Frogboy Frogboy

Fun from last night..

Fun from last night..

Smarter..Faster..

The AI difference from Beta 5 to now is..well just night and day.  Beta 5  as far as I'm concerned had no AI.  Sure, the ships moved around and such but that was about it.

Now the AI reacts to things like influence starbases.  It'll try to counter your influence star bases with its own.  Some races will try to do a culture attack on you. 

And when it comes to combat, no more willy nilly sending ships around ineffectively.  They focus on specific goals and mass up for battles to take those goals.

They will identify a threat, determine what htey need to kill it and then put together the appropriate force.  No more pathetic "Look, it's a lone fighter attacking my mega fleet".

In this example, I had a huge fleet massed for attack.  I was pretty impressed with it.  But then the AI massed an even bigger fleet.  Some AI players go with fighters, others with capital ships, depends on the AI module. 

One late game battle was pretty painful, my brand new "Debbie" (my wife's name) class battle ship was alongside a bunch of "Victory" (older model) class battle ship and the Drengin's fighters were powerful enough to take out the Victory class ships on their own. (speaking of which, I significantly decreased the upgrade cost on ships as a result).

Here's some cool battle scenes from last night:

If you look closely, you can see the Drengin fighters, they're itty pitty but they're nasty little buggers.

Also, the AI is MUUUUCH better at adapting its weapons.  In Beta 5 the weapon analysis went like this:

IF (I am at war) THEN (Find most lethal enemy and counter their weapons).

Sounds good on the surface. But it's stupid.  The Drengin and Altarians had been battling it out all game and the Drengin were finishing them off while I had built a huge fleet of ships that were perfectly suited for whacking the Drengin.

So the algorithm changed:

IF (I am at war) THEN (If my most lethal enemy is much weaker then me) THEN (Look at the most powerful military in the galaxy and factor their weapons and defenses into my counter measures)

18,041 views 42 replies
Reply #26 Top

BTW, this is precisley why we don't have tactical battles.  Because no matter what, someone is going to figure out a trick that we haven't thought of and goof it up.  At least this way, any trick is going to revolve around what types of ships you put into your fleet.

Incidentally, the way it's supposed to work is the AI targets the most vulnerable ship that has weapons and concentrates on it until it's dead.

Reply #27 Top
Now I see why your so against tactical battles. You don't know strategy, so you obviously can't program it in. It's horrible if the entire fleet is going to target one ship and then move to the next. That just means the player with the higher tech level wins. Care to explain why you have an option to view fleet battles? I think you would save so much time and money by just having the battles like GalCiv 1. Or you could at least be able to assign a strategy to a fleet. Like: Targeting priorities - chosing what type of ships your fleet will target first. And whether to engage ship to ship or let the whole fleet go after one ship at a time. That would be easier to program the computer to choose a strategy for its fleets rather than programming the computer to be tactical in battle. It's a lot better to watch two fleets fighting eachother with each ship, (or small groups of ships) firing on another ship. Or at least make the targeting of each ship dynamic. Watching the whole fleet target one ship is boring.
Reply #28 Top
LOL. 
Reply #29 Top
Oh. Drama. Stardock doesn't know strategy.

Do you realise how stupid that sounds?

I agree, LOL.
Reply #30 Top
You don't know strategy, so you obviously can't program it in.

Strange I have always thought that strategy is how win a war while tactic is how to win a battle. But I may be wrong.

That just means the player with the higher tech level wins.

Given the rock - scissors - paper aspect of the battle, it isn't so obvious. If your foe has the correct defense against your weapon but not you, they will inflict much more damage than you. And don't forget that if a cat can kill a rat, it doesn't mean that it will survive if attacked by hundreds of rats. There always has been choice to be made between quality and quantity.

As for battleships and fighter, a battle ship can kill only a fighter at a time (or maybe 3 if you can fire at 3 differents targets when you have the 3 kinds of weapon). So if you need 10 fighters to destroy (hypothesis for this example) a battleship while one battleship can kill one fighter, who will win the battle between 2 battleships and 10 fighters?
turn 1: 2 battleships fire => 2 fighters dead, 8 left
turn 2: 8 fighters fire => 2 more shot to kill a battleship
turn 3: 2 battleships fire => 2 fighters dead, 6 left
turn 4: 6 fighters fire => 1 battleship dead and 6 more shot to kill the last battle ship
turn 5: 1 battleship fires => 1 fighter dead, 5 left
turn 6: 5 fighters fire => 1 more shot to kill the last battle ship
turn 7: 1 battleship fire => 1 fighter dead, 4 left
turn 8: 4 fighters fire => 1 battleship dead

Winner: the fighters.
The attack level of the battleships isn't the key point in this case: if they can kill a fighter with one shot, a better attack rating won't kill more fighters. But the key thing is the attack rating of the fighter against the defense rating of the battleship. And don't forget that there are random number behind attack and defense, meaning that sometimes injuries can be made eve, if the defense rating is gretly and correctly defending the attack of the weapon used.
Reply #31 Top
Give uglycow a job at stardock and she can make the coffee
Reply #32 Top
He'd probably claim that the coffee machine lacked the proper number of buttons and dials that would allow him to brew the all-encompassingly perfect cup of coffee, and as a result, say Stardock doesn't know about coffee, coffee machines, caffinated beverages, or electricity by extension
Reply #33 Top
It is generally easier to build a tactical AI then a strategy AI. The reason for that is that at the tactical level, there are very variables, making it easy to select the "optimal" course. That isn't true for most games that operate on the strategic level.

The trick of battleship versus fighter is a classic problem of balance for SF games. Generally, the balance is heavily set for attack or defense. When it is set at attack, then a fighter swarm will always beat the battleship, because the fighters will have the superior firepower. When defense is stronger, then the battleship wipes the fighters out, because the fighters cannot harm or cannot significantly harm the battleship (due to its superior defense), while the fighters have little to no defense and are easy pickins for the battleship.

To some extent, you can modify the balance through various means, if the game designers see a need to allow for a change during game play. For example, in the MoO series, it allows for certain ship weapons to be "streaming", making that weapon type highly effective at fighting swarms of smaller/weaker ships.

Their are really two questions regarding the balance of attack versus defense, as I see it. They are:
#1) What do the designers of the game envision as the balance point? In other words, at what point can a group of smaller ships be equal in combat capability to one larger ship?

#2) What balance point do the players feel is more fun?

Personally, I prefer the MoO1 sort of balance point. In MoO1, a swarm of smaller ships could be just as effective as a smaller number of larger fleets. In the rest of the MoO series, this wasn't true (Biggest Hull was always superior), and that made those games less fun then their predecessor. Of course, that's just me.
Reply #34 Top
Actually, where those kinds of things tend to fall down are:

1) Coordinating the strategic and tactical AI: The strategic AI needs to know what the tactical AI needs to fill its plans. This isn't as simple as it seems at first.

2) One-trick-pony strategies. AI's just don't deal well with them unless specifically programmed to do so, and even if you do program it to deal with that one, another one will come along. This has been a serious issue with MoO2, Alpha Centari (which didn't even have tactical combat), and many other games. I've only heard of one game-playing AI that ever had even a fraction of the adaptability of the human mind, and that was more because it wasn't blinded by preconceptions. It played the "Trillion Credit Squadron" of Traveller fame and taught the humans a few tricks. I still don't think it came close to winning the tournament, but it did offer a few insights into the game that most humans had missed.

As to the question of what balance point to players feel is more fun, I think I'd like to see the point where both extremes are viable, but with different advantages and disadvantages.

Somehow, I think the upkeep of a fleet of small fighers (in terms of replacements, not in terms of regular maintenance) is going to be higher than a fleet of larger ships, if for no other reason than they'll need replacements more often. Yes, the replacements will be cheaper, but since it takes less to kill one, they'll be needing replacements more often, whereas the larger ships might survive and be repaired.

On the other hand, I don't think that huge hull juggernaughts should be unstoppable juggernaughts, for that matter, since they're only 5 times larger than Tiny hulls.
Reply #35 Top
A comment: Remember that smaller ships will be able to be made on developing worlds, rather then just "core" developed worlds. That allows them to be more easily replaced, as they can be made by worlds closer to the front lines then the big juggernauts.
Reply #36 Top
It is generally easier to build a tactical AI then a strategy AI.


I don't think this is true. It seems to me that tactics are far more complex than the grand strategy. We can mention RTW again. Tactical battles like in RTW (and what some people are pushing for in GC II) are easy for the human. The human has all the information in front of him. The computer really has no good way of tricking the human. Where the human just needs to confuse the computer my moving his units around. Eventually the computer will do something stupid, and then the human tears it appart. If there is an example of a game with good tactical abilities, please let me know. And games like starcraft don't count because the computer has a distinct advantage in that it can do multiple things at once with zero lag time. And, it has fog of war.

On the other hand, strategy seems to get better and better. I have played the Civ series quite a bit. The computer has continued to improve. I have heard great things about Civ IV as well. And from what I gather here, the strategy element in this game will even top that (I really hope so). There is enough fog of war to at least give the computer the element of suprise at times. And seeing as how the designers are trying to eliminate the tunnel vision the computer seems to have, and preventing it from sending units out in a random and disorganised fasion, we will definately see a change from the past. I firmly belive it is easier to program strategy than tactics.

As far as the fighter being stronger than the battleship, I think Peach Phoenix explained it perfectly. Think about it. If a battleship has just marginal attack abilities, it will still be able to destroy a fighter in one turn. However, if it has superb defense abilities, it will last longer agains the fighters. So there is no definate fighters are better or battleships are better. It depends greatly on how you design your ships, and how you deploy these ships against your enemies. Given a generic system like GC I, things were predictable. But now they are not. Stardock has come up with an exceptional method, and they really DO know strategy.
Reply #37 Top
I think its bogus that 4 fighters could ever take out a battleship. Looks like someone should re-write the combat code.

One thing to limit the fighter to only traveling with a "carrier". That way a personal can't cheaper stockpile hordes of fighters in their system.

Another idea - have point defenses fire FIRST before fighters (reguardless who is the attacker). This way cruisers and other ships which are anti fighter truely have a role in fleet combat.

Reply #38 Top
Only takes one little torpedo now a days to take out a battle ship. Look at the overall picture.

Altarians build a nice pocket battleship names CruiseM1K1 or something of the sort, they are continuously at war with the Yor who specialize in missle technology and subsequent arming. Along comes the Arcaen who working with the Hoomans have developed guns and lasers. Primary armament on them little fighters are guns which the Altarians don't have battle armor to really stop. Now you have a half dozen of these guns all going off almost simultaneously up your backside (what fighter pilot would ever try a head on attack or crossing the T with a battleship... or for that matter a suicide port or starboard runby.) Any pilot would prefer an attack approximately 30 degrees off the backside of any superior craft I would believe in all 3 dimensions of space warfare (which is quite similar to submarine warfare I might add, which is why I feel that the modern submariner, any nationality, will be the ideal technicians and crew for deep space flight btw).

Well the first things them bullets come across is possible some shields and a point defense that is pretty good, stops some of them, the heat from the engine may vaporize a few more of them little bullets, but thats it. Generally speaking there is no point defense in the engineering area, some fine damage control exists but if everyone there is fraking dead and the bullets have taken out the power generators as well as the flux capacitors used to keep the fusion engine gas trail in place to the primary ignitors... you get the point.

A few small ships hunting in a wolf pack can take out a single ship no matter what the size if their primary (sometimes only) battery is unstoppable by the single ship even if it takes most of them out with it.

W/R
Suralle

PS: This is only my humble opinion. Ask my wife, she will tell you I have never been in combat or seen combat. Me, my response is "I dunno".
Reply #39 Top
Since the game DOES have battles, I so wish they would be Nexus-style... Point defences frantically trying to destroy missiles and fighters, big guns pounding on the enemy ships in proximity, missiles flying to distant targets and lasers trying to destroy individual devices on enemy ships... All happening at the same time while the ships are slowly dancing in space, surrounded by asteroids and nebulae... This is a treat... Naturally, that's also why it's a game on it's own.
Reply #40 Top
BTW:

To see what a modern (by that I mean a weapons system first designed in the 80's) can do to a battleship from a small submarine in a diesel electric navy (please remember that many, many, many countries possess these small submarines with similar torpedos) please check out the link below.

W/R
Suralle

PS: The ship in question although DIW would still be a new reef if under power. DIW is actually safer in some regards as torpedos are passive / aggressive and can be fooled by noise and by decoy simulators both of which are typically not carried on many surface ships for some reason.

http://www.janes.com/defence/naval_forces/gallery/slideshow_010202.shtml#
Reply #41 Top
mutter mutter mutter

http://www.janes.com/defence/naval_forces/news/juws/juws010202_01_n.shtml

Try that or just got to janes.com and type in Mk 48 in the search window.

w/r
Suralle
Reply #42 Top
It isn't about what fighters can do. We really shouldn't use modern day naval warfare to justify the way battles in GC II are fought. This is my understanding as someone who unfortunately found out about this game too late for the beta. There are five base hull sizes (if this is wrong it doesn't matter, the argument is the same). So fighter, is just the smallest base hull size. It will be the cheapest, and subsequently the least powerful, and easiest to destroy. And then the sizes go up. How you design your ships is up to you. How you spend your resources on these five base sizes is up to you. Remember, the computer is limited to the same set of rules (unless you want it to cheat). So, if you think that fighters are too strong, then build nothing but fighters. See how the computer reacts to that. Maybe you can learn something. As for me, I will try to teach the computer.