why arent they talking at all about the AI? [Archive]

NagatsukaShumi

06-04-2006, 14:58

The reason they aren't talking about the AI is because they very rarely have, infact if you read most previews they won't talk about the AI too often other than the obvious "if fired at with arrows they take cover in woods" for example, this kind of AI discussion is usually presented.

They will not talk to you about fighting and running speeds, this is not something that concerns the mass market but I can pretty much garuntee that CA will have read comments about it all and will be taking it into consideration, they just won't be making press releases about it as it doesn't concern the majority buyer of the TW product nowadays, in a sense TW is a victim of its own sucess in that it has to adapt to survive scrutiny from the casual gamer. If you took away the history experts and TW vets I garuntee there'd still be plenty of sales, we are hardly the largest consumer group here.

King Ragnar

06-04-2006, 15:16

Because its gona be poor and not there main selling point, they are looking for people who just love the look of the games and not they gameplay itself im afraid.

Lord Adherbal

06-04-2006, 17:17

yeah the "TW vet" might be a small part of the customers group, but who said STW/MTW style gameplay would make them LOSE customers ? I can't believe they turned RTW into a clickfest without any reason. Graphics were RTW's main attraction and selling point, not the uber fast gameplay. In fact no one knew about this uber fast gameplay until the game was released, and most of us were disappointed. So what is the reason for this change ? I thought there was a saying "if it isn't broken, don't fix it".

What I find strange about CA is that they dont seem to include large groups of beta testers to pretest game concepts. Maybe thats a misconception on my part but it seems that way to me (and I've helped out on other (non-CA) games. For example, I cant believe many experienced game players were positive about the running/fighting speeds of RTW.

the fact that v1.0 of the game crashed when more then 30 players joined the MP lobby clearly proves they didn't test much. If you read developer diaries from Ensemble Studios (makers of AoE series) you hear them taking about excessive (MP) testing sessions and teams, and how they find and deal with balance problems. I doubt these kind of test(team)s exist in the TW game's development.

CBR

06-04-2006, 18:01

Yeah the community only gets involved when CA is making a patch. We did MP testing for 1.2 patch but mainly for connection/out of sync problems.

Of course for M2TW one player is involved in testing AFAIK.

CBR

Callahan9119

06-04-2006, 19:14

They have mentioned the AI quite a few times.
Read the previews.
Also they have stated that the game is pretty much complete and all that is left is the AI which is what they are doing now.

funny...i read through the whole faq and all of the sticky threads on tw.com and previews from major game sites and havent found anything but the usual dopey things like someone said "if you shoot them with arrows they will move to trees"

so from what your saying they are doing a LAST MINUTE job on the AI, if its going to be dropped in 6 months that means they have what? maybe 4 months to work on it? its not like they have some great foundation to work with.

and i dont see why it would be so hard to make a better ai, how about one that doesnt move around behind walls eating arrows, one that stays in formation, doesnt send generals to a quick death, doesnt send slingers charging into my front lines as shock troops, builds better units for its armies and knows how to tech up its buildings :idea2:

but who needs AI when you have.....drumroll.....CANNON ELEPHANTS ~:grouphug:

Dooz

06-04-2006, 19:34

How bout we just wait and see how the AI actually works before sticking a criticism pole up it's wazoo. Movement speeds and such aren't a big deal at all as they are very easily changed. There are more important assets of the AI to worry about that can't be changed by modders. Hopefully those more major areas will be improved upon and the minor nitpicks can be modded as always.

Geoffrey S

06-04-2006, 19:47

One of those videos shows AI slightly better than in RTW, at least right after deployment. Hopefully with a bunch of months left they'll still be working on the AI solidly.

Callahan9119

06-04-2006, 19:58

yes, but mods generally dont help much in the mp department...and most of the ai problems have long and dusty roots, as far back as STW...thats why i'm annoyed....all the power of todays systems and the AI has not developed :dizzy2:

i'm not bashing...i have bought EVERY TW product thats been released and <except for stw> the day it is released....and i'll buy mtw2...but that doesnt change the fact i am concerned about its development.

to the guy below me, he mentioned MODDING the game speed....and the RTW run speed IS a problem for MP and its hard to get everyone to use the same mod...or in some instances even the same patch

Dooz

06-04-2006, 20:30

yes, but mods generally dont help much in the mp department...

But... AI problems don't hinder MP because there is no AI.

NagatsukaShumi

06-04-2006, 20:42

Most interesting fact here is, none of us know what CA are doing right now. I heard somebody was concerned with the time CA left to "dedicate to the AI". Put it this way, 4 months of 5/6 working days is PLENTY of time for a skilled group of programmers such as CA to get together and work hard on the AI and sort it out.

The changing of the AI I am pretty certain was to try and cater to the wider market CA were bringing in after TWs sucess, this wider market are the types of gamers who get annoyed when they don't win easily and will drop a game if they think its hard, or takes ages, hence the dumbing down of the AI and the speeding up of the units. I imagine, however, CA have surveyed the communities opinions of RTW and are probably looking at making a much finer balance between the two markets who buy this game.

I think some people need to start having a bit of faith in CA, they created two very good games in STW and MTW and one has suddenly turned them inept in some peoples eyes. I remember playing Metal Gear Solid, the first was amazing but the second was absolutely terrible imo, MGS 3 however was fantastic, what I'm saying is ONE weak link in a fantastic series is no reason to doubt the abilities of the latest release been fantastic.

Mooks

06-04-2006, 20:48

If mtw2 is as retarded as rtw turned out to be. Ill permantly turn away from the TW games.

If people want a dumbed down game then they should play on easy...thats why its there , for the game noobs who have practically no micro-management skills (like my little brother for instance). I always play on vh/vh, and in rtw it was never much of a challenge (mtw did present some challenges).

Lentonius

06-04-2006, 20:54

Well, for a game producer, to be honest they care about sales more than anything, i mean, who doesnt?

So, logically, for them, CA are gonna present beautiful pics rather than talking about the AI, more people will buy a beautiful looking game than one for its difficulty. The visuals are what lure the buyer in, so thats what they are concentrating on.

Dooz

06-04-2006, 20:57

The ideal situation would be essentially two game modes, which are actually different, unlike in RTW; arcade and sim. Of course arcade being for those who, as Nagatsuka mentioned, shy away from "hard" games and prefer an RTW type experience. The sim mode would be for those of us who like more realistic games and such.

Failing that, I think at least the difficulty settings should do more than give the AI better moral and more attack and defense points. Sheesh. I imagine actual better AI could be incorporated into the different difficulty settings instead of arbitrary bonuses.

SpencerH

06-04-2006, 21:51

Movement speeds and such aren't a big deal at all as they are very easily changed.

They are not 'easily changed'. They're fixed by the animations (AFAIK). All that can be done is to modify the terrain effect to slow all units proportionally (rather than, for example, preventing the warbands from running for 2 miles at the speed of an olympic sprinter).

lars573

06-04-2006, 22:56

Most interesting fact here is, none of us know what CA are doing right now. I heard somebody was concerned with the time CA left to "dedicate to the AI". Put it this way, 4 months of 5/6 working days is PLENTY of time for a skilled group of programmers such as CA to get together and work hard on the AI and sort it out.

The changing of the AI I am pretty certain was to try and cater to the wider market CA were bringing in after TWs sucess, this wider market are the types of gamers who get annoyed when they don't win easily and will drop a game if they think its hard, or takes ages, hence the dumbing down of the AI and the speeding up of the units. I imagine, however, CA have surveyed the communities opinions of RTW and are probably looking at making a much finer balance between the two markets who buy this game.
The guy at the .com who was at CA's stall in the Sega section of E3 said that CA has 2 guys working on the AI, 1 for campagin map AI the other for battlefield AI. But to be fair the kind of coding work that making AI is would really only be a 1 man job. But if AI is all they are both doing then that's several hundred hours for the development of both.

Callahan9119

06-05-2006, 00:48

They are not 'easily changed'. They're fixed by the animations (AFAIK). All that can be done is to modify the terrain effect to slow all units proportionally (rather than, for example, preventing the warbands from running for 2 miles at the speed of an olympic sprinter).

yes you can only modify the animations, like you said by changing the terrain values, and if its like rtw it would be easy enough but would probably make the treadmill effect even more noticable due to the improved graphics and animations<if its done the same way>...like i said before its hard to import this to the MP arena as there is hardly ever agreement over which mod is best ~:(

yet, really all i want is the difficulty to translate better, like Wonderland said... the ai acts pretty much the same on all levels....upping the difficulty seems to only pad the AI's stats, let the nubsauce players use the easy difficulty

i just get annoyed that they seem to be disregarding and neglecting the player base that has supported them over the years, we all put up with the GOD-AWFUL multiplyer/gamespy issues...and franky i didnt mind the ctd, unable to connect, games dropping to the lobby mid fight and the horrible performance "TTTTTTTTT"....cuz the game was so damn good when u could actually play, especially cuz of all u classy euros ~:wave:

i dont think RTW was a bad game either...just the mp...i modded the unit animations to slow em down and added morale values to all the units and BAM the game is harder...i dont mind them taking creative liberty with history and the goofy head hurlers and roman ninjas...i kinda like them :hide:

IceTorque

06-05-2006, 01:14

With the berserker type AI from BI. ( fire and forget ) The new feature of units gaining or losing momentum while in melee. ( less predictable out come of battles ) Flanks gaurded by terrain, battle map features. ( much harder to flank ) Not to mention, made in Australia.
One thing is for sure. That is we are all goin to be in for a nice surprise, battles and the battle AI will be like nothing ever seen before in a TW game.

-IceTorque

Dooz

06-05-2006, 04:13

I gotta say... I'm feeling optimistic about M2TW. According to them, it's been in the works for 7 years now... meaning RTW was kind of like a "tune-up" for M2TW, a beta in a different historical setting. It sure felt like that. Maybe it's just wishful thinking, but I do have a feeling M2TW will be a lot better, a lot better than RTW.

Then, logic kicks in - marketing, mainstream consumer wants, money, whole lotta BS, and boom, right back to realistic pessimism. Ah, if only I could be the optimistic unrealist.

Oaty

06-05-2006, 07:46

f

and i dont see why it would be so hard to make a better ai, how about one that doesnt move around behind walls eating arrows, one that stays in formation, doesnt send generals to a quick death, doesnt send slingers charging into my front lines as shock troops, builds better units for its armies and knows how to tech up its buildings :idea2:

That paragraph sums it up where the AI needs the most improvement. To top it off the A.I. can have superior firepower behind thier walls and as soon as you counter-fire, the A.I.'s missile troops will start running around. If they had just stood there returning fire the effect would be devastating.

Anyways all games that I have played always had a loophole somewhere in the A.I. The main concern is to fix the worst of the problems, where the A.I. just sends lambs to the slaughter.

Ptogramming an A.I. to a human level is no way near the future.

screwtype

06-05-2006, 09:30

What I'd like to see most is AI armies acting in a co-ordinated manner instead of each unit acting completely independently of the rest. It makes it so easy to beat the AI piecemeal. And there's no sense of a controlling enemy intelligence at all. RTW AI armies basically behaved in a totally random fashion.

All they would really need to do is to program the AI to have half a dozen different battle plans for each of the main situations, ie field attack, field defence, siege attack, siege defence etc., and you would have a game that was entertaining, unpredictable, and challenging. But what are chances we will get something like this?

As far as speculations about M2TW AI, I was initially quite hopeful we were going to see significant improvement in this game. However, the movies we have seen so far are not terribly encouraging. Units still seem to be moving too fast, kill rates still too fast, not much evidence of AI smarts. So I'm beginning to grow cynical again and to think we are going to get RTW with prettier graphics.

However, I'm trying to suspend judgement until I've played the demo.

Mooks

06-05-2006, 12:49

However, I'm trying to suspend judgement until I've played the demo.

Most likely the demo will be dumbed down for nooby players.

screwtype

06-05-2006, 12:59

Yeah, I remember people saying the same thing about the RTW demo as well. They turned out to be wrong.

Apart from the fact that the RTW demo was scripted, it gave a pretty good indication of what the actual game was like. I expect it will be much the same for M2TW.

But I probably won't just rely on the demo. I'll wait to see what the consensus about the game is from experienced TW'ers.

Rodion Romanovich

06-05-2006, 14:30

the graphics are pretty, they love to point this out....but what about the AI?

6 months till release<rumor> why no talk of the AI? :furious3:

getting annoyed with them talking up the graphics and animations and nothing about improved AI <which is really what any game is about> :wall:
It's very difficult to program much of the AI before finishing most of the graphics engine, so it's natural that graphics are presented first. I'd guess they've made most of the AI by now, but it's probably not yet in a finished state. As it's the last thing done before testing and tweaking, it'll probably be finished around 3 months before release or something like that.

Callahan9119

06-06-2006, 07:22

It's very difficult to program much of the AI before finishing most of the graphics engine, so it's natural that graphics are presented first. I'd guess they've made most of the AI by now, but it's probably not yet in a finished state. As it's the last thing done before testing and tweaking, it'll probably be finished around 3 months before release or something like that.

good argument...unit stats etc etc must be tested and such, but couldnt ai be developed fairly independently? i just feel like it should be something constantly worked on from concept to actual design...ai has become a major focus point for game designers, look at for example many fps games when stw launched now we have halo...and many other games that have ai miles away from 4 years ago

yet this type of game does not have the market fps have...so i have and will cut them some slack...i have a powerful system i want the best

but you do have a good point regardless

EDIT

and before anybody crucifies me, saying its easier to do ai for a fps: as TW is dealing with thousands of units....its just really 10 as its just a big blob of animations...not that its easy to do this :)

Rodion Romanovich

06-06-2006, 09:07

good argument...unit stats etc etc must be tested and such, 1. but couldnt ai be developed fairly independently? i just feel like 2. it should be something constantly worked on from concept to actual design...ai has become a major focus point for game designers, look at for example many fps games when stw launched now we have halo...and many other games that have ai miles away from 4 years ago

yet this type of game does not have the market fps have...so i have and will cut them some slack...i have a powerful system i want the best

but you do have a good point regardless

EDIT

and before anybody crucifies me, saying its easier to do ai for a fps: as TW is dealing with thousands of units....its just really 10 as its just a big blob of animations...not that its easy to do this :)

1. after graphics have been planned and implemented enough it's possible to implement AI. But sometimes when graphics are extended it might affect ai. Probably they could start working on AI after getting the basics of the grahpics engine ready. After all it's necessary to be able to preview how the AI looks in practise to be able to program much of it. But yes, it's possible to do it independently to some extent, but preferable for the developers to do after all graphics are done
2. yes, if the developer has made the same type of game several times, they can learn what dependencies there are between graphics engine and AI, i.e. what organization they need for the data from a fast graphics perspective. Due to the high framerate requirements it's often necessary to adapt data structures to graphics. Therefore if graphics are upgraded a lot, like from RTW to MTW2, there might still be problems figuring out AI before making most graphics features.

sunsmountain

06-06-2006, 09:14

The guy at the .com who was at CA's stall in the Sega section of E3 said that CA has 2 guys working on the AI, 1 for campagin map AI the other for battlefield AI. But to be fair the kind of coding work that making AI is would really only be a 1 man job. But if AI is all they are both doing then that's several hundred hours for the development of both.
I have a feeling who this 1 man is, and though I trust in his abilities, I would appreciate it if he shared his design problems with the STW/MTW veterans, to see if they can find solutions. They can't be purely programming problems, ie there must be some puzzles to which we can perhaps supply answers.

What I'd like to see most is AI armies acting in a co-ordinated manner instead of each unit acting completely independently of the rest. It makes it so easy to beat the AI piecemeal. And there's no sense of a controlling enemy intelligence at all. RTW AI armies basically behaved in a totally random fashion.
No, not random. And the AI armies do tend to stay together before engaging. Not talking about the three way battles of course, or players deliberately tempting the AI army to split up (which is of course, only natural). On a unit level, the unit AI seems dominant. What it needs is indeed higher level command, that can synthesize ALL situations and provide answers in ALL situations.

Though the AI was designed from the ground up to allow such higher level decision making, the time needed for tweaking this was no longer there, and patches have already begun as far as i can see to add what i call "formations before engagement".

I remember everything that has ever been said about the AI because CA isn't going into details about it (copyright and stuff), but I don't have the links anymore to prove what i've said. Safe to say that there's a lot more to the AI than you may think. A game has to be finished, anyway, before any AI can be written. That kind of puts a stop on new ideas (and Thank God).

All they would really need to do is to program the AI to have half a dozen different battle plans for each of the main situations, ie field attack, field defence, siege attack, siege defence etc., and you would have a game that was entertaining, unpredictable, and challenging. But what are chances we will get something like this?
I suggested as much with a combination of MTW and RTW AI, giving initial starting formations to the AI, and different strategies for each one. But in RTW, if you give the AI a certain mix of units, it will always apply the same initial formation & strategy, no matter what. That has to change, well if i am to buy it anyway.

As far as speculations about M2TW AI, I was initially quite hopeful we were going to see significant improvement in this game. However, the movies we have seen so far are not terribly encouraging. Units still seem to be moving too fast, kill rates still too fast, not much evidence of AI smarts. So I'm beginning to grow cynical again and to think we are going to get RTW with prettier graphics.

Yeah I just hope they turn down the kill speeds so we get to enjoy the effects more of what we do. This is the main point for me to buy MTW2 or not. Followed quickly by a cunning AI (better than RTW, although RTW is not THAT bad).

Callahan9119

06-06-2006, 09:16

sunsmountain

06-06-2006, 16:27

Wait a minute, Rome had 11 programmers. Surely more than 1 could do & help with battlemap AI??
http://www.gamespot.com/pc/strategy/rometotalwar/tech_info.html

Odin

06-06-2006, 19:34

ok i see your point, yet i still want to hear SOMETHING regarding AI concerns that have been around since stw....anything :help: they just keep upgrading the graphics...the mori <sp> clan randomly breaking my alliance was no different than when any other stupid rtw faction does it...same with units eating missles etc....all i want is them to throw me a frickin bone~:pissed:

While I understand your frustration level, and I know its shared by many, CA has a well established model for releases. Since STW, I dont recall the AI as being the seller of any of thier games.

The graphics seem to drive the initial release/sale process and then, after the feedback from the masses a patch or two gets released, then an expansion with a last patch.

I dont forsee this changing, as its seemingly a successful formula for this developer, otherwise sales wouldnt justify the continuation of the series. You're best bet might be to hold off on preorder or purchase at release and let it hit the masses first, maybe wait for a patch to come out before leaping in.

Furious Mental

06-07-2006, 06:58

"I dont forsee this changing, as its seemingly a successful formula for this developer, otherwise sales wouldnt justify the continuation of the series. You're best bet might be to hold off on preorder or purchase at release and let it hit the masses first, maybe wait for a patch to come out before leaping in."

I understand this, but there is a way around it and personally I really wish CA would make use of it. All they have to do is, once a game is released, licence the engine to other developers who can then work on making games with a new setting, whether it is ultra-accurate history or over-the-top fantasy, or on making powerful battlefield AI, or any other things that CA has not enough time for. People who want the most graphically dazzling TW game could get whatever CA makes itself, those who are prepared to accept less advanced graphics but with a setting that they like or with better AI or whatever could get second-party games. And I think alot of people would get both- any Total War is best seller.

screwtype

06-07-2006, 08:43

One thing I wish and that is that game code would go public after a certain time. There are so many great old games, often from defunct companies, that could just do with a little tweak here and there to bring them up-to-date and make into perennial all-time classics.

It's such a shame to see all this code just go to waste. I wish somebody would pass a law to say that gaming code has to be preserved and released to the public domain after a certain time period, say, five years. Then all those brilliant old games could be revived and built upon by entreprising programmers.

In fact, I think I'm going to start a political movement with this as my central platform ~:)

Furious Mental

06-07-2006, 12:22

I'd vote for that.

Mooks

06-07-2006, 12:39

Same here.

Odin

06-07-2006, 13:03

I understand this, but there is a way around it and personally I really wish CA would make use of it. All they have to do is, once a game is released, licence the engine to other developers who can then work on making games with a new setting, whether it is ultra-accurate history or over-the-top fantasy, or on making powerful battlefield AI, or any other things that CA has not enough time for. People who want the most graphically dazzling TW game could get whatever CA makes itself, those who are prepared to accept less advanced graphics but with a setting that they like or with better AI or whatever could get second-party games. And I think alot of people would get both- any Total War is best seller.

In sentiment, I agree with you. In practice though, there is no need to get around anything. CA seems to be doing just fine with the current business model. What is there incentive to change their current practices? Like it or not the TW series has been a commercial success, and CA is a business.

Ideally it would be a more open source coding so the moding would be easier (AKA Paradox titles) this way, end users like you and I can actually tweek the AI and come up with mods free to the community. I dont know if CA would license the engine, an intriquing idea but not something I have heard a lot about.

Is there precident for this in the gaming industry? I honestly dont know, but again, reality suggests that CA has a proven business model going on the 4th title, I dont see it changing. The only thing that is a variable at this point, is the end user purchase.

If that changes, then thier development process will change.

screwtype

06-07-2006, 13:14

Same here.

Maybe we should all start a gamer's lobby group or something ~:)

sunsmountain

06-07-2006, 13:17

Already now, when 5 years have passed, most companies no longer pursue copyright infringement actively for certain titles. Reverse engineering them to create new titles however, always gets a response, especially if you stand to gain money from it.

While money is an important aspect, you can not totally alienate your original fanbase. You could, but some of us would eventually post a negative comment on Gamespot, hurting their sales & image.

On a human level, the programmers can identify themselves more with us than they can with WCIII players, though they do also have to cater to those or face their investors. They have Quality Assessment Groups, in which they test all new ideas on average gamers, and bug testing is thorough. Though we may not like Rome:TW, all intended ideas were implemented (not necessarily balanced) and the game is relatively crash free. I have yet to crash version 1.5...

The AI remains a tricky point. It cannot be marketed, so we hear nothing about it. Who wants to be smart instead of pretty? The latter sure sells...
But I am confident they will take their time this time, and if they don't, our judgement will be accordingly.

If you want to help, become a playtester (beta release) for Activision, and submit feedback on the AI, IF you get to play beta MTW2 (which isn't a sure thing).

TB666

06-07-2006, 14:12

If you want to help, become a playtester (beta release) for Activision, and submit feedback on the AI, IF you get to play beta MTW2 (which isn't a sure thing).
That wouldn't help since CA are with SEGA now :laugh4:

sunsmountain

06-07-2006, 14:25

oh yes. :stupido2: So help SEGA then. Do they even have playtesters? Oh dear...

Dead Knight of the Living

06-07-2006, 14:52

Forget the AI. WHy aren't they talking about what advertisement flyers they're going to put in the box with the game.:laugh4:

Furious Mental

06-08-2006, 02:50

"In practice though, there is no need to get around anything. CA seems to be doing just fine with the current business model."

I was referring to the fact that the reason invariably given for the underdone AI and bugs is that they have know no time therefore no one can make a Total War game with computer players who were not dropped on their heads as babies. That is what they are supposed to be "getting around". As far as incentives go, well I would have thought the opportunity to make even more money licencing the engine to other studios would be an ample incentive. In the FPS development business it is commonplace. Of course it's entirely up to them. But the fact is that their engine will not maintain a virtual monopoly in this field forever, unless they licence it. If they don't eventually someone will develop a credible rival.

Odin

06-08-2006, 15:54

I was referring to the fact that the reason invariably given for the underdone AI and bugs is that they have know no time therefore no one can make a Total War game with computer players who were not dropped on their heads as babies.

True enough, but after 3 titles, and expansions players keep purchasing the product, underdone AI and all. While the AI might not be up to par with our expectations, clearly, someone at CA (or whomever makes the publishing decisions) feels its adequate enough to sell.

As far as incentives go, well I would have thought the opportunity to make even more money licencing the engine to other studios would be an ample incentive.

I honestly dont know the numbers, however licencing the engine would allow someone else to develop the product line, maybe better then you. So unless you get a new or improved engine you have effectively killed that product line development for yourself. Unless that agreement cover the original developer, like I said I dont know.

Of course it's entirely up to them. But the fact is that their engine will not maintain a virtual monopoly in this field forever, unless they licence it. If they don't eventually someone will develop a credible rival

Well that might be true, but here we are years later after STW and you would be hard pressed to find a "credible rival". If you know one please post a link, I would love to check out a new game. Im not suggesting CA has cornered the market, but they consistantly have used the same model of development, release, patch, expansion with a patch, new title.

Seems to be working for them, and while I want an AI thats top notch as much as the next person, my faith in gaming companies is down the toilet. I'll wait on the sidelines see how the response is from the community, wait for a patch or two then make a purchase.

sunsmountain

06-08-2006, 16:56

I honestly dont know the numbers, however licencing the engine would allow someone else to develop the product line, maybe better then you. So unless you get a new or improved engine you have effectively killed that product line development for yourself. Unless that agreement cover the original developer, like I said I dont know.

Sales of all TW products fall on or below the 1 million copies sold line, which does not consist of enough market for other companies to buy a license from SEGA, who would have to spend money first licensing it.

My estimate, for total sales:

Shogun 200,000 sold
STW mongol invasion 100,000 sold

Medieval 300,000 sold
MTW viking invasion 150,000 sold

Rome 1,000,000 sold (eventually)
RTW barbarian invasion 300,000 sold (eventually)

These are quite optimistic numbers, I might add. Of those 700,000 new members to the TW community, only 1 in 10 registers with the fora, while almost 50% of the STW/MTW group is registerd. These are again my own personal estimates. A lot of the sales for Rome:TW have nothing to do with AI, but simply a sound commercial marketing strategy, and the fact that it's unique amongst strategy games.

doc_bean

06-08-2006, 17:23

My estimate, for total sales:

It's nearly impossible to make an estimate of sales in my experience. They're bound to float around the web somewhere, if only we oculd find them (CA enlighten us !)

Callahan9119

06-09-2006, 05:10

improving the ai should be an obligation over 6 + releases and expansions

what has really changed?...where is glorious achievments? seems we lose something everygame with minimal improvements besides the graphics engine. the more they dummy the game the more they throw themselves to an area where companies already have the market cornered. it might work, seems risky to me though. u know they wont be going anywhere in the mp department with gamespy as is :laugh4:

not many buy a game for the singleplayer experience if its not challenging and deep

and the clickfest rts mp gamers will laugh and laugh when CA tries to market to them with its awful 4 year record of MP, but maybe they can actually trick some into buying it....we'll see. Last night was in MP for 5 hours, actually managed 4 games 15 failed games <not from my end>:balloon2:

rome was as talked up as this, and the ai is worse than it was in mtw and they did nothing to the mp stability <if anything its worse> why will this be any different

Furious Mental

06-09-2006, 09:36

Odin

06-09-2006, 13:55

"So unless you get a new or improved engine you have effectively killed that product line development for yourself."

Did you actually read what I typed? I said they should licence previous iterations of an engine after they have developed an improved one, e.g.licence the RTW engine sometimes between the release of RTW and MTW

How is it a rival product line if they're paying to use the engine? The terms of a contract can easily be written to preclude certain changes being made to an engine, such that if CA wants to ensure that its own games are always the most impressive in terms of graphics, numbers of troops etcetera all it has to is stick a stipulation in the agreement. If you actually look at the "competition" to Total War games it's pretty clear that it consists chiefly of developers and a market which either doesn't care for games on such a large and realistically rendered (in terms of graphics) scale or hasn't the skill to make them and would be interested chiefly in changing the setting and gameplay. Thus I find the idea that they could or would use a licenced Total War game to make a clone somewhat dubious. If/ when some studio pops up which has the skill and intention to create a genuine Total War rival themselves then they would make such a game whether they can use the Total War engine or not.

Yeah I read what you typed.

All they have to do is, once a game is released, licence the engine to other developers who can then work on making games with a new setting, whether it is ultra-accurate history or over-the-top fantasy, or on making powerful battlefield AI, or any other things that CA has not enough time for. People who want the most graphically dazzling TW game could get whatever CA makes itself, those who are prepared to accept less advanced graphics but with a setting that they like or with better AI or whatever could get second-party games. And I think alot of people would get both- any Total War is best seller.

Thats what you typed. No where did you mention:

I said they should licence previous iterations of an engine after they have developed an improved one, e.g.licence the RTW engine sometimes between the release of RTW and MTW

You state they could license the engine once its released, not after a new one is developed, if you meant to say that fine, but yeah pal I "actually read what you typed" maybe if you were a little more clear on what you meant it might make your point quicker and you wouldnt have to take 3 posts to do it.

How is it a rival product line if they're paying to use the engine?

The fact that someone else is now devloping your engine, allows them to take it any direction they want. Thus direct competition with the liscensee unless they develop a new engine and liscense out the old one. So you develop a new total war engine, and you allow someone else to use the old one, well what are they going to use it for? To develop games?

Why the hell would anyone give someone else the means to compete directly against them with a product they developed, even if its an older version? Maybe it happens in some FPS or other joundra (sp?) of gaming but I havent seen it much in wargames that I have played.

In addition to that, I am sure there are people in the industry, and at CA who are far more intelligent about the gaming industry then you and I, and they havent done it yet. Why? My contention is very simple, there is no need to, gamers keep buying the product, no need to change the current formula based on a few hundred complaints about AI, of which about half of those people will buy the game anyway.

econ21

06-09-2006, 14:06

Odin, please let's keep exchanges in this forum friendly. ~:grouphug:

Byzantine Mercenary

06-09-2006, 14:13

Lord Adherbal

06-09-2006, 14:47

with the dramatically reduced audience for such a game

my goodness, is it THAT bad with the current youth?

I can't remember the term "historical accuracy" ever scaring me off

Furious Mental

06-09-2006, 16:28

"Thus direct competition with the liscensee unless they develop a new engine and liscense out the old one. So you develop a new total war engine, and you allow someone else to use the old one, well what are they going to use it for? To develop games?"

Umm yeah. Are you going to tell me how that is dangerous competition any more than the existence of other Total War- like games is? It is in fact a way to make money from competition.

"I havent seen it much in wargames that I have played."

And? It is done. I can think of a few licenced engines off my head- AOE II, WoW, CnC, etc. Besides which the fact that it is more common in FPS games is essentially immaterial. Your reasoning seems to be "they don't do it therefore they shouldn't do it". Interesting.

"no need to change the current formula based on a few hundred complaints about AI, of which about half of those people will buy the game anyway."

First of all, I said, I think, more than twice even, that it is not just about the AI. Second of all, it's also not about "changing the formula" or "need". It is simply an obvious way to make cash by capitalising on the existence of old engines. It is hardly the only way to do it either. Another would be to expand the possibilities for modders and subcontract prominent mods to make retail add-ons, which has proved to be immensely successful in FPS games. The very fact that it has been under-exploited in the strategy genre means that it is big fat opportunity screaming to be seized. Yet another would be to expand the number of ancillary studios which are responsible for expansions, which eliminates the issue of competition completely.

Mooks

06-09-2006, 21:37

Im 14 and I read books in class. Im not considered the most social person but im not a nerd. And well, most of the people in my grade say books are for idiots and "who cares about history". My english teacher told the class about the romans and it was genuinely their first time ever hearing about them. Hell, 90% of them have NEVER read a book just for fun.

I think the problem is, is that people dont know history. They cant really tell if a game is historically accurate or fiction. They dont know that cannons were too heavy to mount on elephants or where russia is (seriosly, alot of people in my classes cant even find britian on a map, I doubt they could find russia if they tried) and who the mongols were and what they did. If someone says its accurate, then they will beleive them.

:dizzy2:

Gaulgath

06-10-2006, 02:01

Adherbal']they were also big for RTW, and look what we got - certainly not something the overall TW vet appreciated

From the first things I heard and saw of MTW2 I had a good feeling about the game, but after noticing they didn't even fix those rediculous run speeds I just lost all hope. This will be "RTW: The Middle Ages", not MTW with awesome graphics.
If there is one thing I hate, it's bashers. The game hasn't even come out yet, and everybody is butchering the game! Why don't I sit you down in a chair and watch you program this? TW is MASSIVE, and not every little detail is going to be perfect. I don't understand what is so hard to grasp about that. :furious3:

Cowhead418

06-10-2006, 02:52

If there is one thing I hate, it's bashers. The game hasn't even come out yet, and everybody is butchering the game! Why don't I sit you down in a chair and watch you program this? TW is MASSIVE, and not every little detail is going to be perfect. I don't understand what is so hard to grasp about that. :furious3:
THANK YOU. I can understand wanting the game to be as historically accurate as possible, but all the nitpicking that goes in is absolutely ridiculous. Every time there is a picture of a new unit, I read a post that complains about the color of a patch of clothing or something else that shouldn't really matter. These are little details and it should be OK if all of them aren't 100% accurate.

As for the AI, it isn't nearly as bad as everyone makes it out to be, especially in mods such as RTR in RTW and XL mod in MTW. It would be a hell of a lot more challenging if the AI would just stop sending their units to the slaughter, especially the general suicide charges. Sure, the army composition could use work as well as a lot of other less than stellar tactics the AI uses, but I think that if the AI would stop letting its units get isolated or if it would have its general live through the first clash of the battle, then there would be a lot less whining. Just my .02.

Callahan9119

06-10-2006, 05:44

If there is one thing I hate, it's bashers. The game hasn't even come out yet, and everybody is butchering the game! Why don't I sit you down in a chair and watch you program this? TW is MASSIVE, and not every little detail is going to be perfect. I don't understand what is so hard to grasp about that. :furious3:

how long have you played tw? the simple fact is they have not addressed problems and design flaws that have been around since the first release of STW. i also dont give a #$^* about historical accuracy, they do a good enough job of balance between whats real and what sells...the same AI from 4 years ago is a laugh...if anything it gets worse.

remember all the RTW hype? remember how the history channel used the rtw graphics engine when it produced a series of shows? remember how groundbreaking it was supposed to be? it wasnt, MTW had sooo much more depth...they fluffed up the graphics at the expense of everything else

if i was a person who started playing at rtw i could understand you being bright eyed and bushy tailed, those of us who have played this series for 4 years seem to be a bit jaded by ca's neglecting its fan base and its rude bahavior on the official forums

with every release and expansion i couldnt wait to get my hands on it, now this time its more "eh i guess i'll get it" its like dragonlance novels...didnt like the way it was going but i bought em anyway...and thats all they want you to do, buy it...who cares if you like it

maybe you like a shallow game thats just pretty to look at, if so i expect you will get just this....i could be wrong...but i wont hold my breath, even if you might want me to :)

Count Dracul

06-10-2006, 07:23

Well,
I hope they do reveal some details about the AI because this was one of the major problems in RTW: bad diplomacy system, bad campaign AI, horrible battlefield AI...
I have my doubts if they will look into it. The old fanbase that is asking for better AI is just a small consumergroup; That means we are of little importance for their sales...

But maybe there is hope. I've searched on the internet and discovered a game ' XIII Century: Sword & Honor' and it looks very promising. I have send a mail to the publisher to ask for more details and here is what they said about it:
Hello Pieter-

Sorry for the delay with the answer, but I was out of the office and unfortunately had no access to my e-mails.

I’ve asked the developers and they have provided me with the following information:

1. No. The campaigns in the game will be battle oriented. There will be only battles.

2. Battle AI is complicated and pretty hard to beat and we are currently working on it and making it possible for a human player to win a battle :) There is no diplomacy in the game except for the sword.

3. This game does not have any unit construction because we are oriented on historical accuracy and historically accurate battles. Castles are map objects and are not constructed during the game. But we plan to supply the game with an editor to allow players build their own battles and castles.

4. We plan to release the game in Q4 2006.

Sincerely yours,

Anatoly Subbotin

PR Manager

1C Company

suba@1c.ru, http://int.games.1c.ru/

I mean there is no campaign map but the AI sounds very promising and that is the most important part of the game for me. + maybe this could be a competitor for Activision in the future. I'm going to check this game out when it is released. Don't know about you?

Dooz

06-10-2006, 09:34

Might as well plug this game too.

http://www.madminutegames.com/

The absolute best battle AI currently available. Take Command 2nd Manassas is the newest game from the site (set during the American Civil War), you can check out the demo and be blown away. AI is great, battles are great. You just need to try it to believe it basically. Huge maps, huge armies, couriers relaying messages between commanders, tactical battles and strategies, man it's amazing. Oh and a great community to boot. This recent discovery has made me a hell of a happy gamer, with some new found optimism and encouragement from the gaming world.

-edit-
Oh sweet Moses, that game Count Dracul mentioned looks spectacular. Those screens especially tickled me in ways I ought not be touched...

Lord Adherbal

06-10-2006, 09:53

Why don't I sit you down in a chair and watch you program this? TW is MASSIVE, and not every little detail is going to be perfect.

if the problems with RTW were caused by programming difficulties then they should fire a lot of people and get the old STW/MTW programmers back.
But no, it's the new design approach that alienates the old fanbase.

And what do you want us to do, whine AFTER the game is released, when the 1% chance we have in having an influence on the gameplay has turned into 0% ?

Count Dracul

06-10-2006, 12:51

If forgot to post the link to some screenshots of XIII Century: Sword & Honor

http://www.3dgamers.com/screenshots/games/xiiicentury/

I have also seen some screenshots where the lances (carried under the armpit: correct way) actually break off after a charge and they change to sword. But unfortunately I don't remember where I've seen it... If you look good It can be seen in this pic

http://unicorn-games.com/screenshots/original/cav.fight.6.0-04-14.392.jpg

doc_bean

06-10-2006, 13:21

That is what they are supposed to be "getting around". As far as incentives go, well I would have thought the opportunity to make even more money licencing the engine to other studios would be an ample incentive. In the FPS development business it is commonplace. Of course it's entirely up to them. But the fact is that their engine will not maintain a virtual monopoly in this field forever, unless they licence it. If they don't eventually someone will develop a credible rival.

Hmm, i seem to have missed this discussion last time I checked the thread :dizzy2:

So far CA is the ONLY company that has access to an engine like Rome's. In the fps scene there have been rival engines from the start, sure most weren't as impressive as ID's, but there was always a competition. Even so licenses don't happen that easily, ID almost always licenses to raven first, and probably makes a whole lot of money from those games to. Every company seems to make sure that no other games with 'their' tech are released around the same time. VTM: Bloodlines might be the big exception but then it was an entirely different game.
Which brings us to another point: fps engines are (considered) more versatile than an engine like the TW one, they make shooters RPGs, third person action games, MMO games and even strategy games with those engines. A TW engine is more limited in scope, although it could probably be expanded upon more to make it more versatile, that would require a serious investment however. The main point of the tw engine is that it can draw lots of people on screen, that isn't always what's needed in other games. CA tried to see what their tech could in other genres with Spartan i think (was it the rome engine ?) but that wasn't the most (commercially) successful attempt afaik.
These days it's all about making pretty close up screen shots, and the TW engine (certainly before MTW2) wasn't really up to that.

So in conclusion: they have a monopoly in a niche market right now, no need to share.

econ21

06-10-2006, 14:11

Dooz

06-10-2006, 22:40

Just a note: mentioning other games in passing is ok, but extended discussion belongs in the Arena. There's already a fair bit there on Take Command 2nd Manassas recently. Not sure about XIII Century: Sword & Honour.

Sorry econ21, I knew it might be pushing it, but considering the nature of this thread and the people who post in it I really thought it'd be something of great value. Plus, I guess I'm still way too over excited about having discovered the damn game. I shall subdue my urges to introduce war gaming pleasure to the desiring masses. :bow:

econ21

06-10-2006, 23:02

No worries, you make a relevant point - to identify other games that have realistic and/or challenging battlefield AI.

I haven't tried Take Command 2nd Manassa myself. Most computer wargames I've played (1990s vintage) did not have noticeably better AI than TW games. An exception may be Sid Meier's Gettysburg. But that game lacked the excitement of Total War battles, as well the strategic layer, and so never grabbed me.

I'm just saying the Arena exists if people want to discuss other titles in detail.

Ignoramus

06-11-2006, 08:04

I doubt that we can expect a better AI for Medieval Total War II. If they are using Rome's engine, then I would say that they can only do as much as the Mods have done.

Callahan9119

06-11-2006, 09:45

its not so difficult, we have amazing recourses with todays systems. its about priority, mtw had a far better AI than rtw, to say it cant be done is nonsense, it was better and it could be amazing if they actually made it anything but an afterthought...spare me the stuff about how hard it is and precedent...ca has failed the LOYAL fanbase that has supported it it for 4 years, it has forsaken it to sales....but hey thats capitalism :skull:

rtw is fun and i'm sure mtw will be too, but they have regressed instead of looking forward and separating themselves from the rest of games, conformity over progress, any dope company can make a pretty game

Ludens

06-11-2006, 10:41

ca has failed the LOYAL fanbase that has supported it it for 4 years, it has forsaken it to sales....but hey thats capitalism :skull:

rtw is fun and i'm sure mtw will be too, but they have regressed instead of looking forward and separating themselves from the rest of games, conformity over progress, any dope company can make a pretty game
Though I share your dissappointment over R:TW, I think you are being unfair to CA here. They did make an effort to improve things. They released five patches which all improved the A.I. (except for 1.1) and, what is more, they kept R:TW supported after they released BI. This is more than they did for either M:TW or S:TW. Failed? Perhaps, but not forsaken.

sunsmountain

06-12-2006, 10:01

We should indeed give the CA a little more credit. It just turns out that the AI is the heart of a game. Making a game beautiful, and giving it 20 extra dimensions and ideas, also complicates the game. So writing a good AI will take even more time than before.

Time is something you have lots of when you are unknown & small, like before Shogun:TW. But then all of sudden, the marketing people start barging in, the managers want to have a say and make sure this product is sold in the way they want to. These people don't listen to "what about the AI?". These people don't care. I doubt they even know what a game is. The CA does know what a game is, but it's hard to keep everybody pleased.

I just hope the people of the CA have the strength to ignore marketing goals and simply focus on what's right: making a good and intelligent game.

Callahan9119

06-12-2006, 10:24

look, i understand it isnt easy to make a game, but the ai needs to build up and get good units...facing scrubs every battle gets old, secondly it needs to form a cohesive battle line...it has been sending units 1 at a time to die since stw...mtw was better slightly. maybe a prioritized unit selection for ranged to fucus fire on units

and on a side note dispose of gamespy, its horrible and has been for 4 years, we deserve better than the mp we have

Puzz3D

06-12-2006, 12:34

I just hope the people of the CA have the strength to ignore marketing goals and simply focus on what's right: making a good and intelligent game.
That group of people has already lost. They didn't have the strength to resist the gameplay damaging ideas.

sunsmountain

06-12-2006, 12:57

That group of people has already lost. They didn't have the strength to resist the gameplay damaging ideas.

That was because they were new. They're not new this time, in fact a lot less "New Ideas" have to be implemented in MTW2, leaving more time for AI. Remember how CA treated new ideas? They wined them, dined them, took them back home and then jumped them? Remember those developer diaries? It's not as black & white as you think. They simply may not yet have had the time to do things you would like them to. They sure can. They hooked you to Shogun, didn't they? They can hook you to MTW2.

Odin

06-12-2006, 13:37

the fact that it is more common in FPS games is essentially immaterial. Your reasoning seems to be "they don't do it therefore they shouldn't do it". Interesting.

Its immaterial? Wasnt that the point of reference you made initially as a justification for lisceneing the engine, and further in this post when you state its been
immensely succesful in FPS games? Your talking in circles then.

My justification is rather simple, why would they change something if they dont need to? I've heard all the arguments for better AI a thousand times on multiple games, but it rarely gets addressed, and why is that? Because gamers still buy the games. While your ideas have merit and seem logical, there is no expliantion as to why they would do it.

"no need to change the current formula based on a few hundred complaints about AI, of which about half of those people will buy the game anyway." First of all, I said, I think, more than twice even, that it is not just about the AI. Second of all, it's also not about "changing the formula" or "need". It is simply an obvious way to make cash by capitalising on the existence of old engines.

What old engine? Please be specific, because it seems to me your talking about a what if scenario. And again, why give technology to someone else to develop? You think its lucrative? Well okay, maybe it is, I dont know but other then the FPS games (which you reference, but claim is immaterial) I havent seen a precendent for this. I own an avarage amount of strat/wargames and have had an avarage amount of expirence with them and I havent seen many liscenses go out to others, if you have, then I defer to your expirence.

Another would be to expand the possibilities for modders and subcontract prominent mods to make retail add-ons, which has proved to be immensely successful in FPS games. The very fact that it has been under-exploited in the strategy genre means that it is big fat opportunity screaming to be seized. Yet another would be to expand the number of ancillary studios which are responsible for expansions, which eliminates the issue of competition completely

Perhaps your right, I honestly dont know, however there is one very simple and direct way to inpact the development of games. That is the power of consumption, if CA thought that thier AI, graphics, or whatever were not good enough for the consumer they would change the formula.

I have seen absolutely no evidence to suggest that is the case. Gamers will buy this title and justify its development with thier purchase. No matter how its developed, or by whom, if it sells these companies will put it on the shelf.

Suppose for a minute MTW2 does crap for sales, what do you think will happen? The process by which the game was developed will be reviewed and enhanced. Who knows, maybe you're prediction of ancillary studios will be the solution, but up till now, that isnt a reality because MTW2 looks to be on course like every other game in the total war series.

sunsmountain

06-12-2006, 13:59

Who knows, maybe you're prediction of ancillary studios will be the solution, but up till now, that isnt a reality because MTW2 looks to be on course like every other game in the total war series.

We could of course, under a democracy, start a smear campaign against SEGA/CA demanding better AI and drop our support for the game if they don't. But it's easier to give them one more chance... I didn't buy Shogun because i had Medieval. I probably won't buy MTW2 because i have Rome... if the engine doesn't change, why buy it? No copies in my collection...

ie if the AI remains the same there is not a single reason for me to buy MTW2, and it's cheaper for me to download a mod for Rome and play it.

Odin

06-12-2006, 14:54

We could of course, under a democracy, start a smear campaign against SEGA/CA demanding better AI and drop our support for the game if they don't. But it's easier to give them one more chance... I didn't buy Shogun because i had Medieval. I probably won't buy MTW2 because i have Rome... if the engine doesn't change, why buy it? No copies in my collection...

ie if the AI remains the same there is not a single reason for me to buy MTW2, and it's cheaper for me to download a mod for Rome and play it.

Its been my expirence that the best way to make a good game purchase these days is to wait for the game to be out, the masses report bugs, AI inefficencies. Then comes the enevitable patch or two, a price reduction then you buy it.

I would love nothing more then to support gaming companies at release of a game, I know thats how they get the most value out of my purchase. Sadly the state of affairs for game releases is poor now, maybe technology has made it harder to make a solid game at release, I dont know.

I do know there is one absolute slam dunk way to get companies to change thier habits, and that is not purchasing thier products until it is to your liking.

I bought MTW right at release because STW was an excellent game (still is) MTW needed some work still, but I never bought RTW. I waited, read the boards and came to the conclusion it wasnt much better then what I already had.

Callahan9119

06-14-2006, 09:18

i like the concept, so i'm a fan, though i feel i have been let down. FOR SURE in the MP, the fact that it looks to be in line with previous mtw games is a blessing and a curse. still fairly original gameplay, yet for its updates and graphics it hasnt evolved besides minor failed design features.

i really liked glorious achievements, added a new aspect to the usual 'destroy all' campaign and the ai really needs to develop for any progress to really be mentioned, as in my opinion, its not at all uncommon to have beautiful games

and as ca grows more brainless corporate entities control it, thus the obvious dumbing down of the game.

sunsmountain

06-14-2006, 10:01

Well, I'll save my pitch forks for now, but when MTW2 is:
- more beautiful
- more nice ideas
- more concepts implemented

without
- a significantly better AI than RomeTW AI

Then my fury or anger will probably result in some negative comments which I will refresh during the first 3 months of sales. "To kick 'em in the nuts where it hurts". Har, har, You're warned, CA:furious3: :wall:

doc_bean

06-14-2006, 10:02

We could of course, under a democracy, start a smear campaign against SEGA/CA demanding better AI and drop our support for the game if they don't.

pff we would be a few hundred people, perhaps a thousand or so, but nothing that should register with CA, Rome was by far the best selling PC game when it was released. There were shortages where I live. A certain week it sold sold twice as much copies as the next best selling game in the UK iirc.

People complained about Rome, a lot. Here and on TWCenter en probably in a few other places, we'll see whether our complaints have been heard by CA. I don't think so, the best we can hope for is that reviewers come to our side, once a game loses critical acclaim, it can lose a lot of sales.

But then again, recent reviews about Rome (gold) all said what a great game it was...

Odin

06-14-2006, 13:47

I don't think so, the best we can hope for is that reviewers come to our side, once a game loses critical acclaim, it can lose a lot of sales.

This is a decent point and one I used to believe in firmly. However I have had expirences with some online gaming sites that do reviews for games and it is my opinion that there is a degree of an incestual relationship going on.

Reviewers normally work for sites or mags that review games that provide advertising capital for said publication. Here in lies the incest, you would be hard pressed to find a company who gives a negative review that gets income from the company producing the game.

I understand this, and its reasonable business, but, in my observations the gaming industry (reviewers included) have a vested intrest in plugging games and increased game sales. More sales of games equates to more readership for the reviewer, incestual if you ask me.

There are some decent sites out there that do reviews of games, but I have yet to read anything unflattering on any release in the total war series. Most of the reviews I saw on RTW were glowing, it wasnt until I read what end users like me had to say until I could form a much clearer picture of the game.

Puzz3D

06-14-2006, 16:17

Reviewers normally work for sites or mags that review games that provide advertising capital for said publication.
That's why the potential customer shouldn't bother reading those reviews or watch the History Channel's Decisive Battles or the Time Commanders TV programs. Even the videos and still pictures that the company releases are designed to conceal flaws in the game. RTW had a massive number of problems when released and yet got a 92% rating. The rating is meaningless.

econ21

06-14-2006, 16:57

That's why the potential customer shouldn't bother reading those reviews or watch the History Channel's Decisive Battles or the Time Commanders TV programs.

I agree the reviews did not catch stuff they should have done. But their tilt was not that of the grognard - I am not sure the reviewers care about the "too fast" complaint that seems one of the most common here. Plus, there was also the time factor - I thought the vanilla game was great on my first campaign. I'm not sure this is complaint is unique to TW though - reviews are generally useless (at least for top name games); reading dedicated forums often gives a much better idea of a games strengths and weaknesses.

I disagree about the TV programmes. I found them fun and they showed a lot of potential with the engine. At their best, the videos in the TV programmes showcased a combination of epic scale, great visuals, historical detail and dramatic action. I agree you might feel let down coming to the vanilla game after that. But things have picked up since then with BI and especially with mods RTR and EB. I personally am satisfied that the potential shown by the TV programmes has been realised. I'm playing a RTR Platinum PBM right now in the Throne Room and it represents one of the best historical gaming experiences the TW series has ever provided me. The WRE PBM we just finished was also excellent. And we haven't even started on EB yet.

sunsmountain

06-14-2006, 17:24

About those TV programmes, in particular Time Commanders seemed to be using an old (but pretty good) version of the AI while using the new battlemap engine. ... maybe it was using MTW AI !! ~:) what we wouldn't give to play with that, eh?
What also strikes me as strange is that the maps on TC seemed to be bigger than they are now, like they have been limited to allow playing on the low-end pc's as well. Not saying that is bad or anything, but combined with the faster battle speeds and the quicker morale breaking routing effect, the experience of Rome:TW was totally new, and not in any particular pleasant way for most MTW players.

Mods in particular address two things which enhance play: physical parameters, ie unit mass effecting charge effects, unit marching & charging speeds & morale, and the other is often modifying battle_formations.txt, to increase AI effectiveness in defining and forcing starting formations to be used in some way.

It's strange though, that almost all the MTW players need a mod before they can start to enjoy RTW. Especially if you prefer to play a game like it's intended to be, like me, ie vanilla.

ps.: I was wondering.. if i create custom battles that are exact copies of the Time Commander battles, and i share them online, am I infringing any copyright issues here? Not saying i have the time, but..

Odin

06-14-2006, 17:43

That's why the potential customer shouldn't bother reading those reviews or watch the History Channel's Decisive Battles or the Time Commanders TV programs. Even the videos and still pictures that the company releases are designed to conceal flaws in the game. RTW had a massive number of problems when released and yet got a 92% rating. The rating is meaningless.

To be honest, I dont think the rating is meanigless at all. On the contrary, it is indictive of the incest situation I spoke of. There in lies the problem for the gamer, this is an industry problem, not CA exclusively.

Your post is spot on, until the end. RTW had problems and a 92% rating is clear evidence that the reviewers (a part of the gaming industry) either didnt have the expertise to find the issues, or had another adjenda in not revealing them.

Eitherway the end user (you, me and every other gamer) is left to hope it will improve with the next release, or slog through message boards and provide examples of bugs for developers to correct.

And thus my initial point in this thread, its an ideal situation for a developer, why change the current process when units sell, reviewers give positive reviews, and endusers provide the specific details of what is wrong with the product.

Why would any company change that? Its ideal

Spino

06-14-2006, 18:42

Sales of all TW products fall on or below the 1 million copies sold line, which does not consist of enough market for other companies to buy a license from SEGA, who would have to spend money first licensing it.

My estimate, for total sales:

Shogun 200,000 sold
STW mongol invasion 100,000 sold

Medieval 300,000 sold
MTW viking invasion 150,000 sold

Rome 1,000,000 sold (eventually)
RTW barbarian invasion 300,000 sold (eventually)

These are quite optimistic numbers, I might add. Of those 700,000 new members to the TW community, only 1 in 10 registers with the fora, while almost 50% of the STW/MTW group is registerd. These are again my own personal estimates. A lot of the sales for Rome:TW have nothing to do with AI, but simply a sound commercial marketing strategy, and the fact that it's unique amongst strategy games.

Actually your sales estimates are quite pessimistic and are way off the mark. Some time ago one of the developers posted some rough sales figures for Shogun & Medieval. It turns out since its release Shogun & its expansion have sold well over 500,000 copies (the actual number may have been closer to 700-750,000 but my memory is hazy). However I distinctly remember this person mentioning that Medieval & its expansion pack went on to sell roughly one million plus units. I think it's safe to say Rome & its expansion pack exceeded this number in a much shorter period of time.

sunsmountain

06-14-2006, 18:50

Yes, Spino, you're probably right, the problem is that it's quite difficult to google that information back at this time. Since i haven't seen any TW game appearing in the 1 million plus list, i took that as a reference. I'm sure Rome:TW will break that mark, while MTW and MTW:VI may have come close to 1 million individually (can't really add them together, as the latter needs the first to play).

So while I believe your numbers, my order of magnitude estimate isn't so far off. A little pessimistic perhaps, but then again i consider 300,000 copies sold to be a lot already.

doc_bean

06-14-2006, 21:50

reviews are generally useless (at least for top name games); reading dedicated forums often gives a much better idea of a games strengths and weaknesses.

I disagree, reviews are usually good for games that last about 10-20 hours. reviewers have a tight schedule, and they don't have time to play one game that much. They can hardly assess the replay value. Much of the problems with RTW came only to light after a few months (the missile bug !). Rome made a damn good first impression, if I had to have rated the game based on initial experience I might have given it a 92% also.

Spino

06-14-2006, 22:18

I almost forgot to comment on the thread's main subject...

CA is talking about the AI, certainly moreso than they have in the past. Clearly this is in reaction to the countless complaints leveled at Rome and its lackluster AI. However keep in mind that CA's generic and pithy responses regarding Medieval 2's AI is pretty much standard in the industry as most developers & publishers treat the open discussion of AI as a taboo subject. I'm not exactly sure why developers shy away from the discussion of AI in their upcoming games. It might have something to do with the level of understanding of the average gamer regarding AI in general. Any dolt (or game reviewer) can and will drone on about pixel & vertex shaders, surface mapping, etc. after a quick read of a video card review and that basic knowledge of the relevant technology translates into expectations which developers know their game must match or beat. Understanding the conventions, principles and jargon involved in AI programming is no simple task and certainly cannot be done via the casual perusal of articles written in layman's terms (good luck finding AI articles written for the common man).

Basically I think developers avoid going into detail on AI related questions for two reasons; 1) Using technical AI jargon in a conversation with the uninitiated will probably lead to countless headaches (i.e. developers would then be expected to explain the meaning of the technical term they throw at the press and public) and might risk drawing unfavorable comparisons to failed games that utilize the same AI design approach (i.e. neural net blah blah); 2) Giving gamers and the press an inkling as to the overall effectiveness of the AI could seriously undermine the game's reviews and affect overall sales (i.e. don't raise people's expectations unless you're certain you can match or beat them).

However, all our doom and gloom regarding the AI means nothing until Medieval 2 hits the shelves. For all we know CA may have heard and heeded our cries about the AI and have taken the MTW2 development team to task on this issue. One thing for sure is that when MTW2 is released most of us in the TW community are going to take every positive review and praise from an unfamiliar source on a forum with a massive grain of salt. I honestly don't expect to see too much of an improvement in MTW2's AI because when push comes to shove Sega & CA really don't need to change anything. Whether or not MTW2 provides any kind of meaningful challenge is completely irrelevant now thanks to the success of Rome. Rome's phenomenal sales have made it painfully obvious that there are more than enough casual strategy gamers out there who clearly do not care about such things.

Brighdaasa

06-15-2006, 01:15

Understanding the conventions, principles and jargon involved in AI programming is no simple task and certainly cannot be done via the casual perusal of articles written in layman's terms (good luck finding AI articles written for the common man).

a good example of a developer that's doing exactly that is Stardock, for their Galactic Civilizations II. Their AI programmer from time to time gives updates on his AI algorithms, telling the customers how they are tackling ai shortcomings or giving the ai new "routines". Also they're constantly updating their ai according to player suggestions/known tactics through their numerous patches.

As a consequence reviewers actually talked about the ai in their reviews, and communicating about the ai programming undoubtedly gave a boost to sales.

take this for example: http://www.galciv2.com/Journals.aspx?AID=0&p=3&s=1 the On the road again... article

sunsmountain

06-15-2006, 09:31

Good post, Spino, I tend to agree with you on most points. Let's just hope CA will take a leaf out of Stardock's book, and give us a little more insight, or a chance to help them.

Galactic Civilization III is a painfully predictable game though, in which you can calculate a lot of what will happen in the future based on current parameters. Making a good AI out of that is easier than writing one for TW, where the situation on the field can change from moment to moment.

Even though I'm no programmer, I've helped programmers write AI code for games. I read the articles that are out there on the net, complicated or not, know about the A* pathfinding algorithm, it's effectiveness, why it's a little simplistic at times, etc.

In solving AI problems, it usually boils down to translating a game question into code. You first need to recognize & ask the question. That's a hard part, for anybody, but not something programmers are by definition better at than you or me. So we can safely say that the level of AI sharing is too low right now. AI programmers think they can solve big problems themselves, find some sort of solution, but it doesn't work. In a patch they try to literally patch up the algorithm, but it's too late: a better algorithm to start with would have been better.

Lucky for them and us, they make a clear division of tasks. Everybody has their own part in the project, which are fused together in a planned way to get a bug-free game. Rome is remarkably crash-free lately...

Anyway, maybe it is all too complicated, and only a few can solve the problem. But please oh please let them get to it and give them the time...

IceTorque

06-15-2006, 11:59

Whats an algorithm ? Is it like a really complicated arithmatic thingy ?

sunsmountain

06-15-2006, 13:09

Yes. Everything you see on the battlefield has to be expressed in numbers.
This is the hard part. It's tied to the question of what you want to do with those numbers, and how they compare to each other.

The algorithm starts:
Then, those numbers have to be added, multiplied or substracted and then they have to be compared somehow. After comparison, some of the numbers can go to some other formula, while others are stored and called upon later. This results in different commands for different units, who each have their own set of numbers.
The algorithm stops, and goes back to the start.

This already quickly explains the fact why some units go to that direction while other units go in the other direction, sometimes resulting in meaningless incoherent marching back and forth on the field, each time processing their own numbers, ignoring their fellow units. The AI does try to maintain some coverage of its flanks, but in practice is quite sloppy in doing so.

Recognizing this sloppy situation on the field is VERY difficult to express in numbers, because how do you recognize that your units are not in formation? You only have a position and a facing. A unit of chosen swordsman has a smaller flank than a unit of macedonian pikemen. When is a flank covered? One man of another unit? 5? 10? The entire unit (only happens in single line formation)?

Answering these questions once, in the context of this thread, may be done, but answering these questions 100 times or more in complex, dynamic and interdependent code is a Herculean work. Each day I hope to find one of the CA AI programmers begging for help from the forums, but they keep it internal. Let's see if they have more to say.

Lucjan

06-15-2006, 13:23

Lucjan

06-15-2006, 13:26

:embarassed: *cough* ..that was..um..total war* junkies. :embarassed:

Odin

06-15-2006, 13:59

While speculation does seem to be the order of the day, it would be safe not to let ourselves get carried away. Whether the AI will be significantly improved or still have one or two of Rome's squallor rats stuck to it, I think we should all take into account that even though Rome's AI was mediocre at best, that didn't stop us. We played it anyway. And not only did we play it, but we made it better, we modded it, we improved on it, and we made our contributions free and available to everybody here on this forum. And I'd be damn more proud of myself if I had a hand in making someone else's work better than I would if I let myself get all upset about how their shortcomings disappointed me and let me lose faith.

Well no matter how long you have been here, welcome and continue posting. A broad contribution is required to really gain a very good sampling of the opinions out there.

Your quoted post above really says a lot. Why would CA talk about the AI when they dont need to? I have purchased many wargames, and have participated in several wargame forums and enevitably the topic of AI always comes up.

Its never adequately addressed (See Spino's post), but the games get produced anyway. Why is that? Why do war/strat games keep getting made if developers arent talking about the AI? Its rather simple to me, the product sells as is.

Yep, would be spectacular if more was done, more polish but there are very few precendents to support this hope. Maybe I am jaded or to cynical but it seems to me the bottom line is the bottom line, AI issues certainly didnt seem to hurt RTW sales (given the speculated sales figures in this thread) so why would the formula change for MTW2?

What we are doing here is effective, we are communicating as best we can our hope that AI issues from the past will get addressed, but at the end of the day it has to be backed up with your willingness to purchase the game. We can sqwauk all we want about AI, if we buy the game straight away anyways the effort is wasted.

Consumption is your power, consumers are the most powerful force in any business enterprise, sadly, we, the consumers have lost sight of that power and rarely employ it.

doc_bean

06-15-2006, 14:04

Making an AI that can make good decisions isn't that hard (see Deep Blue) in comparison to making an AI that make those decision with a limited amount of processing power (also see Deep Blue actually). I'm hoping MTW2 will be HEAVILY dependent on graphics cards for it 1337 graphics so that there is enough 'power' left on the actual processor for the AI.

The problem with AI in games in general is imo, that you will always have to make a trade off between graphics and AI, since they're both processor intensive. Graphics provide pretty screenshots, AI doesn't, that's why the balance has shifted to the graphics side imho. A game like Galactic Civ II doesn't have great graphics, and it uses the 3D card for most of its gfx processing I believe, so enough room is left over for the AI.

Answering these questions once, in the context of this thread, may be done, but answering these questions 100 times or more in complex, dynamic and interdependent code is a Herculean work. Each day I hope to find one of the CA AI programmers begging for help from the forums, but they keep it internal. Let's see if they have more to say.

Come on, you'd never hear Carmack or Sweeney talk much about the internal workings of their new engines, you'd have them mention some abstract concept once in a while (NURBS, Bezier curves !) to make it seem all a bit more impressive, but they don't go around asking for advice either. Code is almost always kept internal. (Sometimes the source code is released, yes, when the tech is outdated).

Brighdaasa

06-15-2006, 16:49

The thing is: they don't need to go into the details of how they achieved better ai. They don't need to get all techy about it.

For example: they could very well tell us things like

We made the ai check wether they are under fire from guard towers and make the ai retreat more often when they're taking casualties from it
We optimized to code to maintain a more coherent battle line
Added counters against the most common human tactics
Made the ai form bigger armies before deciding to attack in dibs and drabs

This is all talk about ai, not techy at all, and things we're waiting to hear for such a long time too. Even if they don't deem it necessary to relay that in press releases and on the website's front page, they can very easily tell such things on the forums (providing they're actually working on this). Why not? It doesn't hurt and tides over the hardcore community.

So the lack of communication about this means what? They don't care to comunicate this to us? They're too busy to post on the forums? They're not allowed to by the publisher/project managers/...? Or are they just plain not working on it?

sunsmountain

06-15-2006, 20:08

Making an AI that can make good decisions isn't that hard (see Deep Blue)
Ehmm, you mean that team of IBM scientists working on Deep Blue which is at heart a massively parallel, RS/6000 SP-based computer system that was designed to play chess at the grandmaster level. And still hold a straight face saying it isn't hard.... :no:

in comparison to making an AI that make those decision with a limited amount of processing power (also see Deep Blue actually).
Right. Besides the obvious comment of where do you buy a RS/6000, the comparison gets even worse. The number of different pieces in chess is 6 while the total is 16, the number of positions on the field is 64. The freedom of movement in number of dimensions is also limited for most of these pieces (3 to 8). Now consider Rome:TW. The number of different units can go up to 20 and so can the total, the number of positions on the field is practically infinite but let's say you can reduce it into squares of attack or something (this goes horribly wrong at the detail level but never mind), still leaves you the big problem of all the units being able to move like Queens (skip the jokes), ie 8 or more dimensions, actually as many as there are facings (which is 32).

Different units
6 vs 20 (overexaggerating somewhat here, though unit rosters support it,
actually you have only 4: Missile, Heavy, Light, Cavalry)
Total units
16 vs 20
Number of positions
64 vs infinite (the width of the map in yards, squared. A lot of them won't
get used but half of them will which is still a lot)
Number of dimensions in freedom of movement
3-8 vs 32

Just talking about how to move the units, never mind using a special ability (or not), or any ability for that matter.

I hope you begin to appreciate the difference in complexity, Rome is more complex than chess. Fuzzy rules and logic, pragmatic programming solutions and just basic trial-and-error are probably better ways to go about writing the AI than thinking and measuring it all through. That said, a lot can be learnt from Deep Blue in that it uses a lot of previous match data, which MTW2 could theoretically do as well (mainly on the campaign map, the battlemap would need a library of 10000 battles before any reliable intelligence is to come out of it, see complexity of problem described above).

I'm hoping MTW2 will be HEAVILY dependent on graphics cards for it 1337 graphics so that there is enough 'power' left on the actual processor for the AI.
Though I agree with this in principal, more important than CPU power is simply a smart algorithm. And see above RS/6000. That's a mainframe we're talking about.

The problem with AI in games in general is imo, that you will always have to make a trade off between graphics and AI, since they're both processor intensive. Graphics provide pretty screenshots, AI doesn't, that's why the balance has shifted to the graphics side imho. A game like Galactic Civ II doesn't have great graphics, and it uses the 3D card for most of its gfx processing I believe, so enough room is left over for the AI.

This is all true.

Come on, you'd never hear Carmack or Sweeney talk much about the internal workings of their new engines, you'd have them mention some abstract concept once in a while (NURBS, Bezier curves !) to make it seem all a bit more impressive, but they don't go around asking for advice either. Code is almost always kept internal. (Sometimes the source code is released, yes, when the tech is outdated).
See Brighdaasa comment, I don't want code specifics, I know they are not giving those anyway, but it would be nice to hear stuff like:

We're finding it difficult to recognize situations in which the AI should always retreat, and in which they should always attack. We're finding it difficult to get units to listen to army commands like: "stay together". We are succeeding in stopping units from walking back and forth without firing arrows or engaging.
etc.

doc_bean

06-15-2006, 21:33

Ehmm, you mean that team of IBM scientists working on Deep Blue which is at heart a massively parallel, RS/6000 SP-based computer system that was designed to play chess at the grandmaster level. And still hold a straight face saying it isn't hard.... :no:

The comparison part is what mattered most :oops:

Deep blue had massive (computing) resources and so a decent AI could be developed for playing chess, admittedly a much simpler game than TW. I realize how everything scales with complexity, but what I was trying to say is that a computer program can be written that can challenge a very skilled human player in a *simple* game. Deep Blue would likely have beaten 99.99999% of all people trying to play against it, there's no need for such an efficient AI in a game aimed for the mass market. But the reason it can do that is that it has a *very* powerful processor and doesn't have to make real time decisions.

I hope you begin to appreciate the difference in complexity, Rome is more complex than chess.

Yes, I was just looking for an example of good, effcient AI, that's hard to find :embarassed: . I didn't want to go into complexity and efficienty theory and such, though they play an important part, of course.

Fuzzy rules and logic, pragmatic programming solutions and just basic trial-and-error are probably better ways to go about writing the AI than thinking and measuring it all through. That said, a lot can be learnt from Deep Blue in that it uses a lot of previous match data, which MTW2 could theoretically do as well (mainly on the campaign map, the battlemap would need a library of 10000 battles before any reliable intelligence is to come out of it, see complexity of problem described above).

The problem with every method is that it has to provide both an easy, quick calculation and an efficient solution. Completely analyzing the situation and coming up with a true 'optimum' is going to be nearly impossible imo, so I agree 'simplifications' are necessary.

Though I agree with this in principal, more important than CPU power is simply a smart algorithm. And see above RS/6000. That's a mainframe we're talking about.

Of course, but there's a limit to how much you can simplify things while keeping a reasonably good AI. Less processing power means more simplification is needed. Assuming the best known methods for simplication are used. More processing power means more variables can be taken into account, or more quick calculations can be done, so, in theory at least, it could lead to a better AI.

See Brighdaasa comment, I don't want code specifics, I know they are not giving those anyway, but it would be nice to hear stuff like:

We're finding it difficult to recognize situations in which the AI should always retreat, and in which they should always attack. We're finding it difficult to get units to listen to army commands like: "stay together". We are succeeding in stopping units from walking back and forth without firing arrows or engaging.
etc.

Well, they're never very open are they ? They just have the marketing boys talking about the graphics all the time, barely discussing the gameplay, and as CA employees pointed out here, they can get in a lot of trouble for discussing their work here. I think it's an industry phenomenon. You only here the really big names publicly talking about their games/engines don't you. Nobody is going to fire Will Wright for lack of discretion.

But I agree it would be nice if they occasionally made comments about how great the AI is and why instead of just going on and on about the graphics.

Duke John

06-15-2006, 21:45

IMO you're giving the CA AI programmers too much credit. Sure programming AI is difficult, but there are loads of AI mistakes that could have easily been avoided by adding a single if ... then....
The current tactical AI can be designed by anybody who can think logically, it pales in comparision with the feat of displaying so many 3D soldiers. From what I have seen most of the effort seems to have gone into path finding. General army advancement is by a simple formation placed in a text file. Once the AI gets near it finds targets and just charges towards it. It doesn't even check wether your units already targeted resulting in easy wins as 3 AI units charge 2 of yours allowing your 3rd unit to flank, without doing anything you have outmanouevred the AI!

Couple it with the fast movement and you have a game that appears challenging and intelligent on the surface but beneath it boils down to run in a strict formation towards your army, send cavalry 200 metres left of your army, find targets, charge, missile units in range fire.

Puzz3D

06-15-2006, 22:17

So the lack of communication about this means what? They don't care to comunicate this to us? They're too busy to post on the forums? They're not allowed to by the publisher/project managers/...? Or are they just plain not working on it?
They used to talk about the AI during STW and MTW, but that stopped abruptly with RTW. Now all they say is that it's being improved. Sounds like management to me.

One thing they did tell us is that Total War is not for hardcore gamers. Suggestions that strike them as hardcore don't stand much chance of being implemented. Remember, they didn't think players would notice that the AI forgot what it was doing when a savegame was reloaded. That's the level you have to think at now with regard to the Total War series. There is a window of opportunity to point out problems and get suggestions implemented when a patch is being done. The window doesn't stay open for very long. Buy the game and test away so you'll be ready to make your suggestions when the window opens. I did that for 5 years, but I'm not going to do it anymore since the game no longer warrants that kind of effort.

Lord Adherbal

06-15-2006, 22:40

doing good tactical AI doesn't need massive CPU power. Things like collision/distance checking does, but that's already covered in the game (hence why a powerfull CPU is much more important then a video card when you want lots of soldiers in the field - even for MTW and STW). So that certainly isn't what's limited them.
All you need to write good AI is time, the time it takes to turn human tactical decisions into logical scemes, then turn them into code.

sunsmountain

06-16-2006, 10:03

Yes, I was just looking for an example of good, effcient AI, that's hard to find . I didn't want to go into complexity and efficienty theory and such, though they play an important part, of course.
No problem mate, I was a little nitpicky there. Even the problem of position can be reduced by only considering positions that are close (still about 200 for 20 units...) or ignore that and only consider targets that are close.

IMO you're giving the CA AI programmers too much credit. Sure programming AI is difficult, but there are loads of AI mistakes that could have easily been avoided by adding a single if ... then....

If... Then.. what? I know what an if then statement is, but how are you going to recognize the situation that it's crap, in numbers, and what are you going to order then, in numbers.

If it were this simple you are underestimating programmers that have been programming games since 1980... I doubt it would be this simple.

General army advancement is by a simple formation placed in a text file. Once the AI gets near it finds targets and just charges towards it. It doesn't even check wether your units already targeted resulting in easy wins as 3 AI units charge 2 of yours allowing your 3rd unit to flank, without doing anything you have outmanouevred the AI!

Not exactly. The AI tries to do more, and there's more in there than you're now suggesting. I don't know why individual units ignore their neighbours all of a sudden, but there must be a reason. If the AI has a lot of skirmish troops, they can be difficult to catch (ie you have to really herd them or face casualties), in siege battles, you can no longer fire at them and get away with it, they'll charge your archers this time. They keep their formation better before engaging. The improvements are minimal, but they are there.

I did that for 5 years, but I'm not going to do it anymore since the game no longer warrants that kind of effort.
Poor Puzz, you're taking this poor AI the hardest i think. I agree if they don't put the effort into it that we are looking for this time, they'll lose well at least 2 of their fans this time. And then we'll have a budget buy in a few years from now and come back for a patch or 2. Hopefully others can do the work for us :)

IceTorque

06-16-2006, 12:19

Well, I reckon their algorithm thingy's are pretty good, and it's their gameplay settings that need to be improved.

Lucjan

06-16-2006, 12:47

I see a schism opening between loyal fans who are just happy to see the series continueing and fans who feel depreciated because CA isn't catering to the wants of the more involved gamer. I just hope CA can build a bridge.

The best thing I think they could do is take the best of their ai coding from their previous games and expand upon that in the ways they see fit, then do something along the lines of...

1 Seperation of difficulty levels placing the ai at more heightened levels of awareness and tactical knowledge across the board. This, as per every game with alterable difficulty levels, has already been done.
2 Implementing a new option that would differentiate between 'realistic' and 'stylized'.

Thereby allowing for a total of 8 different difficulty levels, those four found in the stylized difficulty setting would be on par with the skills of casual gamers looking for some quick fun or those who are new and just not quite as picky as us. And those four found in the realistic difficulty settings would adress our concerns like unit speeds, strength and defense balancing, questionable tactical movements (the ever irritating 'macedonian army besieges your city then the next turn runs away for no reason whatsoever'), and sometimes just downright stupid ai. (I remember on more than one occasion having ai chase a unit of mine up to my walls and then running back inside, the ai would camp at my walls and I would just up the speed and let the towers pick them all off one at a time because the ai was actually too freaking stupid to move out of the tower range.)

3 Take into consideration that the group of people who built their strategy game empire were the loyal 'hardcore', though I don't quite like that term, gamers like us, to whom the total war series was a dream come true, and try to increase the challenge of the gaming experience for us while still building on their new found 'casual' gamer market.

Lord Adherbal

06-16-2006, 13:53

If... Then.. what? I know what an if then statement is, but how are you going to recognize the situation that it's crap, in numbers, and what are you going to order then, in numbers.

if "general is not the single unit in army" then "dont charge enemy"
if "archers have ammo" then "fire from max range while rest of army waits"

if they can't turn that into actual code they can always give me a phonecall.

econ21

06-16-2006, 15:42

Adherbal']if "general is not the single unit in army" then "dont charge enemy"
if "archers have ammo" then "fire from max range while rest of army waits"

In a similar vein, on the strategic map:

"if multiple armies can attack an enemy stack, get them to do so simultaneously (one adjacent to support another) rather than sequentially"

"if an army can kill your army, move out of range"

"don't attack if you are going to lose"

(Ok the latter two are going to have to be more complex, but it helped make homm3 quite competitive)

Sometime back I posted a list of simple tweaks that would make the RTW AI better. IMO, a few simple changes would improve it alot.

maxpriest

06-16-2006, 17:03

I sincerily hope they will be able to fix the simplest Ai problems rtw had, or at least give the modders a chance to fix it.
With 1.5 (battle diff very hard) : A unit charges at me, I defend. The unit then proceed to stop 2 meters in front of mine and faces a different direction.
That happens way too often.
I mean...what the hell.

Callahan9119

06-17-2006, 06:19

Lucjan

06-17-2006, 12:35

what you said....oh and by the way, there are 7 people right now on the various !!mtw!! boards at 1:30 am est, thats loyalty.

Gotta love us nightowls. ~;)

sunsmountain

06-19-2006, 09:13

they didnt have to dumb it down to sell it
Amen. Let's just hope they didn't get around to making it smarter, yet.

Puzz3D

06-19-2006, 18:25

I see a schism opening between loyal fans who are just happy to see the series continueing and fans who feel depreciated because CA isn't catering to the wants of the more involved gamer. I just hope CA can build a bridge.
That's not going to happen. According to them, RTW is not aimed at hardcore gamers which is their reason for not implementing many of the suggestions made by the involved gamers, not keeping all of the features of the STW/MTW battle engine and not providing a speed setting that matches the older game. The older fanbase has now become a liabilty for them. I even saw a post by Dr. Jambo at .com about the Alexander mini-expansion fully deleted by MikeB.

Lucjan

06-19-2006, 21:52

Do you think it's because sega's putting some pressure on them to make the 'casual' gamer the new priority?

Personally I don't care much either way, when the game comes out, whether it's a work of art or otherwise, somebody will put together a team and mod their hearts out, and make an extremely enjoyable game for all of us.

Spino

06-19-2006, 23:15

That's not going to happen. According to them, RTW is not aimed at hardcore gamers which is their reason for not implementing many of the suggestions made by the involved gamers, not keeping all of the features of the STW/MTW battle engine and not providing a speed setting that matches the older game. The older fanbase has now become a liabilty for them. I even saw a post by Dr. Jambo at .com about the Alexander mini-expansion fully deleted by MikeB.

Did you manage to read what Dr. Jambo wrote before his post was deleted? Do you have a link to the thread he posted in?

Given the pedestrian and inhospitable atmosphere over at the official forums why on earth would any TW veteran even bother wasting their time over there? If you want sensible and reasonably civilized discourse on anything related to the TW games the Org & Twcenter.net are your best bet. Once RTW was released and the negative feedback from the TW community started to pile up in Com's forums the place became intolerable. Intolerable not because of the typically immature attitudes but because of the knee jerk, sledgehammer tactics utilized by the mods to deal with any kind of negative feedback.

sunsmountain

06-20-2006, 11:38

Intolerable not because of the typically immature attitudes but because of the knee jerk, sledgehammer tactics utilized by the mods to deal with any kind of negative feedback.
Well, with most of the forum audience being aged 16 and below, the awkward bulletin board format and a lot of posts per day, most topics didn't stay on a page for long at all. Meaningful discussion was and is almost impossible there, which is why true fans quickly find totalwar.org and twcenter.net

The mods of course wouldn't take well to "CA die" threads, so those were deleted & banned. And rightfully so. Be glad CA don't use the totalwar.com forum very much to find out what the fans really want. Then again, they might just visit gamespot and browse the reviews.... 9 9 10 9.5 8.8 10 10 etc...

caravel

06-20-2006, 13:17

Well, with most of the forum audience being aged 16 and below, the awkward bulletin board format and a lot of posts per day, most topics didn't stay on a page for long at all. Meaningful discussion was and is almost impossible there, which is why true fans quickly find totalwar.org and twcenter.net

Unfortunately true. Any decent threads are buried under a deluge of spam, and soon disappear off the first page and are lost. The mods over there just close anything that may be controversial. Religious and political discussions may cause flaming so they're not allowed. Even nationalistic threads are banned. Anything that even looks like it may become a controversial thread is closed just in case. As a result the place is stagnant, heavily restricted and dull. The closing of threads is very random. Spam is usually closed if it's a thread that was deliberately started as a spam thread. Other threads that go off topic are often not spotted for a long time, often the mods will join in and not bother closing it, then another admin will finally arrive and close a thread that other mods have been posting in! Anything that doesn't fit CA's policies is usually closed in an instant though...

I don't care much about the messageboard format over there, ok so they've got no signatures and avatars, that's not the issue. The content is more important, but any decent content is being stifled.

The mods of course wouldn't take well to "CA die" threads, so those were deleted & banned. And rightfully so. Be glad CA don't use the totalwar.com forum very much to find out what the fans really want. Then again, they might just visit gamespot and browse the reviews.... 9 9 10 9.5 8.8 10 10 etc...

Any criticism of CA, that goes beyond mild criticism, is snuffed out. Fair enough they left the great turns debate open, but that was because they knew that they couldn't supress it without closing the board altogether.

Any questions addressed to the CA staff about the AI were ignored or responded to in this fashion: "I was sieging a castle and I just had to stop and look at the in awe...". This is what has led people to believe that CA are not interested in improving the AI. People only have the words posted in the press releases, interviews and the forums to go by, so you have to understand that people may make [i]assumptions. Some people's responses were way over the top, and caused a further tightening of the rules. The " should be included in the game" threads caused the nationalist threads ban. This is the problem, instead of addressing problems with subtelty their heavy handed administration just go straight for the throat and simply outlaw anything that [i]may cause a problem. But the situation at .com is: Juvenile idiots make a mess of the forums: Everyone, including the repsonsible members have to be bound by the same childish restrictions brought into effect to control these idiots.

Puzz3D

06-20-2006, 16:12

Do you think it's because sega's putting some pressure on them to make the 'casual' gamer the new priority?
The shift in direction occured under Activision. Sega has paid for more patches than Activision, but the direction the gameplay is being taken seems to be the same.

Personally I don't care much either way, when the game comes out, whether it's a work of art or otherwise, somebody will put together a team and mod their hearts out, and make an extremely enjoyable game for all of us.
Not for multiplayer. Even for singleplayer there are certain things no mod team can change because they are being kept inaccessible.

Did you manage to read what Dr. Jambo wrote before his post was deleted? Do you have a link to the thread he posted in?
I don't know what he posted. I've been trying to get a hold of him to find out. I can't find the thread now, but I did find this posted by an admin at .com in a thread about the Alexander mini-expansion: "There doesn't need to be a thread commenting on the problems people are having. That is simply spam and serves no constructive purpose. Closed."

caravel

06-20-2006, 16:26

The MTW gold edition issue was treated similarly. When the version 2.0 problem was exposed. An admin posted a day later and basically said: "use the VI 2.01 patch" (which you can't) and then closed all the topics relating to the issue.

Sabuti

06-25-2006, 14:25

Mainly because the mass-market seems to be more interested in games looking great instead of playing great. CA also have the problem of having to balance the game so people new to it or people with very little knowledge of tactics (which I'm betting is the vast majority) won't be completely over-run by the AI. Unfortunately this means the game is less challenging and often more frustrating for the more seasoned players who expect AI improvements in every new iteration of the game.

That's what difficulty settings are for. So this is no reason to not improve the AI. Without substantial AI improvements this game is not even worth buying.

GFX707

06-25-2006, 21:25

Yes....personally I will probably try a "friend's" copy first to see what the improvements to the AI are like. If it's just more eye-candy like RTW was to MTW, then I will not be buying it.

Brighdaasa

06-26-2006, 00:44

a "friend's" copy

Nice euphemism, got to rember that one.

Although that's basically what the demo is for, to see if you like the game.

Duke John

06-26-2006, 07:01

Normally yes, but CA heavily scripted the demo and let the AI come so close that there was hardly any time for manouevring left once the battle started. So you cannot really get a good idea of the AI. However if they do that trick again it will be for me a clear sign that they want to hide something. And of course the unitspeed will play a big role on wether I will lend a friend's copy for the first time :wink:

Lord Adherbal

06-26-2006, 10:58

I borrowed a friend's copy of RTW too, just to get it a day or 2 before the release over here (belgium). But then I bought it to get a CD key for MP. But I doubt I'll be needing one for MTW2.

sunsmountain

06-26-2006, 15:04

If the multiplayer balance is similar to RomeTW at first, you won't be missing out: All Cav armies at first...

Plus a quick 1.1 patch to get started in the first place.
Plus a gamespy arcade that seems to drop random people while other (or the same) people are lacking in connection.

But at least you won't have the Romans with automatically the strongest melee troops and cav and pretty good missiles, which is nice i guess...

The Wizard

06-26-2006, 15:31

CA might defend itself with saying that the AI shouldn't be much of a challenge -- but that's where the learning curve comes in, huh? Games, if they are on computers or not, are supposed to be challenges. In video games that challenge is either provided by the AI or online.

As such, everything else is secondary and is there merely to provide a credible platform from which the AI or another human player can provide a fair and straightforward challenge that is not insurmountable.

What does this imply? That the AI can be cunning like never before and still a game doesn't have to be a drag for 'newbies'. Part of the fun in strategy games, is, after all, learning and growing into a fearsome general.

As such any argumentation going along the lines of "let's not make this too hard on the new players" is null and void. If you don't want a challenge then don't play games.

Oh... and on the .com: it reminds me more of North Korea than anything else. ~:)

Kralizec

06-26-2006, 15:37

Hey, you know the origin of the disclaimer "game experience may change in online play" comes from right ~;)

:wall:

DensterNY

06-26-2006, 16:51

Its a shame that CA doesn't open up the source code for some parts of their Total War games so that the devotees can fine tune their games beyond Mods. I know the ability to Mod already has allowed for some phenomenal revisions to the Total War games but the AI issue has been raging since the release of Rome.

I know its wishful thinking but I could imagine having learning and adapting enemies or at least ones that didn't do moronic things that allows me to slaughter them.

Lucjan

06-26-2006, 20:02

Sabuti

06-27-2006, 04:29

I read earlier in this thread speculation that CA is trying to appeal to the casual gamer to widen its audiance. Let me put it this way, There are people who like and play strategy games, and there are those that don't. Any attempt to make a strategy game appeal to people who don't usually play strategy games is delusional. You would have to change the game so much that it is no longer a strategy game. CA should consider the wants of their core audiance first. If they lose them, its all over.

sunsmountain

06-27-2006, 08:30

If they lose them, its all over.

No, it's not, but then CA become dependent on having hit titles. That's not a problem as they are dependent on that already, as is any company in the (for consoles, still lucrative) business of gaming.

As a sidenote, CA is being awefully quiet on fora, that means they must be working hard on the game.

DensterNY

06-28-2006, 16:24

At the very east, CA could provide the game with some kind of code that can be switched on or off via use of the 'arcade battles' button that would allow the computer to store a small database of the human player's movements in battle, and search it's own coding for a reasonable counter to that movement, so that when the computer sees it happening again in a later battle, it could provide more of a challenge by attempting to counteract a tactical movement it already recognizes as having lost to once before...lets face it, no matter what you do, the only real tactical movement the computer makes, is realigning its lines with your own to prevent you from having one flank significantly stronger than theirs.

I think I mentioned this before on another thread but I wondered if CA could outline for the AI generals a database the same way that you'd have if you played chess against the computer. Depending upon the scenarios you put before the computer it would do its best (according to the play level you selected) to anticipate and counteract your moves. A great general would have full access to this strategic base and try to plan ways to smash your defense or neutralize your offense. I imagine it would be like playing against the computer at Grandmaster level where within a few moves the AI already envisions its victory against you.

Overall though, I would like to say that I'm not bashing CA or its AI... I still enjoy their games immensely but just wish there was a little more fine tuning. I imagine that they read these forums and refer to us with a roll of their eyes as the AI-Fanatics.. hehe

HarunTaiwan

06-29-2006, 07:28

Remember, they didn't think players would notice that the AI forgot what it was doing when a savegame was reloaded.

Thank you for reminding me.

Duke John

06-29-2006, 07:55

More importantly, instead of making an announcement that they would try to fix as quickly as possible, they shoved it under the carpet. And giving idea that the community overreacted and that it wasn't a gamebreaker. Only to fix it in a patch as if it was a bug.

Lucjan

06-29-2006, 17:59

Puzz3D

06-30-2006, 18:39

If AI is just a bug, then it's akin to those old b movies "Them" or "Earth VS The Giant Spider". But, in light of that, in both of those movies people ended up squashing the bugs anyway.
SEGA/CA didn't squash the siege bug in RTW/BI before they cut and ran. They even have the nerve to sell an add-on to a game that will get no more debugging. Apparently, the same bug is in the RTW Gold Edition and will be in the Eras Edition. SEGA/CA have brought a new meaning to the term "Gold Edition". Just look at the MTW Gold Edition fiasco. Hoping that M2TW is somehow going to follow a different path is just wishfull thinking.

Lucjan

06-30-2006, 18:51

...I'm going to start referring to the people in this thread as the "gloom-n-doomers"

BeeSting

06-30-2006, 23:53

i don't expect much from an AI to make the game more challenging.... it's simple as: keeping it's main line solid, which means that individual units do not act independently and choose their "weak" targets--or a group of units of that line will act as ONE unit; and keeping flanks secure with mobile forces having them be the deciding factor, acting more independently, free to make sorties from the main line, charging a unit with more than one unit if possible, and always flankng the enemy's line and not being lured into charging the center of it.... basically if the AI is more reserved and keeps solidity of its units exposing the least possible flanks, i would have a difficult time winning.

as for a group of units becoming one unit, let's say a group of five phalangite units for example, an attack from a single enemy unit on its line should stop the entire group's advance, so to keep its integrity as a solid line and not have its non-engaged units continue to advance and or merge into that single enemy unit. As they do in defensive stance, a group of infantry units should keep it's form, be a single unit--make a unit's flank glued to other units on its sides or something. That is all i ask.

Lucjan

07-01-2006, 00:22

econ21

07-01-2006, 00:47

By asking multiple units to 'glue' themselves together the way you suggest you're actually asking for worse ai than the computer already uses.

An interesting debate. I certainly use the phalanx wall against the AI, but I can see it might be harder to pull off against a human - which is why historically phalanxes need strong flankers like hypaspists and Companions.

RTR Platinum improves the hoplites a lot by just getting rid of the phalanx formation altogether. They are a big, dense unit of spears. They move fast, so are not easily flanked and destroyed piecemeal. Plus they only have spears, so you don't get the "switching to swords" vulnerability of pikemen. I think the AI just struggles with the phalanx formation - it's too slow and vulnerable.

Making phalanxes fast may even have some authenticity with the medieval Swiss, who are supposed to have attacked at the run.

BeeSting

07-01-2006, 01:49

By asking multiple units to 'glue' themselves together the way you suggest you're actually asking for worse ai than the computer already uses. A group of five hoplite units being stopped by one unit of whatever is utterly pointless. It would cause all five of those units to sit and do nothing as you flanked or wrapped around and charged them from the rear, this is easy enough to do now the way it is, but you're asking for a free, resistanceless shot to the back.

Good point. i would imagine it won't be wholly unrealistic for the entire mile of phalanx line to stop for some suicidal unit to tackle it, so it won't break, if that ever happened. so what do you suggest then, to have all units in the line converge thus caving in the flanks and opposing no walls to its opponent? i see your point tho, a suicide unit can tackle/hold the entire center line while you can envelope around to the flanks and to the rear. For instances like these however you have as i suggested flexible flanks that are free to make sorties from the base (center line) and clear the nuisance and then quickly return to its original position; the distress caused by the number of enemy plus cavalry units hitting it on its sides will quickly cause a route so the line can proceed to engage the enemy's main body in good order.

AI should organize its units in two categories, flanks and center: center being the platform/base or operation while the flanks work off of it--flanks of course the opposite, as being flexible--and units should have this labeling so AI could easily organize its line...

My point in previous post is that rtw ends up being a wild geese chase; units ending up all over the friggin map. And it is true, whether a force of men is static or highly flexible, if they are consolidated in a sound form, it will be difficult to defeat--not to mention the fact that it will keep the game experience from being a major pain of meaningless chase from one end of the map to another to end the battle. given the technical limitation, AI should follow such fundamentals as keeping form or consolidation of force, as its priority. keep it simple, stupid--so to speak.

Lucjan

07-01-2006, 12:58

The AI in rome has the fault of having no hierarchal structure by which it should give commands. It solely gives commands to each unit individually and tries to maintain them within a fair distance of one another. What should occur is a simple coding of hierarchy into the way the ai thinks, with formation, solidity, effectiveness of unit placement against the opposing unit type and mobility of formation taken into account when it engages the enemy. Instead of just moving x number of units with individual agendas towards the enemy but keeping them within a close enough distance of each other to appear as if they're moving in unison with one singular plan.

CBR

07-01-2006, 13:15

Good point. i would imagine it won't be wholly unrealistic for the entire mile of phalanx line to stop for some suicidal unit to tackle it, so it won't break, if that ever happened.

If we are going the historical way then it would actually be very difficult for the whole line to stop just because a part of is attacked. It would be near impossible to control a line that is so wide. Polybius mentions the problems the Macedonain Phalanx has against the Roman legion, as it doesnt have any reserves when holes appear because of local combat (units either pulling back or advancing) or differences in terrain when advancing.

CBR

Lusted

07-01-2006, 13:52

SEGA/CA didn't squash the siege bug in RTW/BI before they cut and ran. even have the nerve to sell an add-on to a game that will get no more debugging.

Your kidding me right? The load/save bug was fixed in 1.3/1.4, then Ca bough out patches 1.5/1.6 that fixed bugs with 1.3 and 1.4. Alright not all bugs were but some were. Plus the ai got improvements. Now i hardly call that cut and run tactics do you?

Puzz3D

07-01-2006, 16:47

Your kidding me right? The load/save bug was fixed in 1.3/1.4, then Ca bough out patches 1.5/1.6 that fixed bugs with 1.3 and 1.4. Alright not all bugs were but some were. Plus the ai got improvements. Now i hardly call that cut and run tactics do you?
If they hadn't fixed the load/save issue, they would have been a laughing stock in the gaming industry. If they leave the seige bug in RTW/BI, then they have cut and run since it's not just an AI or playbalance issue, but it's probaby deemed not very damaging to their image and future product sales. Alexander is version v1.9. Is this a marketing ploy designed to make you think RTW/BI will get a v1.7/v1.8 patch?

Silver Rusher

07-02-2006, 14:50

IF a company decided to make a historically accurate game of the order of EB, with the dramatically reduced audience for such a game then they would have to charge a lot more, say theres 10% of the market that there is for RTR, they would have to charge £250 just to make the same kinda profits as CA.

Maybee CA could be incouraged to allow mods to use their engine like UT2004 has, the main reason i brought it was because of all the cool mods there are for it, its like buying 6 games in one!
How would charging £250 allow them to make the same profits as CA? No, seriously, I don't think there is a single person who would be willing to pay £250 for Europa Barbororum, sorry to the developers it's a good mod but that price is insane.

BeeSting

07-03-2006, 21:40

sorry about this edited below... didn't mean to spam

BeeSting

07-03-2006, 21:40

edited below

BeeSting

07-03-2006, 21:40

Instead of just moving x number of units with individual agendas towards the enemy but keeping them within a close enough distance of each other to appear as if they're moving in unison with one singular plan.

I think the problem occurs with having the attack/charge bonus for units via selecting or clicking the individual enemy units. This problem is most evident for a line of phalangite units, they have a hard time keeping a solid unbroken line as human players can expose this game feature by stretching them out, especially in defensive or guard mode the units become as in broken blocks tackling enemy lines in all sorts of angles instead of tackling parallel to the enemy line. in order to avoid this, a player can normally clicks behind enemy line and hit backspace as the line engages, sacrificing the attack bonus--but it's better than breaking up the solidity of your line--but for the AI.... let's just say my emersion is ruined when ever i see AI use phalanx units. This is just many problems with phalangites with AI--simply because of the game feature and AI is not spohisticated enough to think out side that feature to keep form. How to resolve this is beyond me. Just to take some pot shots at this I think they would have to recode the phalanx to approach the enemy differently more in reserved manner, not to order individual units to tackle their own targets but to appraoch the center of the line as a whole as if--as mentioned in my previous posts--say five or six phalanx units were as one unit. So they as one unit when selecting their target to attack it will choose the very center unit of opponent's line, as in phalanx unit in guard mode, so to not cave into that enemy unit in the center line, giving the attack bonus to all of it's line yet keeping the solid unbroken parallel line. I hope i made sense. FYI, English is my second language. Back to the point, this can be applied to all formations if the AI can organize its units in 2 flanks, 1 center, and perhaps reserves to fill in the holes.

Duke John

07-04-2006, 08:11

BeeSting

07-04-2006, 10:17

If AI were to organize its units in 2 flanks, 1 center, and reserves to fill gaps in the line--it gives each unit specific roles that fall under these three categories--will give cohesiveness to the battle formation. This would be far superior to the seemingly random orders given out to "seek the weakest unit and charge it with your strongest" or "flank your opponent no matter if you are at the very center of the battle line" leaving a gaping hole in your own line to flank the enemy.

Center would be formed mostly of infantry, mainly to hold the enemy line, and the flanking units on its side are given the order to make sorties from the base (the center) to neutralize the enemy flanking units before charging the centerline on its sides. Current RTW AI just charges whatever that exposes its sides and rear, easily falling for traps. Cavalry units should not just charge at enemy infantry without engaging the enemy cavalry first. AI use of cavalry often if not always trashes its own first by throwing it away at the nearest enemy infantry units with the weakest stats or ones with exposed flanks or rear, which leads to its mounts being wiped out before the player's cavalry, by players cavalry, for exposing its flanks from engaging the player's infantry. Its decisive forces are gone within minutes after the engagement and the fate of AI's army is obvious--repetitive process begins for the player--clean up time.

Again flanks and rear reserves should work off of the center, the base. Specific roles for units in reserve to stay put till gaps occur in the battle line or to back up a wavering unit whether it is on the centerline or on the flanks. I don't know how hard it would be to code this to prevent battles from ending up to be a meaningless goose chase or all units doing their own thing oblivious to what others of its own are doing. I may be wrong in my observation but whatever method CA uses for its solutions, I hope the AI in MTW2 will be "improved" as promised to such extent that we will forget its faults in RTW.

sunsmountain

07-04-2006, 17:35

I don't know how hard it would be to code this to prevent battles from ending up to be a meaningless goose chase or all units doing their own thing oblivious to what others of its own are doing.

Exception leads to exploitation. In other words, if programmers program anything to prevent anything from happening, players can always find the condition on which this rests, and exploit this. This is particularly true in STW and MTW, where the AI simply relies on certain constraints in the map (defined pointers, initial formation before engagement, etc.) that cannot be exploited by the player. For randomly generated maps, and any possible army of 20 units, these constraints become impossible to declare.

You're describing the "standard" military strategy of deploying in the center and on the flanks, but there are many more of course. And what about 3-way battles? What you need, and what CA have tried to program (and the basis IS there), is a AI from the ground up, ie:

1. unit level AI responds to threats nearby. This is the most important part of the AI and needs to function before you can program the rest. That means: Archers must skirmish properly (they don't always), Hoplites must select their target properly (they don't), etc.

Unfortunately the development time for Rome TW was gone and they didn't get to finish this part, so that they could continue with:

2. army level AI, consisting of group level AI (if groups are defined or recognized by the AI, which is very hard to program) and if not, AI that all units must respond to.

This last part AI is in fact programmed, in battles where the AI has already lost half of its army, it gets the withdraw option which it uses. (the enemy army is routing... enemy reinforcements chose withdraw)

My criticism in all of this is that CA aimed too high when initially designing this AI, in particular with the second part of the AI, the army level AI. Recognizing situations on the battlefield and knowing what to do is typically something only humans can do, while I would give the AI a static set of strategies and formations to choose from, and simply make a somewhat educated pick.

This ensures coherence on an army level, gives the AI a strategy, but doesn't ensure all units will attack at once (if you outmaneuver it), which is also essential. This is why i understand they have left the RomeTW AI as is (without the army level AI), knowing full well that players will already be challenged by unit level AI, ie a good skirmishing AI with lots of horse archers is difficult to beat by any new player.

MTW veterans however, find it simply annoying, and expect to be challenged by army level AI, which was present in MTW and STW (though very static and constrained, most vets didn't and still don't notice).

Fwapper

07-05-2006, 13:40

I think people are really giving CA a hard time over the AI. A lot of what people say is just rumors which have overgrown.

It is very easy for CA to show off their cool graphics and new animations and so on, but much harder to demonstrate the AI. We'll have to wait for the demo or actual release before we can see it for ourselves and know for sure.

Of course the marketing will be largely driven by the graphics - it is the most obviously improved and visible part of the game. They're not going to charge some cavalry into a city and say:
"Oooh look! They went where we told them to go!! They didn't run around like headless chickens!!"

What I'm trying to say is don't bach CA too much over the AI as we haven't actually seen it yet.

Lucjan

07-05-2006, 22:08

What I'm trying to say is don't bach CA too much over the AI as we haven't actually seen it yet.

Oooh, another voice of reason attempt. hhehe, we've tried, it's just so much more fun being all "GAHH, CA WE WANT BETTER AI" *charges CA headquarters with angry viking horde like capital one commercial*

The Wizard

07-07-2006, 23:02

...I'm going to start referring to the people in this thread as the "gloom-n-doomers"

The end of the world is nigh!...

Regardless, we love this game dearly. We want to see its full potential realized. What makes us sound bitter is our disbelief that we are in any way listened to, which is quite frustrating as far as I'm concerned.

It's quite bad style, too, not listening to your hard core. This -- in a world where there are developing teams that develop patches with professional gamers (Blizzard) or actively helps and supports the modding community, sometimes even offering jobs to top mod makers (Bioware)?

Sad. :no:

The hard core is your friend. Listen to the hard core. They are your access to real, active player experience.

Nemeo

07-08-2006, 02:46

Unlike many people here, I think that making a good AI for a game is far from impossible. On the battlefield we don't ask it to simulate feelings or anything arbitrary. The game has many restrictions and a great set of rules, like "units can't fly", "units can't fight underground", "units can't go through other units and walls", "cavalry beat infantry", "pikemen beat cavalry"...etc. From then on its easy for a coder who knows his game to have it check things like "I'm a cavalry, there's a phalanx unit in front of me, I must not charge" and then have it attack another unit or flank the phalax unit. The AI can check every units and have them moving at the same time in the same second, it should be able to beat a casual player from time to time. The AI IS the game, it should know better than the player how it works, what would not play in its favor. If coders don't tell it to analyse the situation it faces and just put a routine to have front unit making a straightforward charge and side units attacking from the side (they don't really flank in RTW as they often charge before the rest of the army, thus their target is rarely fixed before they charge), then this is no more a strategy game. It's like being a cat toying a mouse.

There can't really be a unit over another in TW. So it's almost like a chess game. There are good chess game around there, with very potent AI. I've never beaten one of them in the normal difficulty. So no one will ever persuade me that it's impossible to have a decent AI in any game, as we don't ask computers to simulate artificial life.

Tellos Athenaios

07-09-2006, 13:15

With designing AI there are always a few problems:

Who are you designing it for?
How much does it take for a slow system to make use of it? People with slow systems should be able to play your game by simply tuning the graphics down - and not the AI.
(Ever tried to create a snake game on a TI84plus with some AI to make the food move in order to escape from you? If yes, you know what I mean, even a relatively simple AI (with the above mentioned system and program this means: advoid obstacles, and if your route is obstructed try and flee the second best way away from the snake) slows down a lot.)
Do you really want it to be ALL seeing, and ALL knowing (this isn't realistic, since on a true battlefield most commanders wouldn't have been able to see ALL of it...)? If yes, how do you get it to respond in a sensible way?
How long do you have to develop it? I mean, an ALL seeing AI would require a self learning program - that's what they are into at universities and this has lead to 'speaking' systems. For companies however it would be nice if a new version of a series didn't take several years to complete...
How, when, and where should AI change it's battle formations? You can't just change your battle formations at any time, you can't do that at any place, and the key to do it succesfully is to do it in the right order, and in the best way adapted to the new given situation. To humans with a little sense of tactics this all comes natural but how do you copy the human way to the AI?
And to refer to my prior point of slowing down your system: to make effective use of battle formations and if needed to change those, the AI has to constantly check it's troop positions, movement directions, speeds, enemy troops and so on. And not only checking those, it has to adapt to new situations too, wich often requires entire new formations and always requires a new forecast. And so, a real killer AI in battle would also be one that takes processor minimal requirements to a whole new level.
And probably loads of other problems, to make it even worse. Such as the game is based on an earlier game wich had it's specific faults and bugs, and since this is some way just a new version of older software many of these faults and bugs can't be edited out without having to completely rewrite either very large portions or the complete game. Rewriting would then lead to a new game that simply doesn't fit as successor in the TW series nor as a start of a whole new series. (Often rewriting such large amounts of code requires rewriting other code not concerned with AI too - to make sure the comletely changed AI is going to work you might need a new 'enviroment' as well.)

Now why don't we see improved AI (yet)?

This is, in my opinion, simply a matter of REALLY good AI can't be seen (as it would seem as if as human was in charge and not the AI), improved AI is hard to notice (since this is mostly noticed in detail or by trying it - but there is no demo yet so this way in cancelled out for now) and to show AI you need to show a lot of your game. And all we have seen for now is just to build expectations not to give us answers on these sorts of questions: a company probably doesn't like when it's new products lose all of their 'surprise' values. So we don't get to see everything that's in the new game - which I feel confident about will include an improved AI. I'm very positive about this, since I doubt CA wouldn't make use from some of the excellent work Modders have done on AI battle formations like DarthVader's formations mod.

sunsmountain

07-24-2006, 18:24

Do you really want it to be ALL seeing, and ALL knowing (this isn't realistic, since on a true battlefield most commanders wouldn't have been able to see ALL of it...)? If yes, how do you get it to respond in a sensible way?

On Very Hard, I would LOVE the AI to cheat and know the exact make up of my army before we even deploy, and have its battle line already in an ideal counter position at the start. Responding in a sensible way needs 2 parts: A good unit level AI, which will not do stupid things while carrying out the second part which is army level AI, trying to get the best match-ups, best flanks, best path of attack etc.

They tried to finish the first part and didn't succeed enough. Basics were just not right. Things that could have been prevented by a better program, but were not. They certainly will have to be with MTW2, or a lot of RTW/MTW/STW owners will not be buying it. At least I won't. I will not play a game that screws up my generals, phalanxes, or archers marching back and forth playing their bagpipes and licking each others ...

... the second part will be hard enough as it is.

And to refer to my prior point of slowing down your system: to make effective use of battle formations and if needed to change those, the AI has to constantly check it's troop positions, movement directions, speeds, enemy troops and so on.

Ahum. Right now they probably check no more than enemy troop position, and troop type. Movement direction is a variable that adds too many dimensions into the problem, speed is redundant and inferior to troop position, as the AI has to evaluate its decision every so many seconds anyway.

I'm very positive about this, since I doubt CA wouldn't make use from some of the excellent work Modders have done on AI battle formations like DarthVader's formations mod.

Fair enough. CA have already implemented a measure of army level AI in patch version 1.5/1.6, where starting army formations were implemented, and if you look closely, the AI does try to maintain them while engaging. Of course, when it comes close to your troops....:wall: :wall:

caravel

07-26-2006, 14:13

"units can't go through other units and walls", "cavalry beat infantry", "pikemen beat cavalry"

This is how MTW and STW AI, for all it's flaws, worked. Cavalry did not hurl themselves at a formation of Chivalric Seargents, it tried to flank the more vulnerable units. It would also try to adjust it's formation to ensure that it could have a go at countering/beating the player's. For example if If your cavalry were on the left wing, it would move it's spear units to engage/obstruct them. Simple stuff. STW was in some ways better,as the AI seemed far more intelligent in battle. If it couldn't win it would withdraw immediately after inspecting your troops, it was more effective at flanking and far more cautious, often deploying much better defensively and not being easily drawn out. MTW AI was better at bridge defending.

Also, and very importantly, the AI, and the player, had the time to deploy correctly and change formation using some kind of tactics. This is unlike the RTW cavalry that all appear to buzzing about on motorcycles, not horses, changing direction like flocks of birds. This is just a few of the reasons I simply cannot get into RTW, and despair at the fate of M2TW.

The campaign map is a different issue of course. There is a sort of AI on the campaign map, well a 'diplomatic engine' for lack of a better description, and this has never been that hot, but the RTW map uses a different system that was totally new at the time, unlike the MTW risk map which had previously been in STW. Personally I didn't like it, and it's RTS/Civ style look. I prefferred the risk style general's map to manage my campaigns. MTW and especially STW had a sort of 'sim' feel about them. You didn't feel as if you were playing a game at all. RTW felt and looked like a game. In the ending video of STW, there is a highly polished fmv sequence of a risk style campaign map with pieces being pushed towards a castle to assault. This is the sort of map for M2TW (and RTW) that I would like to have seen. (I will post an image later)

I agree 100% with Puzz3D also and most of the other doom 'n gloomers for that matter.

:2thumbsup: