Is this the worst game I have ever bought?

[wodin]

NO!

Dont start making "gamier" games. Keep with the simulation.

I love more pieces, more rules, long games and loads of detail! Though I also need a damn good tutorial manual.

[Ursa MAior]

Well the real fun is always playing against a human opponent, the only problem is is the time. So AI plays a part. I would love to see a 'simpler' game, with a very challanging AI. I dont know whether the old Battles of Napoleon fits into this category, but I stil play it from time to time.

IRL the commanders were not faced with a quzillion of solutions only a handful one. I'd like to see a game similar to this

http://www.pbs.org/empires/napoleon/n_war/ibs/game.html

only a bit more complicated, with small movies depicting the outcome. IMHO it would worth a try to produce a similar game (low budget) and test the audeince with it.

Just my 2 cents.

[carnifex]

but the opening move libraries for chess are taken from books written by chess grandmasters and so it seems unlikely that any game I might design would have such a readily available resource.

Unless the game helps you build such a library. For a hypothetical example, when COGII runs, it asks the player if he would like the game to submit it's local knowledge to the central tactics library, and after it uploads it's own data, it would download the collected wisdom from all the other COGII installs that reside in the library. Stuff like "I am Austria and I concentrated on economic builds and got my ass kicked, so now all economic choices get a -1 priority rating etc" would get propagated to all the COGII installs and everyone's Austria AI would learn from that.

If you sold X copies of COGII they would all work together on improving the AI. Upon release the AI would be pretty standard, but after a couple of months it would have the accumulated wisdom of all the other AI's.

[Reiryc]

I'd say no to a simpler game with better ai...

So long as the multiplayer is good and robust.

[Hard Sarge]

Well the real fun is always playing against a human opponent,

you know, as a Old time Board gamers, I got to disagree, most times playing a human, is more pain then enjoyable, even worse when it is PBEM

I would much rather have a decent AI then be able to play another person

I can rule lawer with the best of them, I can out play many people, just because I know the rules better, or what is needed to win better then they do

(I had a friend trying to teach me his new Waterloo game, and I totally got smashed, so bad in fact I only had 3 Cav units left from the Allied armies, and he just marched across the board to the exit points, while I danced away from his couple of Cav he sent out to watch me, he was getting mad as I wanted to play the game out to the end, he ended up moveing all his troops off the map and then followed with his Cav, to gain even more VP points for exiting the map, so I moved my 3 Cav were I want them, and end of the game declared I was the winner, he blew up, how could I even try to say that, I told him to read the rules, he needed a uncut supply line to every unit that left the map, there were only 3 roads that connected his supply lines with his exit points, I had Cav sitting on each one, he wouldn't play me any more)

give me a decent AI anytime

[msaario]

Ok, got a good discussion going on, though not a new subject in the Matrix forum...

Anyway, I am not one of those 20 year old pimple faced boys spoiled with UT shootemups even though I do enjoy the Call of Duty series quite a bit (MP44 rules!). I have found (well this is a personal problem with the amount of free time in my hands) that one of the most important factors in a game of this scale (or any, really) is the interface - and how it supports the game play. Just look at Civ 4 (ok, a billion-ish dollar budget) and its UI. To play the game is really, really easy, yet there is a lot of complicated stuff there if you want to micromanage. Good, working graphics supporting the game itself helps to put in to the "mood". War in the Pacific has good enough graphics - nothing fancy and the animations are kinda sad really. I have spent a lifetime playing it (even with two kids 3 & 1,5 years old) because it is easy to approach (after all other Gary's games). The game play/flow is fantastic and that really makes is shine.

So far COG has been a bit intimidating with a complex economical and political system (++ really) but as we all know Napoleon's greatest achievements took place on the battle field and therefore this side has to work very well too to make a great game. Comparing (at least the looks, can't really compare the game play) tactical part of COG with Cossacks or what-was-the-other-one game is a bit uninspiring. COG just looks dull - however I do enjoy playing it which has kept me going despite many other features I have not familiarized myself with yet. From WW2 we have all the video footage we can watch and therefore it is easy to understand how things also look like in the real life. From Napoleonic times we only have drawings and paintings, so if a game can "bring it to life" I definitely am interested.

So in short what I am saying is that also looks do count today. It doesn't have to be the latest and greatest FX, but definitely a well done map and some little guys shooting at each other instead of ... dots. Thank god there is no "Strength 10" but the actual number of men!

Thanks & I will definitely buy the sequel!

--Mikko

[Grand_Armee]

I truly believe that in game where there are so many choices like COG, that an AI will always be beatable given time. There are just too many variables on any given detailed battle. An AI can't learn, plain and simple. CAn an AI be programmed whether to KNOW when to choose an all out speedy assault or to KNOW when it is better to build a solid formation, and be a bit more ponderous in moving? That's a lot of thought process for a computer. Sure, you can set things like that to happen any number of times...but can you set it to KNOW? Can you program an AI to recognize advantage? CAn you program one to prepare to have an advantage?

This game has too much going for it to be thrown in the can because you're tired of the AI. Hell, I'm tired of playing the AI. I only play against a human. Fortunately we both have one day a week where we can spend hour after hour flogging the AI states and eachother. And hopefully someday we can find more to make some great MP games.

When we are as far flung as we are, you have to build relationships with opponents...it takes some committment. I set my alarm for 3:55AM every Sunday just so Garoco and I can play.

One of the beauties of this game is that once the issue of diplomats being unable to change their function for anyone but the host, there are no need for house rules. The game tells you when your army is in disarray and you should consider retreating, the game tells you when you have lost the battle...or when you have won. Best yet, it tells your opponent when your nation has surrendered. There is no need to quibble...no chance, or need, to find a gamey way to win the war despite the fact that your army has ceased to exist while your opponent's is capable of further action.

AFter 20 years of boardgames and disappointing PC games, this game is miles in front of the rest. I believe that waiting for the ultimate AI opponent is like waiting for the Second Coming of Christ...you expect it, maybe even pray for it, but it may not materialize in your lifetime.

[msaario]

What if the vendor would release the AI code to public domain and allow seasoned players to provide enhancements? Just an idea...

Just a guess. To me the AI matters, I cannot commit to a human player for a long time.

--Mikko

[Gil R.]

ORIGINAL: carnifex

If you sold X copies of COGII they would all work together on improving the AI. Upon release the AI would be pretty standard, but after a couple of months it would have the accumulated wisdom of all the other AI's.

Too risky -- isn't that how the machines in "The Matrix" learned enough to be able to take over?

[ravinhood]

Would you and other Matrix players prefer games with simpler rule-sets if they had more challenging AI?

Hell Yeah!!

Games I enjoy with a challenging AI to me.

1) Civilization IV (Prince & Monarchy are very tough now if you use the aggressive AI in the options setup and raging barbarians)

2) Colonization Win95

3) Alpha Centauri (higher difficulties)

4) Spartan v1.013 (one of the very best and toughest AI's)

5) Mad Minutes Bull Run

6) Transport Tycoon (heh it's great for a long time, but, is easy to exploit and beat)

7) Warlords III (if you don't use the Dragon Merc cheat)

8) Kohan II (the AI that I created is a killer if not stopped early).

9) Galactic Civilization (still waiting on GalCiv II to arrive but I believe it will be just as good if not better)

10) Age of Wonders (origional) the others in the series suk

11) Tin Soldiers: Caesar & Alexander (these guys have the right idea I think)

12) Medieval Total War (using Medmods & 16 Kings mods and after Vikings invasion was released. Very fun and challenging game on the other hand not a single mod or patch ever made RTW the game MTW was.)

When I look at what games I play most because they have a good AI, Sid Meier always seems to take the top billings. Why? Because Sid realized a long time ago that AI's aren't the brightest thing in the computer gaming world. So, he made them challenging by adding enough difficulty levels and OPTIONS so the player could tweak the game difficulties enough to get a challenging and fun game out of them. Alpha Centauri is just one of my favorite favorites because it has no real history to have to follow and it just seems to give me everything I enjoy about these types of games. FREEDOM to choose and RANDOMness every game. Random AI personalities and agenda's. That's what more games need as well. Because scenarios get old and boring.

I'll give honorable mention to Combat Mission, although the AI sure leaves a lot to want. Of course I can stack the odds highly against me and they have improved covering the map edges from that exploit, but, it still does a banzai charge near the end of the scenario even when it's winning if there are objective flags still out of its control.

The one thing I'd like to mention here about Sid Meiers games and most especially Alpha Centauri, something he puts in his games I wish everyone would consider and use is when a player reaches a certain point in the game ahead of the pack, ALL the AI's declare war against him and it becomes and all out slugfest for the rest of the game. Whereas other games get boring at the middle and end game, Sid's games provide a challenge from start to finish because of this one AI bit of coding.
<If player winning by % all AI's declare war and attack him> or
<If player power < AI power by % declare war and wipe them off of face of the map early> (I've had some excellent games of Alpha Centauri since the Spartan leader just loves to try to wipe me off the map early if we are close together. Miriam of the Believers does as well) lol

In essence program the AI to WIN, not just be a roadblock or a mere time delayer. Progam the AI to take advantage of advantages, wipe the human player off the map early when that advantage is present, don't just have them piecemeal attack a weak human opponent, send in the guards, send in the reserves, at least like many of us did in AOE online, if you can't win at least ruin it for someone else. hehe

This is also a reason I don't play many socalled historical simulation games anymore. They are too concerned with historical accuracy and the game ends up boring and with a simulation there's not much room for variations or randomness. It plays the same every game. In other words when playing an eastern front type of game, it's always the same line, the same front, the same push with the only difference being where the main push will be, stronger north, central or south with the outcomes considerably the same.

[carnifex]

Too risky -- isn't that how the machines in "The Matrix" learned enough to be able to take over?

I'm not concerned.



I don't want to remember nothing. Nothing! You understand? And I want to play against a real AI. Someone important. Like Zhukov. You can do that, right?

[pixelpusher]

SMAC is one of my all-time favorite games. I still dust if off and play it now and again. It's a great game. (I happen to know that it's one of Eric's favorites, too.) Really like the novel non-cheesy approach to aliens, all the flavor quotes, and the real depth of the opponent players. Also liked all the information-control aspect, with the probe teams and the network linking, and the PSI combat which balanced normal combat.

Sid sez that you want the player to be genuinely torn in between his available options. Usually it's guns or butter type decisions: 'Yes, I really want to build a Library ... but I also really want to build 4 chariots and squish my neighbors.' Hard choices. Meaningful choices. = fun. Also, the player is the star of the show.

The thing I like most / find most fun about SMAC is that I still come up w/ novel strategies to cope with a given game situation. There really isn't a 100% optimal solution every time. For example:
1) playing a completely non-economic game, and then passing all kinds of UN deterents - with the intent to violate them w/ impunity, but prevent my opponenets from doing same.
2) Creating huge flying chain of cargo planes that linked a series of islands spanning half the planet. Kind of an air bridge.
3) When all my cool land cruisers kept getting killed by enemy air incursions, I started just making tons of them, but with very basic options... then transiting them to the front (1/4 planet away) and upgrading them there to fighting configuration. Enemy still got some, but they cost me less.
4) TONS of wicked possibilities with terrain editing / terraforming. My favorite is the 'former rush': Build a LOT of good sea formers (IIRC, it takes 10), pref. w/ psi defense or armor. Rush them into single square of enemy coast, preferrably next to city and lower the terrain. Goodbye city. Do the same to cut off narrow peninsulas linking one enemy continent to another, or to create sea inlets to rush in your amphibious troops and attack from the sea. One of my other favorite parts of this game.

My only complaints w/ smac are:
-the prototype / upgrade manager interface is very hard/ messy to use.
-the units all sort of look the same. (although I appreciate the novel voxel approach to the 3d units.) I'd kind of want a tank to look like a tank and not a corvette.
-the really cool buildings and upgrades come only at the end ... by which time it's usually over (I think i've only used the teleporter thing once, and don't really build the hover tanks). The later techs are kind of esoteric and star-trekky, also.
-wish that it could continue off-planet somehow.
-in the sequel the aliens are cheesy.

[ericbabe]

ORIGINAL: ravinhood
When I look at what games I play most because they have a good AI, Sid Meier always seems to take the top billings. Why? Because Sid realized a long time ago that AI's aren't the brightest thing in the computer gaming world. So, he made them challenging by adding enough

I concur that in the Civilization series there is not much challenge until the AI receives several factors of production and support bonuses. We give production and support bonuses to the AI in COG at the harder levels, but even at our hardest levels it's not quite as drastically as in the Civ games, at least from what I remember from CivIII. But then players complain that the troop levels are unhistorically large! So can't really please everybody.

When I asked about whether players would prefer much simpler games with a more challenging AI, I guess I meant more challenging without recourse to material bonuses for the AI -- "AI cheats." Just about any AI can be challenging with enough material bonuses.

[ravinhood]

But, if the AI can't be made to be challenging on the harder difficulties "without" the material bonues, then the game has no longevity or great value. It's always been those types of games with practically unlimited difficulty either in levels (GalCiv II has THIRTEEN!!!) or in AI material bonuses at least at the top two tier difficulty levels (Slitherines Spartan has Hardest and Impossible that are extremely fun and challenging for someone like me).

I'm like this and really always have been, make the game as fair and balanced and eqaul and all that stuff for the "normal" difficulty, but, any levels higher make it a monster at LEAST at the VERY LEAST the top difficulty level. Or give us OPTIONS and/or SLIDERS and let us make it more difficult like Combat Mission and SPWAW. A game doesn't have to have a great interface and great graphics to me to be great. It just needs to be challenging almost to the point of unlimited challenge by OPTIONS. I'll buy just about every game like that, especially those where I can piddle with the AI script (Kohan II and Spartan) or at least enhance it with options and sliders.

I know games have to be made with the 13 year old in mind, but, do they all have to be made where the 13 year old can beat the highest difficulty? I'm asking for one difficulty level at the very least for older more experienced strategy gamers like myself and many others here. It's not like we are asking for BLUE (though that would be nice )

It just has to say "something" whether Sid's AI's are good or not, his games are the most "challenging" of all the games I own. I find myself going back and playing old Sid Meiers games than I do any other (cept MTW). If people didn't spam about it, I'd never know or even think the AI "cheats", the way I look at it, it got smarter than the lower difficulties, even if it is given extra resources and other advantages. I don't look at it as cheating, never have. It's all in the "frame of mind" you look at the game in. Where there's "challenge" there is no cheating to me.

[Ursa MAior]

Whats wrong with simpler games? A short vers can be as meaningful as a 2000 pages baroque novel.

[Zakhal]

complex vs AI

Without AI complex game is useless because there is no game to play.

Galciv had a great AI, but the whole game was built around it (its simple) so it was expected. Galciv2 is suppose to be complex. It will be interesting to see if they can still have a challenging AI.

[ravinhood]

Psychologically I'll play a game over and over that I seemingly can't win, vs a game I can win constantly at. Back when I used to could play chess with my friend P., he would thrash me game after game after game, there was nothing like the feeling when I finally beat him at the game. I enjoy seeing that in Sid Meier games. I have yet to beat one of them on the most extreme difficulty on a HUGE map. Now, I know how people cheat and exploit the game by playing a tiny map with only 2 ai's. That's not beating the extreme difficulty to me, thas just exploiting the mechanics of the game in favor of the human to the extreme. I don't know of anyone who has ever beaten the highest difficulty of Civilization on a HUGE map and using ALL the AI's. And I wouldn't believe it till I saw it IN PERSON hahah, none of this hacked saved game bs for me, I GOTTA SEE IT! Alpha Centauri is the same way and even though I haven't won on the most extreme difficulty doesn't mean that I never will. At least I still have that hope in the back of my mind that someday I can beat it. To many games today right out of the box on the most extreme difficulty are just too easy (RTW most especially) I'm still pissed about that one. When I play Alpha Centauri I always use a 128x80 map as well. Lots of computer AI towns and totally a blast trying to take them out. Same for Civilization IV, I immediately found out how to increase the size of the HUGE map in that game and play on a pangea map (AI plays better when it doesn't have to use naval units), wonderful game simply wonderful. Can't wait for GalCiv II now.

[Mr. Z]

In general, I agree that there are some fundamental design differences between the Sid Meier-style games and ours. We're probably all fans of Sid Meier, but as ravinhood pointed out and Eric emphasized, there are fundamental differences between open-ended strategy games and historical simulations. We are definitely bound by certain constraints--if CoG were permitted to be open-ended, there should be no compalints about how far North and West the Turks get! But we get them all the time, so...we see that we have some self-imposed limitations. If amping up the AI means Turks in Scotland and million-man armies, then that's just something we can't do. We have to stop somewhere and make the advantages more subtle, like economic bonuses and tactical and morale bonuses.

So, one advantage of an open-ended game is it's easy to make the game simpler in general, and beef up the AI by just giving it ever-increasing multipliers to everything as you go up the difficulty levels. In CoG, at least, that's something we can't do (and probably is impossible for any historical simulation.)

In historical simulation, a lot of the fun, imo, tends to be either at the level of "I want to see how the map could have been different" or at the level of "I want to see how the individual campaigns could have been different." Either you think that kind of thing is interesting, or you don't. In CoG I guess you could say we tried to accomplish both. Now if you don't think either is interesting, and just want a brand-new challenge every time you play, well, then I guess CoG just isn't your style of game! (BTW, don't forget about the Balanced scenario!)

OTOH, take a game like EUII--that's strategy-level only, but it does provide the kind of open-endedness that you get out of a Sid Meier game. One of the advantages they have is the scale of time they're dealing with--four centuries, whereas we only have 23 years (in a time-limit game, that is.) EUII can be much more flexible with development schedules and so forth, in a way that CoG can't. Furthermore, EUII is not a historical simulation! Its closest comparison is just a game of Civ on the Earth map. So they're actually two fairly different kinds of games. We did try to build a little bit of that into CoG, but as you can see we always have to deal with the limits of history--and the "No Turks in Paris!" fans are always going to be at odds with the Civ/EUII fans. We can't really control that, though we could change the kind of game CoG is. I think it would be hard to strike a balance, though.

So I think the decision about how to tweak the AI is kind of a decision about what sort of game you prefer, and what sort of game CoG is/should be. Just my two cents.

[O.O. Howard]

Friends,
I am a runaway from civ4 to CoG. I really wanted to see a good napolean scenario with commander units that could choose to buff their stack with either movement or attack or healing(depending on the ability of the commander). i had some other ideas too. People really like to make too many unique units which get overpowered and imbalanced, yet instead i think unique Commanders would give a wargame feel to a scenario. But i am not a programmer, so all i could do was hope someone would work on it. what they came up with is a good try but it doesnt work for me. So.... i got lucky and found CoG. i havent played civ4 since. Tho i will sometime.
EU is ok. I much prefer this. CoG combat is enjoyable, EU's is bland. They focus on global empire management. In CoG I especially like how the battle map is the correct terrain type but not the same map every time. Although if i may claim this: years ago chatting with some friends we had a disagreement, they thoughtt the perfect civil war grand strategy computer game would have every hill and river in exact place. i said the terrain type in your hex would be exactly correct, and it would then determine a randomly created but appropriate battle map. that was a looong time ago, but my idea was a good one. You would be bored quick if the terrain was fixed in each province the same each time you fought there.
I think the Leaders in this game are so much more personable. it pulls you into the game.
Now the explaination of why the AI isnt so great taught me alot. I guess we shouldnt to expect too much. But the fact that it can play at all is enough for me. The idea that designers should make simple games with great ai is not appealing to me.
The game has some rough spots. is the game 'done' or is it going to be improved? I have read about how complicated it is to play a live game over the internet. True? and thats where it will stay? i realize it is unreasonable to expect super ai in a game this complex, but is it unreasonable to expect the game to be a finished and completely working product? Maybe within a year of its release?
And on to the next project... american civil war using the CoG engine. Make the battlemap much bigger so you have to hunt for your opponent more.

[Grand_Armee]

ORIGINAL: Mr. Z



...beef up the AI by just giving it ever-increasing multipliers to everything as you go up the difficulty levels.

This is the trouble with any AI. It cannot learn. It cannot detect a weakness or a flaw in a (your) plan. It can throw this or that forward...according to the number of variables it has been given...but it cannot know whether this will work or not.

For this kind of activity you need a living, thinking brain across the map. No AI will please anyone after a short while...simply because it cannot do things that we humans can do.

The cagiest thing I've seen the AI in COG do in DB is go after my supply wagons...easily taken care of by placing them behind a unit.

The only PC game I never completely mastered was that nasty classic Waterloo by SSI. And the only reason it wasn't beatable repeatedly was that it gave the human player so few options to choose from that the capabilities of the human player were hamstrung.

AI's can only win when they have so much mass on their side that the number of options you have is too limited to account for all of their options/variables...it is this way with every game I've ever seen. To me, that is no fun.

It still surprises me how few out there actually go to the trouble of finding human opponents. There is no better test of your skill...even if you occasionally are beset by an annoying bug.

To me, there is no fun in bragging that I defeated an AI army of 200K Turks with 50K Frenchmen. It was cool at first...but now it's just old hat. Humans always find a way. It is only when you are beset by somebody else who must also find a way that you will face a challenge. Until then, you are just learning to master programmed variables. Boring.

Try a real opponent...or wait until Skynet comes online. I think with the former you have a better chance of a good time before Alzhemiers sets in.