My very abbreviated take on what has happened

[Pak40]

Shrapnal's All American still wins the award for longest game in development. They were developing it way before CMBO was in development. I think I was still in college at the time and I graduated in 1994! Well, maybe not that long ago but it was well before the turn of the century.

[Peterk]

Attempting to mathematically solve a game and simply providing an opponent which is capable of delivering a decent challenge are two different cans of worms.

[Hub]

Although I've played wargames for decades, I seldom ever play online - mainly because nine out of ten experiences have been unpleasant ones, for reasons already stated by others.

No, the AI experience is not as fun or as challenging as against another human. I look for games that have editors that allow map making or scenario creation, in order to get more mileage out of them, since MP isn't doing it for me.

[John Kettler]

Michael Dorosh,

B-17 Queen of the Skies is emphatically NOT the first solo wargame. SPI's Wolfpack, for example, dates back to the Seventies (Panzerschreck Review 1-4 below, plus additional link).

http://www.bouldergames.com/gamenotes/gamenotes_703.htm

http://members.cox.net/mboone5/Period_WW2.txt

Regards,

John Kettler

[Baron von Beergut]

As a compromise, maybe Hunter could release a "lite" version of CMC without an AI. This would gain him some revenue and get some folks playing and able to test it in effect. Then when or if a finished version came out these people, for a small fee, could upgrade to the full version.

[target22]

I am one of those people who learned to play these games in the 1960's in a rural area where finding opponents for any board game was hard, let alone mention a "WAR" game. I'd have no problem with the Campaign Game having no AI. If at the Point of Contact I could setup up one or both sides and give the least chalenging side to the AI would be fine (much like it can be played now). I do play online games like Day of Defeat, Red Orch, But I can come and go as I please without felling like I am comitted to finish something. By the way I think War in Europe has sold as a No AI Board Game style PC game for 2person or Solitaire Play.

Well, Hopefully it does make it without the AI, or else it may not work on this PC by the time it comes out. LOL

[DavidFields]

I go back to, barely, Tactics II (There must have been a Tactics I? I have never heard it mentioned.)

I tried to play all types of games solo--all the Avalon Hill stuff (even something as ridiculous to solo as Luftwaffe--had to plot out multiple possible Allied bombing targets, then choose randomly at some point.) I even set up the ultimate "board" version of World at War...that was exciting, even if things went no further.

But I also played human to human, granted we often faked "Fog of War"--duplicate Panzerblitz boards with a barrier and a mediator, who told you what you could see.

Now I only play solo, for the detailed reasons pointed out in a much earlier post. At first I was slightly offended about the idea that this is being a "closet gamer". But...mayber that is right. Maybe, at 48, professional, with a lady, I just want my habit to be between me and the computer. If I am going to interact with people, which I happily do, I will do it over a martini at a sushi restaurant--and likely will not again make the social faux pas of discussing Russian WW2 tanks.

This may be an obvious question, but would releasing a multi-player version of CMC over the internet be the answer? The net proceeds could then fund the completion of the solo version? I probably don't understand the financial issues with internet releases....or maybe they are just "so close" to the completed release?

[Michael Dorosh]

Originally posted by Rankorian:

I go back to, barely, Tactics II (There must have been a Tactics I? I have never heard it mentioned.)

There was; I have a copy. It was reprinted for, I think, the 25th anniversary.

[FinnN]

Actually I think board games are having a bit of a resurgence these days - I know more people that play them now than ever before and there seems to be steady stream of them being released (mostly from Germany). What you don't get though are board games with obtuse and difficult to understand rules requiring weeks to play and a zillion counters on some poorly laid out hex map. Miniatures are also quite visible still for example, although I wouldn't touch them with a bargepole, there seems to be plenty of 'Games Workshop' shops floating around in this country (UK) at least. These are certainly far more widely available than historical miniatures ever were in my memory at least. On the other hand I don't know anyone that plays with historical miniatures any more, and haven't heard of anyone for ages either.

What I personally think is happening both with 'real' games such as those above and computer games is that those with accessible and clear rules/gameplay have become popular and those without have fallen by the wayside and left to niche markets. I think that games aimed at mass market aren't going to put in lots of realism because the extra effort isn't going to pay back in equivalent extra sales. On the other hand, if 'hardcore' games put more effort into having a better user interface and genuinely scalable difficulty/complexity settings then there are plenty of people out there who'd buy them. I think there's too much self-defeating "it's a niche game, anyone who buys it won't care if it's easy or hard to understand". A good (but rare) example of a game that did manage to pull this off are the Silent Hunter games, II was a surprise hit despite its bugs and II sold well too. There are lots of people out there gagging for more intellectually challenging and immersive games, but no-one seems terribly interested in providing them. I'd include CM:SF in that.

As to AI, well I think we're probably not too far away from seeing hardware come to the rescue there. In the not too distant future as multi-core processors become standard and games designed to utilise them from the get-go start appearing I think brute force will start delivering where current games can't with the computer power being split around doing everything.

Have fun

Finn

[Lt Bull]

Originally posted by Peterk:

Attempting to mathematically solve a game and simply providing an opponent which is capable of delivering a decent challenge are two different cans of worms.

Not really. There is NOTHING simple about "simply providing an opponent which is capable of delivering a decent challenge". Both of them are based on an understanding of mathematcial models to "describe" a game (a language that CPUs are good at understanding), though in the former case, it just happens that the game lends itself well to being 100% accurtately modelled using mathematics (step #1) and to also then have a convergent solution (step #2).

Poor CPU opponent coding comes from a poorly conducted step #1 (mathematical model doesn't describe the game enough) which in turn makes any work done in step #2 redundant, or from a porrly conducted step #2.

Originally posted by FinnN:

As to AI, well I think we're probably not too far away from seeing hardware come to the rescue there. In the not too distant future as multi-core processors become standard and games designed to utilise them from the get-go start appearing I think brute force will start delivering where current games can't with the computer power being split around doing everything.

Haha! If only that were true. Do not look towards hardware as a saviour of deficiencies in CPU/AI opponent programming in games. What is lacking is an ability, skill and understanding at the software level to mathematically model the ever increasing complexity of games and to then code this is in a way that can be manipulated and utilised by the CPU.

What you are saying is equivalent to saying the solution to all the poor car drivers on the road is to give them the fastest most modern cars. Hardware is nothing without software. Junk software with great hardware just increases the rate at which junk data is output.

[FinnN]

I have to disagree to an extent, if you can find a better example of something that could be better handled by multiple processes than things like pathfinding and LOS calculations then I'd love to know. In games like the combat mission series those are going to be two of the biggest sinks of CPU time and all you need to do is pop over to the CM:SF forum to see what the result of the abstractions made has been.

There are two ways to get a solution to a game - some sort of elegant model, and simply taking a sledgehammer and crunching some numbers. In reality both will always be required, but as your sledgehammer gets bigger you get more options when preparing your model as more and more things become computationally feasible. Unless of course you think that a really elegant model would have got an army from Medieval Total War II walking around in realtime on a machine from 20 years ago...

Have fun

Finn

[Peterk]

CMBB could have had a dramatically better AI without any hardware improvements necessary.

I mean, deciding if the computer needs to take a flag or not for the win is not rocket science but CMBB doesn`t do it.

Not leading attacks with the HQ elements. Preserving HQ elements. Keeping sub units close to their HQs. These are all things that don`t need more powerful processors, but CMBB doesn`t really do them well or at all.

I think the will of the programmers to provide a challenging AI is the biggest obstacle not hardware.

[Michael Dorosh]

Originally posted by FinnN:

Actually I think board games are having a bit of a resurgence these days - I know more people that play them now than ever before and there seems to be steady stream of them being released (mostly from Germany).

I have been doing a survey of a limited niche of games - board wargames, tactical level (units represent platoons, squads or single men), 20th Century subject matter - and have found that the number of titles released since 2000 exceeds the number of titles introduced in the first decade of tactical board wargaming (1969-1979). Advanced Squad Leader currently has 4000+ published scenarios both official and unoffical, 13 official modules, not counting two "deluxe" modules, three starter kit modules and a half dozen "historical" modules. Their fan base is hard core and rabid, but definitely there, and spending money all the time. There are also many companies producing third party products from scenarios to entire modules with die-cut counters and mapsheets.

The other big series appears to be Panzer Grenadier, with several titles in print now. Not sure of the game's popularity but apparently enough to keep these many titles in print.

It's not like the early days were you had two or three companies competing - especially now with desktop publishing, anyone can produce a professional looking set of scenario cards from their home office and sell them on ebay.

What you don't get though are board games with obtuse and difficult to understand rules requiring weeks to play and a zillion counters on some poorly laid out hex map.

The main tactical series are ASL, PG and another I didn't mention, Advanced Tobruk System, an offshoot of the original Tobruk which actually predates Squad Leader (1976 vs. 1977).

Miniatures are also quite visible still for example, although I wouldn't touch them with a bargepole, there seems to be plenty of 'Games Workshop' shops floating around in this country (UK) at least. These are certainly far more widely available than historical miniatures ever were in my memory at least. On the other hand I don't know anyone that plays with historical miniatures any more, and haven't heard of anyone for ages either.

What is the difference between "miniatures" and "historical miniatures"?

What I personally think is happening both with 'real' games such as those above and computer games is that those with accessible and clear rules/gameplay have become popular and those without have fallen by the wayside and left to niche markets. I think that games aimed at mass market aren't going to put in lots of realism because the extra effort isn't going to pay back in equivalent extra sales.

This I can agree with - Combat Commander seems to be more commercially popular than ASL etc. - you can teach it to your kids in an evening. I think even the "expert" ASL players still need a rulebook beside them to consult every second turn.

On the other hand, if 'hardcore' games put more effort into having a better user interface and genuinely scalable difficulty/complexity settings then there are plenty of people out there who'd buy them. I think there's too much self-defeating "it's a niche game, anyone who buys it won't care if it's easy or hard to understand". A good (but rare) example of a game that did manage to pull this off are the Silent Hunter games, II was a surprise hit despite its bugs and II sold well too. There are lots of people out there gagging for more intellectually challenging and immersive games, but no-one seems terribly interested in providing them. I'd include CM:SF in that.

Depends on the level of intellectual challenge you're describing - which isn't mutually inclusive with being immersive.

As to AI, well I think we're probably not too far away from seeing hardware come to the rescue there. In the not too distant future as multi-core processors become standard and games designed to utilise them from the get-go start appearing I think brute force will start delivering where current games can't with the computer power being split around doing everything.

I agree with Peterk that we wouldn't need more hardware to see a decent AI developed for the CMX1 line. I think it was unfortunate that so many people demanded 1:1 representation and that BF.C felt it was necessary to include. THAT was the requirement for more computer horsepower. CMX1 was godawful to look at, but the core game itself was an excellent translation of 30 years of tactical wargaming into one product - traces of everything from PanzerBlitz to ASL and everything in between can be found in CM:BB. Unfortunately, the decision has been made that abstraction is somehow bad and that the "mass market" needs to be chased.

Having a game capable of playing out on a 4 square kilometre map with every piece of equipment that was available on the eastern front is a huge bonus. We're told we're not going back to that. Having an operational layer, even a rudimentary one like CM:BB operations, was a welcome addition; we're told that these never worked and not to expect them in future. CM:C will be a welcome addition to CM:BB, but quite possibly won't be possible as an addition to CMX2.

So instead of updating the graphics for CMX1 and providing a powerful editor such as CM:SF's for terrain modelling, we instead see a different direction in design philosophy which guarantees more and more computer horsepower hits. I guess we'll see how it works out when we come to the Second World War titles. I'd give up one or two telephone poles that add nothing to the decision making tree if it meant gaining something back of the "old" CM.

Which brings us back to simplicity - if that is the trend in board wargaming, surely it must be in computer gaming as well? I remember my manual for Falcon 4.0; was larger than many metropolitan phone books IIRC. It didn't stay on my hard drive long, but I played Knights of the Air until my computers no longer had 3.5" disk drives from which to install it.

[bitchen frizzy]

Originally posted by Michael Dorosh:

So instead of updating the graphics for CMX1 and providing a powerful editor such as CM:SF's for terrain modelling, we instead see a different direction in design philosophy which guarantees more and more computer horsepower hits. I guess we'll see how it works out when we come to the Second World War titles. I'd give up one or two telephone poles that add nothing to the decision making tree if it meant gaining something back of the "old" CM.

I'm optimistic that computers powerful enough to do both will become affordable in our lifetimes. Then somebody will pull it all together into one game.

Which brings us back to simplicity - if that is the trend in board wargaming, surely it must be in computer gaming as well? I remember my manual for Falcon 4.0; was larger than many metropolitan phone books IIRC. It didn't stay on my hard drive long, but I played Knights of the Air until my computers no longer had 3.5" disk drives from which to install it.

Hardware sophistication can bring simplicity to intricately detailed games - if CMx1 had been a boardgame, its rulebook would have been immense, but 99% of the "rules" were handled by the AI. It can only get better in the future, and I believe that's what everyone from hardcore wargamer grogs to twitch kiddies wants.

[FinnN]

Originally posted by Michael Dorosh:

[QB]

In terms of absolute number of modules, that may be - but I'm talking about 'mainstream' (I use the term loosely) shops rather than mail order or e-bay. Old-style war board games are still there, but their relative footprint to the other types of game is really small.

Like I said, they're around but relative to other types of board game nowhere near as popular.

Space Marines, orcs and trolls, etc, vs well anything historical.

I agree to an extent, but I think it's the change in the way the calculations have been done to accommodate realtime rather than the 1:1 representation. The nice thing about CM1 games is that the level of abstraction used matched what you were seeing. In CMSF the things that are abstracted and the things that aren't aren't distinguishable.

Agreed - although personally I feel that CM:C has missed the boat. To date the best match I can think of between a strategic level game and a tactical level game was close combat 2. Talk about campaigns in flight sims and people will always go on about how good falcon 4 and red baron 3d was. A fairly big part of the success of the current silent hunter games is the dynamic campaign. My guess is that the campaign side of things rarely gets more than a passing (and usually inaccurate) comment in reviews and as such doesn't get the deserved attention from big publishers. Quite why it's, seemingly, low priority for small publishers too I don't know as I'm sure it would help to maintain sales over a longer period.

Upping the graphics certainly puts more strain a computer, but that's not at all what I was talking about. As computer horsepower increases simpler models can be used for AI but resolved in a fraction of the time.

It certainly is. Games themselves aren't necessarily simpler though - their interfaces are. That goes for board games and computer games. To me new games that come across as old fashioned aren't complex ones, or even ones with below average graphics - they're ones where the user interface (and associated elements like tutorials) has clearly seen little polish. An old fashioned board game (on the whole) would have badly written

rules, few or no diagrams - whereas newer ones I've seen have been much more clearly written (this includes newer editions of old games), better illustrated and often have multiple copies of the relevant pages for each player.

Have fun

Finn

[FinnN]

Originally posted by bitchen frizzy:

Hardware sophistication can bring simplicity to intricately detailed games - if CMx1 had been a boardgame, its rulebook would have been immense, but 99% of the "rules" were handled by the AI. It can only get better in the future, and I believe that's what everyone from hardcore wargamer grogs to twitch kiddies wants.

Agree 100%.

Have fun

Finn

[C'Rogers]

On the subject of boardgames, one of the key reasons they have had a resurgence is simply the cost involved. Design a bad board game and you have lost your time and maybe a little bit of money. Thus if your interest is niche you can "easily" pursue making products for it.

On the other hand a computer game must sell well to cover costs. Board game makers can get a second chance, a young computer company will not.

[rammer4250]

posted by Lt Bull:

1. People buy the game in basically the same numbers or more than other similar themed CPU opponent-centric release

CONCLUSION: You have just proven that if people like "the game" within your computer game, they will buy it and play it (by finding opponents to play) REGARDLESS of whether it has a CPU opponent. Just playing your game is so good, they are prepared to "forgive you " for not releasing it with a CPU opponent. You have basically achieved in much less time and at much less cost (and therefore at much higher profit) what other CPU opponent-centric game developers aspire to do. """"

Lt Bull

You are so totally wrong. I will never buy a game, wargame or otherwise, that doesn't have at least a playable AI. Once the game states multiplayer only I will not even look further. And as Michael says, I am in the majority. For now anyway.

Rammer

Oh yeah, forgot, back into the closet I go.

[ September 02, 2007, 05:45 AM: Message edited by: rammer4250 ]

[Lt Bull]

Originally posted by rammer4250:

posted by Lt Bull:

1. People buy the game in basically the same numbers or more than other similar themed CPU opponent-centric release

CONCLUSION: You have just proven that if people like "the game" within your computer game, they will buy it and play it (by finding opponents to play) REGARDLESS of whether it has a CPU opponent. Just playing your game is so good, they are prepared to "forgive you " for not releasing it with a CPU opponent. You have basically achieved in much less time and at much less cost (and therefore at much higher profit) what other CPU opponent-centric game developers aspire to do. """"

Lt Bull

You are so totally wrong. I will never buy a game, wargame or otherwise, that doesn't have at least a playable AI. Once the game states multiplayer only I will not even look further. And as Michael says, I am in the majority. For now anyway.

Rammer

Hello Rammer,

Thanks for your comments but I don't think any of what you said has any relevance in refuting my conclusions quoted above if people were to buy the game in basically the same numbers or more than other similar themed CPU opponent-centric release.

What conclusions would you instead suggest if people were to buy the game in basically the same numbers or more than other similar themed CPU opponent-centric release?

[rammer4250]

Lt Bull

I guess you would be right. But I don't believe you are correct now. I love all the Combat Mission titles yet I won't buy something just because Battlefront produced it. Just for kicks I downloaded the Theatre of War demo but I am not going to purchase it just because it's from battlefront. I hate RTS and all the pretty graphics is not going to change my opinion no matter how good the game is. I have tinkered with war games of all types (AH boardgames,squad leader and ASL, miniatures, and computer games) since I was 12 or 13. I am now 48. My main interest is military history (mostly World War II). I have never played nor intend to play against a human player for a variety of reasons. Mainly I have no interest in competiton. I want to play at my own pace and not even finish if I don't want to. Yeah the AI is pretty bad at times but at least I don't have to move for the other side like I had to with the boardgames. As far as I am concerned the greatest {sims} I have played are Falcon 4.0 and the Combat Mission titles. My relationship will definitely end with Battlefront if they ever decided to go strictly with RTS and/or multiplayer only. If Battlefront thought they would make more money with what you suggest and went that way God Bless them because I am sure that they are in business to make money. Doing what they love is just a bonus.

I apologize if I have totally misunderstood your statements.

Rammer

[Lt Bull]

Rammer4250,

I think you have "objected" to what I have said in the wrong way and have confused what you think I am wrong about. Maybe you just quoted the wrong part of my post, becasue none of what you quoted me saying has any relevance to what I think you are trying to say.

I think I know what you are trying to say: that a computer game without a CPU opponent will never sell, stating your own personal views on the matter as an example.

But I never neccesarily said it WOULD (though I think it could if done correctly but that is besides the point here). The point I was making in that quote was that if it DID succeed (a hypothetical) then you COULD conclude the following:

CONCLUSION: You have just proven that if people like "the game" within your computer game, they will buy it and play it (by finding opponents to play) REGARDLESS of whether it has a CPU opponent. Just playing your game is so good, they are prepared to "forgive you " for not releasing it with a CPU opponent. You have basically achieved in much less time and at much less cost (and therefore at much higher profit) what other CPU opponent-centric game developers aspire to do.

[FinnN]

I agree 100% with Rammer - the vast majority of people won't buy a game that's 100% multiplayer. Even MMOGs have a significant AI component. The handful of exceptions to that are some action games or fast paced simple strategy games - and how many of those has sold even 1% of what, say, CMBO sold? Maybe you could save 20% or 30% of the development costs but you'd lose 99% of the buyers. That means a significantly smaller profit.

Your argument seems a bit circular, if it was successful (it sold well) then it'd be successful. Or am I missing the argument here?

Have fun

Finn

[Lt Bull]

FinnN,

I agree the vast majority of people won't buy a game that is 100% multiplayer, but commercial success does not need the vast majoirty of people buying your game.

You must agree that MP gameplay (not only for PCs, but for your console games) is becoming increasingly more and more popularity and important to gaming with the advent of the internt/broadband. You only need to look at the heavy investment gone into creating a MP network for both the XBox and PS consoles to see this. MP gameplay is becoming more and more important in gaming and will continue to be so as more and more people get on the internet/get broadband etc.

The potential of a "MP centric" game today is nothing like what it was when CMBO firt came out. The potential of MP only games to succeed will only increase with time.

You suggest 20-30% of development costs if you eliminate the need to code a CPU opponent. Depending on the kind of game you may be making, this value could be much higher. As it happens, the kinds of games that appeal to and are of particular focus for discussion to the peope who loiter at these forums tend to be more complex, detailed and stratgeic/tactical than most other games out there. By their very nature these games present some fo the toughest challenegs to game developers as far as coding a competent CPU opponent.

I recently rediscovered an old post of mine that i had forgotten I had written:

Victory conditons: possibilities + the burden of coding a CPU opponent

I discussed the "burden of coding a CPU opponent" and how it can prevent a game from featuring otherwise cool game features HUMANS woudl appreciate. I used the simple example of how "victory conditions" in agame that are easy for humans to deal with and rationalise might not ever make their way into a game simply becasue it woudl be just too hard to code a CPU opponent to deal with them . It is interesting to note however that many of the things I was suggesting that could be done as far as victory conditions go are in some ways visisble in CMSF.

I also watched a very interesting video interview of the Studio Director from Creative Assemblies (makers of the Total War series), Mike Simpson, who was the inspiration behind the series and has been right there from the start . In a way, CMC is trying to be a game like the Total War series, in that you almost have two games in one, a tactical game layered and nestled within a strategic game.

Even thugh the game concept has been around for many years now, when asked why there aren't too may games that have tried to emulate the success of the Total War game concept (the two games in one), he says the answer is simple: when "game developers take a look at the problem, they run away screaming" as it really is a steep hill to climb. Getting the men on the battlefield is easy he says, any good game developer could do it without too many problems. It is the "behind the scenes" stuff, in particular the AI, and then having to bring it all together in essentially two games, the campaign (strategic) and battle (tactical) games which are virtually standalone games in their own right. So most developers decide that for the ammount of effort to catch up, they could have produced two or three games and end up deciding it is not worth their time.

Throughout the interview, I could not help but make comparisons to how Creative Assemlies managed, developed and grew their "game concept" and their vision going forward and how BFC have (or haven't depending on how you look at it).

He states the biggest challenge in growing a game like the Total War series comes down to resources: having enough really good people with the right skills to implememnt all the stuff in their vision for what they want in the game. This certainly has not been an area in which BFC have indicated they have had any need to address as they still have one guy doing all the hard coding stuff.

Was also intersting to hear his take on the future. He beleives they could keep making Total War series games forever as long as they keep managing to put something new and intersting in each time. Finding a new historical time period was not a problem. However if you keep producing the same game with different content he says, people will get bored with it and in the past that has killed off other game series of games. He says at CA they don't do that. They try to implement a good solid gameplay feature that is above and beyond anything that has come before into every release, even the add ons. The CMx1 series really didn't do that.

He also talked about "evolutionary" changes (same engine but new features/enhancements)and "revolutionary" changes (throw out old engine, make a new one but keep all the stuff that worked). BFC seems to have thrown out a lot of the things that worked well in CMx1 when makign CMSF.