as the advanced factories then, they're immobile so you have to setup defenses in the world; and probably the computer won't like that so much, so they'll be a small-but-non-zero penalty for this (like the homeworder's human planets, but not quite as harsh :) ).

This could be interesting, but I agree that it should be something other than data centers that you capture and use this way -- some sort of new ship.

I wouldn't really want to give them regen, really this sort of thing is what engineers are for. Plus this is one way to force more of the Tech I and II engineers off the battlefield, where the Tech III ones are supposed to shine anyway. :)

Well, that's true -- but it presupposes that everyone wants (and has the knowledge to) unlock higher level engineers.  With increased offensive attacks from the AI, this would pretty much be required.  And you'd always have this one weak point in your armor -- the engineers -- that if the AI happens to pick them off, then your defenses will crumble under the ongoing waves to your planets.  That requires a degree of maintenance that people tend to complain about. 

The idea of given the turrets regen is to make them MORE effective than regular ships at wormhole defense (thus giving a better reason to use them), and also to make it so that the higher-level engineers are not required.  I can see your point, but I think that this is kind of a fitting change.  Basically, most of the defensive stuff would then be non-repairable but would have regen.  Turrets would have a pretty quick regen, compared to say force fields, to make it so that they have to be really overmastered before they die.

Right, which is why you need more enemies pounding at the door to compensate. :)

Oh, absolutely.  But then that's why you need turrets with better health and regen. :)

With the current way the AI works batching it's assault forces up before streaming them through after the death of the command post, more then about 300 ships of any appropriate tech level is about the LD50 for the game (appropriate tech level == the planet is Tech I if you've got Tech I ships, Tech II planet if you have Tech II ships, Tech III planet is you've got Tech III ships, Tech IV planet if you've got your entire attack force sitting at the opposite side of the wormhole).

Point well taken.  This would be the benefit of beefing up turrets for this sort of defense, though.  Then basically the only way to survive such a large wave (while also having any meaningful offensive force) would be to have excellent turrets and other defenses in place.

Granted I've never setup a serious set of static defenses for most, but the last game where I had serious defenses on an 8x special forces AI world, it's 150-ish ships every five minutes or so could really not be defended against in the long term at all despite using every mine I had, mine-layers, all my force fields to protect my turrets, and a couple of hundred ships.

I've used mines and turrets to exclusively defend wormholes before (and do so every so often still).  In my alpha group, we create a little pattern that we call the "ring of death" and it works super well for everything that isn't tractor-immune or able to pass over mines (so, raider AI types are basically immune to the rings of death).

The whole research system is pretty meta-gamey in comparison to everything else

Okay, point taken. :)

And you pretty much restated what I said, just much clearer. :) I'd probably have something like 10 defense knowledge equal to one regular knowledge in that case, but that would be in conjunction with lowering the knowledge costs of the more expensive defensive technologies at the same time, but your option is probably better off to start with anyway since it's simpler and more obvious. :)

Cool.  That seems good to me.  I wasn't sure that I was restating what you were saying or not, since I wasn't 100% sure I understood the subtleties of yours, so good to know we are on the same page.

One thing I forgot to mention in the last set of verbiage is that we now no-longer have "System Defense Boats", ships that can't leave the system but are good for defense. Teleporting things used to be that spot (and especially teleporting engineers which were perfect for this support role), but they were a bit on the hard to use side. If there was a ship type that was mobile, but could only move between non-contested worlds (that is friendly worlds with command centers on them), it would be the "can't attack an enemy world easy, but is useful for defense" that the original teleporting ships were good for, but without their inconveneince of having to build them on the same planet.

Could be...  I'm more thinking that the benefit of defensive-only ships is to be stronger and basically immobile, though.  I'd leave this as a secondary discussion for after the other changes are in place, since then we would have a better understanding of what it looks like.
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 x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #27 on: July 15, 2009, 02:18:23 pm »
I find that attack turrets, in general, are nearly useless for defense (in my style of play/opinion, of course). Then again, I rarely get more than one or two turrets researched, and the enemy rarely uses Mk I enemies against me. Mostly, I put them there just because I can, but usually I load up on tractor beams, mines and snipers (which I can put far away) and ignore the rest. They just don't hurt the enemy (although I don't know about the higher level ones) enough to rely upon them for defense.

That's interesting -- when paired with tractor beams, they are pretty fatal against most ingresses in the games I play.  But I tend to put 40+ turrets on most wormholes with turrets like that.  Even just the Mark I standard turrets can eat through an incoming wave or two if built carefully...
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 x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Upper Level Bombers a Requirement?
« Reply #28 on: July 15, 2009, 02:27:52 pm »
I would make some sort of warning for when that even occurs, maybe even a countdown timer for before they get released like with the waves, to give the humans time to prepare.  I think that could add an interesting angle to the defensive side.

What's with all the warnings? I mean, why isn't the enemy allowed to do a silent, strong attack? If you want warning, shouldn't you post scouts in every system and then watch them moving?

I think the AI needs more unpredictability (and variable sizes which can occasionally have a chance of some minor success) in its offense, not more warning and advanced notice. :)

Well, the game is really a mix of RTS and Tower Defense, the warnings being one of the main holdovers from all the different aspects of Tower Defense that I added.  It all boils down to interesting decisions, and the scale of the game.  With a game of this scope, in realtime, you can't afford to just keep super strong forces on every planet you take (especially if you take more planets than what you personally take).

Thus if the AI is smashing stuff blindly into you, it creates basically a lot of randomness instead of actual strategy.  When you have warning, even a brief bit of warning, then you can try to do something with your forces to react.  It's like troops marching across Europe in the world wars, and the other commanders moving to anticipate them.  Sure there were surprise attacks (as there are in AI War, especially with the special forces ships bunching up before going through wormholes now), but those were generally small in size or else they were not much of a surprise.

Yes, we can point out several examples of large surprise battles from lots of wars, but I would argue that the defending side in those battles were not really using any strategy there, just advance planning and then tactics when the battle came.

So, without advance warnings, I see these possible outcomes for AI War:

1. New players constantly get pummeled and die, then quit the game.  Also, experienced players constantly get pummeled and lose planets, but are good at playing cleanup and rebuilding.  Then there is a LOT of rebuilding.

2. The game is rebalanced so that it is easier to defend all your planets, or so that the AI attacks with less.  Then there very little strategy of where to place your troops, because you just place them everywhere.

3. Extra emphasis is placed on some sort of micromanagement-heavy scouting, which alerts you to attacks.  If your scout dies unexpectedly, see #1.  This then gets rebalanced out so that either scouts are sturdier, or you are back to an automated warning system like you currently have.


At its core, AI War is a game about decision making and logistics, not about reflexes or quick-clicking micromanagement.  The scale of the whole thing pretty much requires this sort of style, but that was also the style I wanted to have.  The warnings provide decision points for the human players, so that they have a chance to shuffle their constantly-outnumbered ships around into an effective fighting force.

Look at it this way:  thematically, the AI has vastly superior numbers, but the human players have vastly superior intel capabilities and the ongoing element of surprise (to a degree).

Of course, if you take issue with this sort of thing, I can always add lobby options that turn off the warnings.  If you can take the beating, and like that style of game, then more power to you I guess.  But I don't see ever having larger-scale blind attacks in this game moreso than what the special forces already do.
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 Admiral

  • Hero Member
  • *****
  • Posts: 547
Re: Upper Level Bombers a Requirement?
« Reply #29 on: July 15, 2009, 02:29:43 pm »
When paired with tractor beams, they are pretty fatal against most ingresses in the games I play.  But I tend to put 40+ turrets on most wormholes with turrets like that.  Even just the Mark I standard turrets can eat through an incoming wave or two if built carefully...

40+ turrets on a wormhole? I'm lucky if I build 4 tractors and 8 each of MLRS and short-range. Too many wormholes to defend to put 40+ on each.

Cheers!