Matrix Games Forums

ORIGINAL: Shannon V. OKeets

I expect to be finished by the end of September, but Matrix Games needs a few weeks after I submit the finalized code to package the game for distribution.

And the good news is: I can postpone selecting a new computer, and also larger monitors to display larger segments of the map simultaniously.

Since my old comuter comply with my present requirements, I'll just wait a little more, I gues I'm saving a lot of money due to this [:D]. But I'm still looking forward to the release [8|]

ORIGINAL: Shannon V. OKeets

ORIGINAL: ptey

About the postponement, I think the writing was on the wall atleast a month ago. Im glad you didnt choose to release a game almost unplayable due to bugs Steve.

Also, 500 hours for the AI seems low. I suspect it wont be competitive against anyone but a novice. Im fine with that though. Are you willing to share what your aim is here Steve?

I have a lot to do.[8|] My aim is to get things to work.[:)]

Steve it is good to get everything perfect but it is not practical with the scope of this game. I dont know how you got this far without help. Maybe you have put too much on your back trying to correct every little bug in the program, I am not saying release it too soon I am saying that when it is released rely on the good people here who really know this game to find minor program errors, when they post these errors you will know right where to go to correct them, your truly tough undertaking is to keep as close as you can to the true WIF board game and bring all those massive options to MWIF as compared to making up rules as you go along as many times it has been done in other war games by many programers, I believe that the best playtesters will be the people who buy this game [no knock against current playtesters] if they try different options and strategies for a week or so for testing purposes instead of jumping right into the game for real. As far as the AI goes I believe it will improve with every new update that you and matrix put out. As long as I am convinced that you and Matrix will stand behind this game for a good period of time and make corrections when needed then none of us has anything to fret about.

Willy

ORIGINAL: Shannon V. OKeets

ORIGINAL: bfontes

It looks like Microsoft might be able to get Windows 7 out before this game is ready. So, will the game come with the "Windows 7 Ready" seal of approval?

I am not sure what that means.[&:]

MS is claiming they will be ready for prime time by October. Bill Gates is now telling us to look for the "Windows 7 Compatible" logo on any new software we buy.

" If you’re buying a new device or application, look for the "Compatible with Windows 7" logo. These products have passed tests that Microsoft designed to install readily and run reliably with Windows 7. "

Only people happy to beta test will upgrade to Windows 7 for the next couple of years.[;)]

ORIGINAL: Shannon V. OKeets

ORIGINAL: Greywolf
Hum, basically what you want is a way to play a saved games into another setting than what it started.

I think there is no real trouble having a player taking on the spot of a AI.

On the reverse IIRC , and I think Steve will correct me otherwise, the AiO use extensive save of his decisionary process not to be taken into a loop. It is clear that taking mid-turn the place of a human player wont probably be possible. Taking back the situation from turn start will be hard and I think not possible because the AI is set to choose a strategy from start not to interpret the map to recognize what the previous player had in mind and started...

...

Yes.

...

My point here is that a game position has an underlying structure that is a function of the logic the players are using. The AIO understands its own logic, but human decisions can be quite bewildering - not only to the AIO but to other players.[;)]

...

Thanks to the both of you for these detailed explanations

Alain

ORIGINAL: sapper_astro

Only people happy to beta test will upgrade to Windows 7 for the next couple of years.[;)]

Anyway even when Microsoft release a new product it's still in Beta. For a Windows operating system, it's best to wait one year, even 2 years, before using it. I suggest wait for the first service pack.

Re the AIO, many of the people making predictions seem to be relatively new here and perhaps haven't read the old AI threads. A tremendous amount of the fundamental work has been done and it seems to me that it's the actual implementation that still requires a lot of work. Given the quality of the work Steve has done on the rest of the project, I think he deserves the benefit of the doubt on this.

You could compare it to a car production line. Each of the assemblies is very complicated and require time to complete but when they come together you suddenly get a car very quickly. Perhaps Steve could give his opinion on this.

There's absolutely no point in whingeing about MWiF not needing an AI because it was decided long ago that it is going to have one, most potential players want one and to sell a semi-obscure wargame that you would either have to play against yourself or need to find online opponents would be commercial suicide.

A decent AIO will generate enough sales to produce a solid mass of MWiFers so that anyone who is after more of a challenge will be able to find human opponents. Who knows? The AIO may turn out to be much better than many assume anyway. I recommend a look at the old AI threads, especially the major power specific ones.

Cheers, Neilster

Well, Neil, all I can say is the AIO hasn't executed one decision yet. And humans have problems making decisions re strategy and tactics in this game. I'm concerned. No one has been able to develope an AI worth a darn to date. If you know of a wargame with a good one, tell us. And Steve can't just make all the AI units have more speed or power or give them more "hit Points" like other systems do. I guarantee you this won't be done by September because the AI won't be done. And don't tell me I'm just a nay saying troll, because I do support Steve and this project. Reality must be acknowledged on both sides.  

Steve you are qualified to be in any branch of the US Militray of your choosing now. After all, you have done so much with so little for so long you now can do anything with nothing. [:D]  We appreciate the hard work and effort you have put forth and we will wait until the game is finished. It has been a few years since I first said to our Troll that it will be done when it is done. So, if this takes a few more months or years I can wait. I just hope your bank account holds up long enough for you to finish this and get paid for your efforts.

ORIGINAL: HansHafen
No one has been able to develope an AI worth a darn to date. If you know of a wargame with a good one, tell us. And

IBM made a chess AI that could give a grandmaster a run for his money. (Cost them a bundle)

There was a PC Game done 10 years or so called OGRE (which was a table top game conversion). It had a great AI and was EXTREMELY hard to beat (but, of course, it was a small game.)

There are several Bridge (see my signature) games that are proficient (without cheating) ... GIB and Jack.

It can be done....

Wargames are different, in many respects. There are many more pieces and many more 'squares' than in chess.

I wish Steve luck and the best of skill.

ORIGINAL: HansHafen

Well, Neil, all I can say is the AIO hasn't executed one decision yet. And humans have problems making decisions re strategy and tactics in this game. I'm concerned. No one has been able to develope an AI worth a darn to date. If you know of a wargame with a good one, tell us. And Steve can't just make all the AI units have more speed or power or give them more "hit Points" like other systems do. I guarantee you this won't be done by September because the AI won't be done. And don't tell me I'm just a nay saying troll, because I do support Steve and this project. Reality must be acknowledged on both sides.  

I agree.

Don't forget that even the IA script parser is not yet fully written. Have you ever tried to write a compiler ? or even a parser ?
That's by itself a work that's not done in a day.

Then, you have to make all the inner tables that will support the decision process, write all the scripts for all countries (general strategies ; count at least 8 complicated such scripts) ; write all the other scripts for generic behaviors (keep troops supplied, fight air-air combat, move troops, move ships, produce, lay down produced troops...) ; that's also a lot of scripts.
Then plug these into the existing decision points in the program. You get an idea when Steve says that there are hundreds of decision points : all of these will have to be backed by some AI script, be it one line long (unlikely!)...

Then, debug and test all of this.
Loop around solving bugs in the scripts and decision process inner engine and improving them (and also changing the decision tables, which the first run will certainly not get right).

Not to forget that all of this has to be saved and restored at any point in any phase.

Wow! That's an awfull lot of work!
I'm not diminishing Steve in saying that it will likely not be completed in (the revised) schedule. I'm a professional programmer and programming architect and I would not sign for this on less than six months, possibly more. Steve working on a three eight-hour shift all by himself, it can possibly be completed within two months...

I greatly admire his work, and I really hope this project gets done and well done and I'm ready to wait two or four or six more months if necessary!

Yves

ORIGINAL: SamuraiProgrammer

ORIGINAL: HansHafen
No one has been able to develope an AI worth a darn to date. If you know of a wargame with a good one, tell us. And

IBM made a chess AI that could give a grandmaster a run for his money. (Cost them a bundle)

There was a PC Game done 10 years or so called OGRE (which was a table top game conversion). It had a great AI and was EXTREMELY hard to beat (but, of course, it was a small game.)

There are several Bridge (see my signature) games that are proficient (without cheating) ... GIB and Jack.

It can be done....

Wargames are different, in many respects. There are many more pieces and many more 'squares' than in chess.

I wish Steve luck and the best of skill.

I believe your right Samurai it can be done but very difficult to do. Avalon Hill did 3rd Reich and messed the AI up to high heaven and worse yet never did anything to improve it. I have all the Strategic Command Series and even though they are beer and pretzles games [fast games] I still enjoy them. The German AI is relentless on land and at sea. It pulverizes Poland, France, Greece, Yugoslavia and many times Russia, but it fails in two areas invasions and supply. [actually the whole game has a bad supply program Axis and Allies] Usually the first part of the invasion is well done, but if the Allies lost a lot of units then forget it, because the Allied AI would keep sending single units into France with no chance of survival. It is a very good game system but it lacks the depth of WIF or at least what I have been able to learn about WIF from the posts here. The AI moved quite a few units around and in general did a pretty good job of it. It can be DONE!
I dont want to pull a Bill Clinton I feel your pain stuff but I have empathy for the players who want to only play another human which is probably the best way to play providing both players are very capable at this game. Now they have to wait for the AI and other parts of the program to be finished, but just maybe in the long run it will be a complete game for everyone. One comment I have read which tells me the AI will be very important is that when most players are losing badly they quit the WIF board game! And one comment was never ever saw a game completed. At least with the AI you will be able to take the game to its conclusion if you want to.

Willy

ORIGINAL: HansHafen

Well, Neil, all I can say is the AIO hasn't executed one decision yet. And humans have problems making decisions re strategy and tactics in this game. I'm concerned. No one has been able to develope an AI worth a darn to date. If you know of a wargame with a good one, tell us. And Steve can't just make all the AI units have more speed or power or give them more "hit Points" like other systems do. I guarantee you this won't be done by September because the AI won't be done. And don't tell me I'm just a nay saying troll, because I do support Steve and this project. Reality must be acknowledged on both sides.  

My name isn't Neil.

TOAW 3 is a wargame with an excellent AI.

I was merely offering an alternative way of looking at the situation to all the doom and gloom. If it takes longer to develop the AIO then I don't care. If it isn't possible at all then that will be a very bad result. I prefer to remain positive. You can suit yourself.

Edit: BTW, re the AIO, didn't you say "Cool. Looks like you have alot of the work done. I await Spetember! " in another thread about 3 days ago?

Neilster

Good grief, lay off the speculation. Let the AI be developed before you decide if its good or bad. Give Steve a chance. We all know the history of IBM and its multi millions poured into big blue and successive programs. So no body really excepts an AI for that would challenge everybody every game. If the AI can challenge novices most of the time and capable players some of the time then its goal will have been extraordinarily achieved. Time will allow it to be refined. As we players want more challenge we will have to gravitate to the best AI of all, the human brain.
 
Look at it this way whiners and complainers need not apply for beta testing.

I agree. I think the goal will have been extraordinarily achieved if the AI can callenge novices SOME of the time...and I am not looking for extraordinary when it comes to an AI. Capable players need not be challenged by an AI other than what you wrote...the human brain!
C

ORIGINAL: abj9562

Good grief, lay off the speculation. Let the AI be developed before you decide if its good or bad. Give Steve a chance. We all know the history of IBM and its multi millions poured into big blue and successive programs. So no body really excepts an AI for that would challenge everybody every game. If the AI can challenge novices most of the time and capable players some of the time then its goal will have been extraordinarily achieved. Time will allow it to be refined. As we players want more challenge we will have to gravitate to the best AI of all, the human brain.

Look at it this way whiners and complainers need not apply for beta testing.

ORIGINAL: Neilster
My name isn't Neil.
Neilster

And I'm not Past, as some seem to believe.

I agree with you, old man. I'll say only this about AI in computer wargames (and I've said it so many times before, I'm sure most of you would prefer that I just shut up): if you don't have one that works to at least some competent degree, your game just ain't gonna sell.

Steve and his team are pulling a heckuva load with this undertaking. I am content to wait for a final, finished product that they are happy with releasing. With that, I grant all those fervent wishes and ... shut up.

I'm putting my 2 cents in. You don't don't see all the WiF boardgame players saying; 'gee, if I gotta wait for and pay for an AI that I'm not even going to use, I won't buy it.' Or worse yet, speculate that it won't sell at all. Why? Because that would be classified as a threat, and the WiF boardgamers are above making threats, least of all, to their own game designer. This is a game that comes with it's own following and thus, it's own set of circumstances. You AI guys really crack me up with your 'PC game expertise'. You're speculating and why? Because you want to threaten Steve into forgoing everybody but you, and concentrate all his effort into the stinking AI. I'm sick of it. Someone has to call you guys on this. You have no idea what kind of sales a game this popular will do. You just don't. I wish you'd just shut up.

ORIGINAL: macgregor

I wish you'd just shut up.

You've got an indefinite pronoun reference working there.

Steady mac!

PoE (aka ivanmoe)

ORIGINAL: pasternakski

I'm sure most of you would prefer that I just shut up): if you don't have one that works to at least some competent degree, your game just ain't gonna sell.

I kinda wish that developers had never gone down the AI path with hex/turn based games. The best AI for wargames have always struck me as hash, anyhow. I want to play against human opponents, not the computer. For that matter, one of the best reasons for porting a boardgame over to a PC is that the move offers folks who are geographically isolated from other wargamers a chance to actually play. Conversely, I really have to question why someone would PREFER playing a computer over a human.

PoE (aka ivanmoe)

ORIGINAL: macgregor

I'm putting my 2 cents in. You don't don't see all the WiF boardgame players saying; 'gee, if I gotta wait for and pay for an AI that I'm not even going to use, I won't buy it.' Or worse yet, speculate that it won't sell at all. Why? Because that would be classified as a threat, and the WiF boardgamers are above making threats, least of all, to their own game designer. This is a game that comes with it's own following and thus, it's own set of circumstances. You AI guys really crack me up with your 'PC game expertise'. You're speculating and why? Because you want to threaten Steve into forgoing everybody but you, and concentrate all his effort into the stinking AI. I'm sick of it. Someone has to call you guys on this. You have no idea what kind of sales a game this popular will do. You just don't. I wish you'd just shut up.

Yo Mac nobody told you to shut up did they, these forums are just what they are supposed to be a place to get your views across without animosity to anybodys point of view. It is not a threat to say no AI no sale, a person is just saying I can do without this game if there is no AI, that is our perogative as buyers of anything that we have to put out our dollars for. If Steve decides to go without the AI fine with me, but Steve is a little smarter than that, he knows and Matrix knows that there is not enough of you board game players to make a profit and isnt that what its all about with Matrix "make a profit".
No AI no sale is a statement of fact not a threat. I guess I will just have to go through life playing Strategic Command with a very good AI. And by the way Battlefront's sales are booming with the SC series. Oh well!

Willy