As has been mentioned, tactical combat is an element of the warfare side of the game. This side does not need emphasising further as it already, as with most strategy games, occupies a pretty dominant position. A much more interesting use of developer time, imho, would be to expand upon facets of the game currently underdeveloped - espionage, technology victories, diplomacy, etc.

Some of those arguing for tactical combat seem to view it's omission as a deficiency by default. Sometimes a feature which adds something to a game actually destroys what was originally appealing about it. 3d in games is an example of this - reviewers are quick to castigate a game for not being 3d, but some games (eg worms, commandos) lose something with such 'innovations'. Tactical combat is the same - it alters the balance of the original premis, making combat the most important and developed element. Making such an addition 'optional' is really ducking the issue. A real-time option could be added to Galciv2, but just because it's optional doesn't make it a good idea.

Computers aren’t any good with tactics, it’s the number crunching they kick you with.

But what if they were? What if it was great, a real challenge. Maybe you got to the point where you let te computer handle things, because you couldn't take the strain? Man that would be heaven!

Actually computers are very good at tactics once they know 'all' the options, remember the computer that beat Kasparov at chess, Deep Blue was it?

Majority? Minority? I honestly don't know how many people want tactical combat, but there are probably more than a few.

To be honest i didn't really consider tactical combat until i'd played a few (okay, a lot of) games, and noticed that the tactical viewer is just a little bit gimmicky, the appeal soon wears off: Wow my red laser is now green and doesn't the ship i spent an hour making kick ass! it cannot go on forever, can it?

I'm going to say it again - tactical combat rules! (If it were done well)

Its also quite refreshing to hear the opinions of professional programmers when describing why it is too difficult or too much needs to be changed or whatever. Is it really? In what way? tell me more....

Take it easy people, if it happens, it happens.

Honestly, I don't think actual tactical combat akin to Moo2 would help GalCiv2 out, simply because of the way the game is structured and designed (the rock paper scissor combat for example). Taking out this would be a serious mistake because it is one of the most interesting ideas I've seen in the genre for a long time. I can live without the black hole generators if I can switch weapons and beat their heavily shielded ships. It is especially interesting because the AI can adapt and use the system fairly well. Sure, it's fun that your ship can do interesting things, but if the AI just can't adapt to your strategy, then it would ruin the game.

With that said, I do believe that they could add more to the actual combat than just a simple "watch your ship shoot" report. For example, I would like the game to pause between each firing round and allow you to retreat from a battle if you're getting hammered. Add a "fight to the death" auto-resolve and a skip button and that would be as much combat interaction as I would want from GalCiv2.

There's a reason I'd play Civ 4 over Rome Total War any day: I prefer to focus on the overall picture rather than being bogged down by tactical combat. Sales figures for Civ 4 and Galciv 2 seem to agree that there are several people who don't mind not having tactical, when the rest of the game is so superb (because of the focus on it).

Voila.

I'm all for tactical combat, but it's too late to put it into Gal Civ 2 now. There's cost, time, and I think it really needs to be designed into the game from the start to make it good. Maybe you should be asking for tactical combat in Gal Civ 3.

Then there are perverts like me who'd like to have ai-controlled combat but with somewhat more random/unpredictable results. Drop in weapon ranges, ship positions + speed and you'll get banged-up fleet after engagement. Not usual mint condition one sans 1 beaten up one.

Surely far easier to implement than tac AI?

hell, even "try to run away if badly damaged"-behavior would be nice addition..

I enjoyed MOO2, but the tactical combat was the hugest player exploit available. If you thought you would lose a battle all you needed to do was go into tactical mode to fight it out. If you wanted to lose, all you needed to do was auto-resolve it.

The one truth is that if this feature were to be put in something else would have to be taken out. Stardock (like all game companies whose name is not Blizzard) has time and money constraints on each project. Every hour spent on tactical combat would come from some other feature.

Sorry for a somewhat lengthy post, blame Frookie .

But first:
Personally I prefer Rome total War to Civ4 any day (Release the DOGS!!! ).
I like Sid Meier’s games, but he was way of with Civ4. I still play Civ3 though.
Just needed to get that out of my system…. Puuh.

But what if they were? What if it was great, a real challenge.

And what if there were fairies and unicorns and elves and ponies!! That would be even greater
Unfortunately it’s not going to happen anytime soon. This is not for the lack of trying but because the tools, the chips inside a computer, is not up for the job.
A chip is not that different from a relay (Link), it just contains a whole lot of relays. Each little circuit breaker inside a chip can have only two positions:
On and Off, more commonly known as 1 and 0 (sometimes also as Yes and No).
This right here is the core of the problem. When humans think, there are also settings for ‘maybe’, ‘most likely’, ‘not a chance’ and a whole lot of other things. This is in turn the reason why it is so hard to make computers act like humans.

I’m sure we will get to a point sooner or later when computers (or, most likely, the next thing) can make actual decisions. I don’t agree with the opinion that this would be heaven, though. Subject for another thread.

Actually computers are very good at tactics once they know 'all' the options, remember the computer that beat Kasparov at chess, Deep Blue was it?

Here is where you are wrong. Computers don’t use tactics. They never have and they never will. ‘All’ they do is respond to a preset number of values, witch in itself is no small task to program. Tactics requires an ability to imagine and adapt, abilities computers don’t have. Remember: it’s not the computer you are playing, it’s the program. Essentially every decision a computer makes is derived from an IF-statement. The IF-statement compares two numbers (although they sometimes are represented by other things, such as strings (note my mad rhyme skills )) and depending on witch number were higher it either do something, like move a pawn, or go to another IF-statement.

To appreciate the complexity of an AI, try to imagine a simple task such as going to an ATM to get some money. This is something most people have done in their life, so I imagine you are all familiar with the process. Then start writing down every single task needed to get that money. After that, look over the tasks and write down the sub-tasks. Repeat until done. I’m sure You get the idea. If You add control functions as well (do I need money, do I know the code, is there a thug behind me in the line, etcetera.) it’s easy to see that the program will grow. This is still a relatively easy task we are doing here, but even so, as soon as we get randomised input from a user the program will grow to a gargantuan size.

Yes it was Deep Blue. George Plimpton (American journalist) wrote a small piece on that worth a read at Link.
Chess is, however, a relatively easy thing to do. That’s why there are plenty of free chess programs ready for download on the net. Of course, to beat Kasparov you’ll need some next level stuff. I don’t know for sure, but I suspect that the programming in Deep Blue was tailor made for Kasparov’s tactics. In any event, there are only so many moves You can make in chess. Sure it’s a great amount of different combinations, but it is finitive. Still it took a while to pull it of, even for the mighty IBM.

I agree that the battle is a bit gimmicky, but the battle report is quite useful in my opinion. At least for as long as You are able to change the ship configuration .
You are probably also right in that there are more then a few that wants tactical combat. I don’t think it is enough to warrant a major rewrite of the game however. It’s just not that kind of game. If you want to make a game like that You will have to make design decisions to support it from day one, so it’s not something you can add later on like spoilers for a car.

I'm going to say it again - tactical combat rules! (If it were done well)

Right You are, good Sir.
It’s the ‘Well’ part that screws things up

Its also quite refreshing to hear the opinions of professional programmers when describing why it is too difficult or too much needs to be changed or whatever. Is it really? In what way? tell me more....

Do I see a bonus rating coming?
Well, I am a programmer, but if You really want the inside scoop of an AI, then I suggest You start praying for frogboy to come and visit the thread. Hopefully he has better things to do, though. Like the 1.1 patch and other things that carters to our craving for bling-bling while crushing an entire galaxy underfoot .

The biggest part of programming isn’t punching the keyboard. That’s just typing. Most of the time is spent thinking of what a problem really is and how to define it. Then you start thinking of what the solution would have to do, much like the ATM-example above. As far as AI is concerned, my experience is quite limited. In a way, every program has an AI to some extent, but most people who hear the word AI think of something rather far from what it really is. Artificial Intelligence only contains the word intelligence but no real intelligence as a human would define it.

Once I was tasked with the writing of a new sorting algorithm. Nothing strange about that, most programs contains one in one form or another. Because of how the rest of the program (a simulator for training military and catastrophe response personnel) was written, and to better support future development, an ordinary sorting algorithm wouldn’t do. Instead it would have to sort in accordance with internal priority compared to other objects. This is monkey-work that any human can easily do without much effort. To teach the computer to do it, however, took me the better part of two weeks just for the algorithm. Planning it and typing it took about four days. Debugging it…. Oh boy!
The end result became rock solid, however. But it shows just how much time and effort it takes to teach a computer to do rather mundane and easy tasks.

Take it easy people, if it happens, it happens.

Not going to happen

Sorry for the long post, I hope some of You find something of interest in it.
Also sorry for riding the OT-border, but it seemed somewhat relevant to the issue at hand.


Rygel XVI - Dominar to over 600 billion (more or less) loyal subjects.

Yeah! Bring on the fairies and elves too! but you can keep the ponies though!

Thanks Rygel ( can call you Rygel an leave out the XVI?) for being an honest to goodness programmer and coming forward with the lesson in Logic.

I do something similar at work (Systems Analyst, yup i'm a data pusher) i take existing programs and make changes to them, where possible, and build new ones where i have to. you are right though, its a major mission to build new algorithms and debug/ test/ test again/ let someone else test it/ get it signed off.....AAARGH! I always take a look at my old progs( and other peoples) to see what i can use before starting a new project....

the thing is we already have the Logic in place at the moment to begin Tactical combat, it just needs to be refined. I mean how do those AI ships move in the galactic screen etc etc.

it 'might' even be a little easier to implement as the area of movement, no of ships, types of ship, no of enemies, is now greatly limited by logistics, battle area etc.

You catch my drift??

As long as we're throwing credentials around, I'm also a programmer with more than 20 years experience (the getting-paid-for-it kind of experience), and while I generally agree that gameplay AI is very complex, it is my opinion that a good strategic AI is overwhelmingly more difficult than any tactical AI, and the GC2 strategic AI is pretty good.

The reasons a strategic AI are more complex are pretty simple, too. There are generally multiple win conditions, multiple strategies available for each win condition, a significantly longer timeframe, a lot more that can go wrong, and so on. Basically strategy is far more complex. A tactical AI simply has to win a single battle, usually using fixed assets and capabilities against a known force.

That being said, I don't foresee it being added to GC2 either -- and I keep flip-flopping about whether I'd like to see it or not. I suppose it would be a neat option, if I could disable it when I didn't feel like getting into the nitty gritty fighting. And I also have to agree with one of the earlier posters that even the best tactical AI still usually isn't very good against the humain brian, so I'm not sure it would be worth the effort.

b] The tactical combat in both the previously mentioned games was pretty simple. I believe the Stardock team could develop a tactical engine as good as either of those without batting an eye.

They could. They won't. One could argue all the issues about coding in an entirely new combat system, etc. but what it really comes down to is the above statement. Stardock won't create a simple AI for tactical combat because they know it will be cheese. Frogboy has stated several times that the only way tactical combat would ever find its way into GC2 would be IF they could create an AI good enough to beat human players on a regular basis. IIRC, "considering the problems with getting the AI to use starbases correctly," is part of the explanation of why tactical combat won't exist in GC2.

I'm sure many people would love the feature of tactical combat in GC2, but it simply won't happen. The Stardock team won't put in a combat feature they know would be more aesthetic than challenging.

A chip is not that different from a relay (Link), it just contains a whole lot of relays. Each little circuit breaker inside a chip can have only two positions: On and Off, more commonly known as 1 and 0 (sometimes also as Yes and No). This right here is the core of the problem. When humans think, there are also settings for ‘maybe’, ‘most likely’, ‘not a chance’ and a whole lot of other things. This is in turn the reason why it is so hard to make computers act like humans.

Well, I certainly don't want to get off on a rant about analog computers or fuzzy logic, so let's just say that's not the biggest impediment to adding a tactical component. I can think of several other more important issues right off the bat:

1) There are currently no decisions worth making. With unlimited range on the weapons, the best tactic is always to attack the ship with the highest offensive/defensive ratio, and the ai already does this.
2) Retooling the tactical model to factor in range, position, etc. is nontrivial, would involve a lot of recoding not only of the combat processor but also the tech tree - need to add ranges to those weapons systems, after all, and in the end wouldn't add much - the best tactic would then be to maintain tight formation and still shoot at the the enemy with highest off/def ratio.
3) The interface for controlling your fleet in true 3d would be a minor bitch.
4) The pace of the game would drop considerably.

As long as we're throwing credentials around, I'm also a programmer with more than 20 years experience (the getting-paid-for-it kind of experience), and while I generally agree that gameplay AI is very complex, it is my opinion that a good strategic AI is overwhelmingly more difficult than any tactical AI, and the GC2 strategic AI is pretty good. The reasons a strategic AI are more complex are pretty simple, too. There are generally multiple win conditions, multiple strategies available for each win condition, a significantly longer timeframe, a lot more that can go wrong, and so on. Basically strategy is far more complex. A tactical AI simply has to win a single battle, usually using fixed assets and capabilities against a known force.

I agree. But the real problem with tactical combat is not related to tactics. For a given ship design the AI can solve the tactical combat perfectly. Probably.

The problem is that the ship design is far more important when you include tactical combat. And ship design is one of the most difficult strategic decisions. And GC2 has already problems there.

I think any inclusion of tactical combat will be a horrible failure when it is not based on multiplayer results. You really need excessive beta here to derive scripts for BASIC ship designs then. Otherwise this task is far to complex for an AI.

Frogboy has stated several times that the only way tactical combat would ever find its way into GC2 would be IF they could create an AI good enough to beat human players on a regular basis.

Agreed - unconditionally. Does that mean somebodies working on it right now? (fairies and unicorns again )

Good call Sloth_DC except that the combat simulatory gizmo is in 3d, but the actual games is pretty much tabletop style, so the combat needn't be 3d. other similar ideas to consider would be 'flanking' opponents and getting them in a crossfire, getting shot from behind is also horrible so would we be having to place our defensive systems strategically on the hull, even weapons could be arranged for similar types of combat such ar 'broadsiding' enemy ships or strafing. the tight formation idea is good unless you get 'enveloped' and then you're toast. And don't forget the marines!

All this would add richness to the combat side of the game where at the moment its almost simply a case of who has the bigger guns and the fastest ships.

Moo2niac - your case for more data is valid (Is that a sultle shade of multiplayer pushing going on? ), so i better start uploading my games then, and get some of those stat thingies people have on the bottom of their posts (they look pretty cool). the enemy ships i face are big on guns, little on defence and de nada on the speed. from what i've noticed the AI wants to crank em out as fast as it can so if the defences were cheaper - spacewise and cost - maybe the first attack wouldn't hurt so much.

REWIND - RYGEL I have to agree that civ4 was watered down a lot, but its not their fault, they are now called atari and that means bad decisions are going to be made a lot from now on So no new versions of X-com and no chance of alpha centauri2, but i hope i'm wrong. Saying that, the great leader aspect was a good touch, but it could have done with more religion specific ideas/wonders/units. I've yet to play Rome total war, its on my shelf, but i haven't upgraded by box to cope with it...soon.

I wonder how 1.1 is shaping up hopefully things are more interesting and the ai now uses planets more effectively and will those yellow food bonus tiles disappear forever??

Skelton42: My statement about the simplicity of Moo2 and BotF's tactical engine implied the ease that I think Stardock could outdo them. This, of course, would be what I would hope for. This added element, along with the ever increasing number of mods, would give this game an infinately longer play life for me and I daresay, many others. Will this ever happen? If you believe the strong opinions of some on this thread [and indeed, on the overall forum] then no. But I still hope for its implementation along with continued development of the AI. Afterall, hope springs eternal!

can call you Rygel an leave out the XVI?

Sure, I have been called worse

Yes, I understand the notion that building a main battle tank from scratch is harder then to just add metal to a Honda Civic.
That doesn’t necessarily means that the end results will be equally good, though.
When You have tried Rome total War you will notice that the tactical combat AI is quite good but the city building AI is a bit daft. It’s quite obvious where they put the focus on that one. In GC2 the planet AI is quite good in my opinion. Maybe not in small universes, but that is to be expected. I think Frogboy wrote something about that in a developer blog.

1) There are currently no decisions worth making. With unlimited range on the weapons, the best tactic is always to attack the ship with the highest offensive/defensive ratio, and the ai already does this.

That’s hardly what I would call tactics, that’s more like slugging it out.
Some times it might be better to start pounding on a supporting starbase although it doesn’t have the best offensive/defensive ratio. I want a tactical AI to consider a lot more then just choosing a target.

3) The interface for controlling your fleet in true 3d would be a minor bitch.

I haven’t even thought about that
Sounds more like a bitch and a half, now that I think about it.

4) The pace of the game would drop considerably.

Yeah, particularly if we consider your third point.

My statement about the simplicity of Moo2 and BotF's tactical engine implied the ease that I think Stardock could outdo them.

From what I’ve heard, that won’t be too difficult

What’s all this jive I’ve been hearing about the removal of food bonus tiles?!
Allow me to illustrate:



Rygel XVI - Dominar to over 600 billion (more or less) loyal subjects.

Joeld 1212 - Don't give up on me man! the war effort needs you!

Rygel - I play large universes and the yellow food bonus tiles always unbalance my setup - green tiles are cool though.
the ai seems happy with the research tiles but not the rest. (1.0X)

Poor kitty! Dude, do you stroke the cat like an evil villain going "MUWahahaha!"??

Btw. speaking of tactical combat, I really think Civ IV is also missing it. Imho Civ IV isn't complete without mortal combat style tactical combat between spearmen and tanks (spearmen for teh win!1!11) )

I love a well-done bit of sarcasm

The real point is, GC is not a tactical combat simulator, it is a turn-based empire management simulator. The only tactics that should even be in it are empire level tactics. However much I would like to see a more realistic fight in the "tactical view", I most certainly do not want tactical fleet and ship level combat in an empire management game. I am Genghis Khan, not Starbuck, not even the original Starbuck.

3) The interface for controlling your fleet in true 3d would be a minor bitch. I haven’t even thought about that Sounds more like a bitch and a half, now that I think about it.

It was done rather well, even elegantly, in Homeworld.

It was done rather well, even elegantly, in Homeworld.

That’s true. The scale was a bit different though, and it required heavy use of the pause key in order to direct the ships to where they did (or sometimes took) the most damage. You’ll lose a lot of speed in those kinds of games, but I’m sure it could be done (now that You led me to the right path ).


Rygel XVI - Dominar to over 600 billion (more or less) loyal subjects.

Frookie: I'm still here and still pushing for the option of tactical combat. Have you wondered why the naysayers so oppose this? I have. Some say this is only a "strategy game" and there's no intent for combat or war even though every single Galciv2 game I've played has ended in war. Others for some reason seem to want to play as many games as possible in the shortest period of time and they think that tactical combat will slow them down. I, on the other hand, enjoy a drawn out campaign against those goody two shoe Altarians. If these people want to speed through a dozen games while I play just one they can turn the combat option off. Still others have stated that any amount of time the devs spend on tactical combat would automatically take away from the AI. I don't see these two as being mutually exclusive. Currently, I believe that Stardock is working on things in addition to the game AI, including enhanced modding techniques, possible implementation of multi-player and a slew of other items. They also take the time to browse these forums to see what the players of the game want. Well, many of us want the option of tactical combat!

I have to throw in with the No groups here. I'm a strategy guy through and through.

That said, I may, and I stress may, be in favor of more Tactical choices pre-combat, similar to the Planetary invasion choices. I would also be in favor of multi-sqaure Combined fleet combat, in the way that you could attack the enemy from two different squares at the same time; thios would give you more Tactical options.

But I would only want this if it could be coded and balanced so the aI could do it too. I would not want a half-a**ed effort.

But as far as finger twitching your way through combat, nah. My arthritic hands can't do that so well anymore. And FPS's designed for that are fine if I want to scratch that itch.

(Off Topic, those of you who like naval combat, (or are true grognards) the Harpoon games provide a good Operational Level Simulation that is pretty close to tactitcal combat in a Strategic sim. http://advancedgaming.biz/ )

After reading Oldstatesman's post, it makes me wonder if we are all thinking of different kinds of tactical combats? I certainly don't want FPS or anything that involves twitching fingers.

I'm thinking more along the lines of BotF's combat but with more options [and of course done better]. Maybe a few formations, each giving a positive [or a negative] depending on what ships in what formation your facing. I'd like to be able to direct my ships fire and give my ships movement orders. I'd like to be able to do this by individual ship or by ship class or by lassoing a mixed group of my ships. After making these decisions I'd hit the turn button. The battle would play out with both my ships and the AI's maneuvering and firing simultaneously [no I shoot then you shoot ala Moo2]. You could pause at any time to reavaluate and give new orders or do this at a preset autopause. If you wanted to retreat, you could.

Of course, you could skip all of this, if you wanted, and auto-resolve. Tactical combat would be an option you could use for one battle but dispense with for the next. Like the current combat viewer there would be multiple camera angles you could adjust .

My idea for tactical combat is all about options; but namely the option for using it if you think it would make the game more fun for you or not using it if it wouldn't add anything to your game.

Want to see turn-based tactical combat done right? http://www.battlefront.com/

Check out Battlefront's Combat Mission -- it's a fantastic series of games.

I guess in actuality it's a mix of strategy and tactics, but if interactive combat had to be added, I'd love to see it follow this model. By the way, Combat Mission is one of the least-buggy games I have ever bought (and I've been playing computer games since 1979). This is especially amazing considering many of their games are produced simultaneously for the PC and Mac, and they're a far, far smaller operation than Stardock. And Combat Mission is an actual simulation, by the way, in that it SIMULATES the real world -- bullet trajectories, armor thickness and material variations, even the angles at which rounds strike tanks -- read up on it if you're at all interested in WWII combat games.

"Borrowing" the Combat Mission model for combat would make GC2 combat interesting, yet avoid RTS-style twitch-gaming. It would also significantly increase the size, cost, and complexity of the game. But hell, we can dream, right?

Combat Mission is the ONLY serious simulation of WW2 tactical combat. One of the best game series' ever.

BTW: did some of you play Space Empires IV ?

For me this was the successor to Moo2 with the exception that it was better in EVERY imaginable way (if you exclude fancy graphics). I don't know how many hours I spent with this game - has to be in the thousands.
I'm still on the fence if GC2 will beat SE4 in the long run motivation-wise.

Back to the topic at hand: combat options: yes ; tactical: don't miss it much.

I'd rather like to see SE4 style multiple research options.

I have to throw my lot with the 'no tactical combat' crew. I see GC2 as empire management game not a fleet combat management game. Good reasons have been mentioned already in this thread. The one that sticks out is that the game balances a certain simplicity with strategic depth and tactical combat would push that balance towards the micro-management nightmare that I ended up avoiding in so many other games I previously tried in this genre.

A second point we all (Both tactical combat naysayers and tactical combat fanatics) shouldn't forget, is that as members of this forum we are very much the vocal minority, when compared to a silent majority who have bought the game and yet never contribute actively to this forum. I wonder what percentage we, the vocal minority, make when we are compared to the overall copies of the game sold? I am of the belief that if those people who have bought the game yet feel no need to come to this forum and ask for tactical combat that maybe, just maybe, the majority like the game as it is. Thats just one interpretation but I have to ask myself that if that many people were unhappy with the game because it didn't have tactical combat you would expect them to make a stand here on these very forums, hundreds of them, nay thousands of them! This is specualtion but they are also my thoughts on the matter.

Have a good one.

AR