In Development

Gamers can also use this forum to chat about any game related subject, news, rumours etc.

Moderator: maddog986

User avatar
zakblood
Posts: 22722
Joined: Thu Oct 04, 2012 11:19 am

RE: In Development

Post by zakblood »

sign of the times is that most players do want eye candy and online Multiplayer more than single player, so the developers imo have given them what they have continued to keep asking for, not that i agree also, being a single player gamer myself etc, who tends to not play online in hardly any game now, well not war games anyway, just not good enough or have the time or ability etc,...

but do understand also that it's what sells, and war games as a whole compared to elsewhere eg FPS etc don't on the same scale, so in numbers terms, it's better to give player on the whole what they ask for, and from a testers point of view and forum mass reader, more ask for eye candy and Multiplayer still than AI for single player.

but saying that, imo the AI has improved over the last few years with some games, war games this is giving me at least more of a challenge than in the past, and when you get to a given age, you can remember what it used to be like etc, with V FOR VICTORY MARKET GARDEN or ARNHEM THE 'MARKET GARDEN' OPERATION as 2 examples etc, against Desert War 1940 - 1942 or lets say Campaign Series Middle East 1948-1985, or even the likes of games from the Flashpoint Campaigns Series, any of which give you a good run for your money on the AI stakes.

with the gronards types giving a real challenge on a given difficulty level WITE / WITW or WITP etc and don't even get me started on the Command Modern Air / Naval Operations series, as the AI is masterful to say the least, so there has been some large improvements on the AI side in some given games, i could name others, which for me do stand out as enjoyable in the single player side, but then again, i guess it also depends on you're skill level as well, which maybe higher than mine,... [:D]
Windows 11 Pro 64-bit (10.0, Build 22621) (22621.ni_release.220506-1250)
User avatar
MrsWargamer
Posts: 1653
Joined: Wed Jun 18, 2014 4:04 pm

RE: In Development

Post by MrsWargamer »

Hmm I couldn't care less if AI was just given the boot from future efforts. I could just as easily respond to the assertion that devs are lazy with maybe gamers that refuse to locate human adversaries are even lazier. Not saying anyone is, but, it seems a fair response.

There is NOTHING wrong with playing a hot seat game as both sides.
And the multiplayer interface here through Slitherine Group is in my view the pinnacle of human vs human wargaming.

Demanding a wargame possess an AI in my view often is more likely to damage a good wargame design instead of aid it.
Wargame, 05% of the time.
Play with Barbies 05% of the time.
Play with Legos 10% of the time.
Build models 20% of the time
Shopping 60% of the time.
Exlains why I buy em more than I play em.
User avatar
rico21
Posts: 3034
Joined: Fri Mar 11, 2016 8:05 am

RE: In Development

Post by rico21 »

In the future, we will buy a game designed for multiplayer without AI.
To play alone, we will purchase an external program that we will link to this game (or another!).
What is certain is that this "miracle" program will not be developed by the video game industry.
Prepare you to play games in no english language.[:D]
User avatar
IainMcNeil
Posts: 2784
Joined: Tue Oct 26, 2004 10:01 am
Location: London
Contact:

RE: In Development

Post by IainMcNeil »

ORIGINAL: vonRocko

Sorry, no insult was intended. I understand it is complicated, but this seems to be the only thing that hasn't improved over the years.

Sorry for over reacting - I'm a little touchy when people use the word lazy in the same sentence as developer. Some of the comments on the Steam forums make my head boil. :)

Its true - sadly AI currently has to be written for the specific set of games rules you've written. Change the game rules and the AI has to be modified to cope with it. So each new game generally requires AI from scratch. Until AI becomes good enough to "learn" itself then I suspect game AI isn't going to make much progress. We're hoping to experiment with something like this in a future release but we'll have to see how it goes...
Iain McNeil
Director
Matrix Games
User avatar
Lobster
Posts: 5300
Joined: Thu Aug 08, 2013 2:12 pm
Location: Third rock from the Sun.

RE: In Development

Post by Lobster »

If you don't make something and then constantly improve that something then that something will never become better. The only way a game's 'AI', or any gaming 'AI' in general, will become better is if it is used or included at all. That will in turn require the money to make it better. If people stop buying games with anything but multiplayer with no AI then the game developers will stop improving gaming AI because the money isn't there. One thing I can depend on. The AI will not drop a game because it is losing.
http://www.operationbarbarossa.net/

Two things are infinite: the universe and human stupidity and I’m not sure about the universe-Einstein

Q: What do you call a boomerang that doesn’t come back?
A: A stick.
User avatar
Mobeer
Posts: 664
Joined: Wed Jan 17, 2007 7:59 pm
Contact:

RE: In Development

Post by Mobeer »

ORIGINAL: IainMcNeil
...
sadly AI currently has to be written for the specific set of games rules you've written
...

This approach is part of the problem; the AI needs to be written whilst the game rules are being developed, not as an after thought. Developing the AI alongside the rules means that the rules can be changed if the AI can't manage them. Leave the AI as an afterthought and it's bound to have problems.
User avatar
MrsWargamer
Posts: 1653
Joined: Wed Jun 18, 2014 4:04 pm

RE: In Development

Post by MrsWargamer »

"One thing I can depend on. The AI will not drop a game because it is losing."

There's not much point playing an AI game once the AI has effectively thrown the game with an outrageously stupid stunt even a blithering idiot 6-year-old wouldn't do.

I think the venerable, awesome Steel Panthers's Long Campaigns are an utter waste of time. I've built forces that AI simply can't cope with. It takes no effort. It takes no skill. It's like kicking puppies playing the AI in the Long Campaigns.

The AI is an overrated necessity. I'll just say it, we ONLY have AIs in our wargames to placate whiny wargamers that refuse to actually do something to find a human adversary, or who simply refuse to accept playing the game vs themselves in hotseat mode.

AI is a waste of money.

I have NEVER defeated the AI even once and felt satisfaction at winning.

The only victories I have ever savoured were vs a human.
Wargame, 05% of the time.
Play with Barbies 05% of the time.
Play with Legos 10% of the time.
Build models 20% of the time
Shopping 60% of the time.
Exlains why I buy em more than I play em.
User avatar
loki100
Posts: 11699
Joined: Sat Oct 20, 2012 12:38 pm
Location: Utlima Thule

RE: In Development

Post by loki100 »

ORIGINAL: Mobeer
ORIGINAL: IainMcNeil
...
sadly AI currently has to be written for the specific set of games rules you've written
...

This approach is part of the problem; the AI needs to be written whilst the game rules are being developed, not as an after thought. Developing the AI alongside the rules means that the rules can be changed if the AI can't manage them. Leave the AI as an afterthought and it's bound to have problems.

He didn't say that.

The comment is that each game needs an AI built for its rules so at the moment there is not a generic 'Slitherine AI' that is adapted to each game but that each game needs its own created against that games' rule set
User avatar
IainMcNeil
Posts: 2784
Joined: Tue Oct 26, 2004 10:01 am
Location: London
Contact:

RE: In Development

Post by IainMcNeil »

ORIGINAL: Mobeer
ORIGINAL: IainMcNeil
...
sadly AI currently has to be written for the specific set of games rules you've written
...

This approach is part of the problem; the AI needs to be written whilst the game rules are being developed, not as an after thought. Developing the AI alongside the rules means that the rules can be changed if the AI can't manage them. Leave the AI as an afterthought and it's bound to have problems.

Obviously we don't make changes the AI cant cope and a lot of great ideas get shelved because the AI can't cope with it, but even changing stats and balancing can impact the AI. Its really is a pain in the ass to get right! Fixing a bug can affect the AI. basically every single code and data change you make has a chance of impacting the AI. The main issue is though that you cant use the AI from the previous game and build on it. Whereas you can use the engine code or multiplayer code, so they continually get better but AI stays where it was.
Iain McNeil
Director
Matrix Games
vonRocko
Posts: 1448
Joined: Tue Nov 04, 2008 12:05 pm

RE: In Development

Post by vonRocko »

"I'll just say it, we ONLY have AIs in our wargames to placate whiny wargamers that refuse to actually do something to find a human adversary, or who simply refuse to accept playing the game vs themselves in hotseat mode"

The whole reason I play pc games is not needing a human opponent or playing "hotseat". I have boardgames if I want that.
User avatar
jack54
Posts: 1428
Joined: Wed Jul 18, 2007 8:25 pm
Location: East Tennessee

RE: In Development

Post by jack54 »

ORIGINAL: vonRocko


The whole reason I play pc games is not needing a human opponent or playing "hotseat". I have boardgames if I want that.

+1... exactly the way I feel.
Avatar: Me borrowing Albert Ball's Nieuport 17

Counter from Bloody April by Terry Simo (GMT)
RichG
Posts: 121
Joined: Sat Nov 28, 2015 3:29 am

RE: In Development

Post by RichG »

ORIGINAL: berto

Not saying AI is unimportant; it just seems that to most casual players (where most of the sales are) having a decent AI is farther down the wish list.

The recent WargamesDesignStudio survey seems to suggest otherwise, where improved AI was at the top of the improvements list and also pretty much 80% of players play solo and never or rarely multiplayer. These results may of course be biased towards the JTS games but I'd suggest they are relevant across most grog games.

User avatar
MrsWargamer
Posts: 1653
Joined: Wed Jun 18, 2014 4:04 pm

RE: In Development

Post by MrsWargamer »

Hehehehe

MOST seems to be an interesting label.

Perhaps if MOST players actually played humans, there would be MORE wargamers, making the hobby MORE mainstream, making it easier to market MORE products.

If MOST wargames are sold to MOSTly solo wargamers, who MOSTly expect an AI even when MOST developers have tried their best to explain it is MOSTly not that easy, then perhaps they should MOSTly finally except it that MOST of their wargames are never going to have an AI which MOST of the wargamers who like a real challenge, will MOSTLY be interested in supporting.

The main reason I don't play most of my PC title wargames isn't the AI situation though. It's the interface. As big a footprint as my board game the Longest Day has, I can look down on it and see all of it just by looking down. I can move any counter, simply picking it up and moving it. I can control the game's many functions without needing a lot of menus. It's all down between my ears :) Just some dice, and a few charts. And once it's set up, it's really just the one chart that controls the action.

If they made the AI fully 10x times as good, it might just barely manage to finally be about as good as I was, when I was 13.
Wargame, 05% of the time.
Play with Barbies 05% of the time.
Play with Legos 10% of the time.
Build models 20% of the time
Shopping 60% of the time.
Exlains why I buy em more than I play em.
User avatar
Curtis Lemay
Posts: 13846
Joined: Fri Sep 17, 2004 3:12 pm
Location: Houston, TX

RE: In Development

Post by Curtis Lemay »

If a poll comes out saying that what players really want is the game to levitate their a$$es so they won't get saddle sores, I still wouldn't expect that feature anytime soon. Wargames are constrained by what is actually possible.

And the sad fact is that the AI is a black hole, down which an almost unlimited amount of coding effort can be pored without any discernible effects. Games can and will have an AI, but at the level of about a three-year old. Far better to focus that coding effort on game features themselves - where actual benefits can accrue.
My TOAW web site:

Bob Cross's TOAW Site
User avatar
JReb
Posts: 298
Joined: Wed Sep 18, 2002 6:24 pm

RE: In Development

Post by JReb »

ORIGINAL: MrsWargamer

Hehehehe

MOST seems to be an interesting label.

Perhaps if MOST players actually played humans, there would be MORE wargamers, making the hobby MORE mainstream, making it easier to market MORE products.

If MOST wargames are sold to MOSTly solo wargamers, who MOSTly expect an AI even when MOST developers have tried their best to explain it is MOSTly not that easy, then perhaps they should MOSTly finally except it that MOST of their wargames are never going to have an AI which MOST of the wargamers who like a real challenge, will MOSTLY be interested in supporting.

The main reason I don't play most of my PC title wargames isn't the AI situation though. It's the interface. As big a footprint as my board game the Longest Day has, I can look down on it and see all of it just by looking down. I can move any counter, simply picking it up and moving it. I can control the game's many functions without needing a lot of menus. It's all down between my ears :) Just some dice, and a few charts. And once it's set up, it's really just the one chart that controls the action.

If they made the AI fully 10x times as good, it might just barely manage to finally be about as good as I was, when I was 13.

Don't hold back. Tell us how you really feel!! [:D]

Good points made by all. I think we have seen about all the AI development we will see in our generation unless there is a paradigm leap in computing technology and programming. Its as good as its going to get unless some major event changes the paradigm.

Having said that, if Slith/Matrix does not have AI in their games their sales would plummet significantly. I do not consider myself 'whiny' for playing against the AI. [:D] That was pretty funny though.

When I get time to play, its a personal hobby of mine that I enjoy on my own. I don't do it for the competition and have no desire to play PBEM or multi-player. Its just that quiet time I get to have to myself that I enjoy most. Its my form of escape, that or a good book. I've given up on television altogether! [8|]

The thing with AI is that if it is too good people will complain its too hard or its cheating, if its too easy....well, we see what the comments are then. So its definitely a double-edged sword with no real solution at this time in history.
My shrink says I have anger management and conflict resolution issues....and I'LL FIGHT ANYBODY THAT DISAGREES!
RichG
Posts: 121
Joined: Sat Nov 28, 2015 3:29 am

RE: In Development

Post by RichG »

I think Rico's last response might be a pointer to the way things could progress.

With the current rise in machine learning technology it's not unfeasible to imagine a computer opponent that learns to play a good game just by allowing it to observe real humans playing for a learning period. The problem would be making sure it didn't become too "gamey", so the programmers effort might end up having to deal with 'house rules' and the like to stop a machine opponent from taking advantage of the system in ways real players wouldn't.
User avatar
berto
Posts: 21461
Joined: Wed Mar 13, 2002 1:15 am
Location: metro Chicago, Illinois, USA
Contact:

RE: In Development

Post by berto »

ORIGINAL: RichG
ORIGINAL: berto

Not saying AI is unimportant; it just seems that to most casual players (where most of the sales are) having a decent AI is farther down the wish list.
The recent WargamesDesignStudio survey seems to suggest otherwise, where improved AI was at the top of the improvements list and also pretty much 80% of players play solo and never or rarely multiplayer. These results may of course be biased towards the JTS games but I'd suggest they are relevant across most grog games.
For grognards like the WDS/JTS audience, yes, AI is highly or even most important. But grognards are few in number. Developing to please the grognards only rarely pays. The real money is in appealing to casual gamers. Developers will usually follow the money.
Campaign Series Legion https://cslegion.com/
Campaign Series Lead Coder https://www.matrixgames.com/forums/view ... hp?f=10167
Panzer Campaigns, Panzer Battles Lead Coder https://wargameds.com
User avatar
berto
Posts: 21461
Joined: Wed Mar 13, 2002 1:15 am
Location: metro Chicago, Illinois, USA
Contact:

RE: In Development

Post by berto »

ORIGINAL: JReb

The thing with AI is that if it is too good people will complain its too hard or its cheating, if its too easy...
And/or will complain that the AI turn takes too long. That the game is too high priced (to compensate for the extra development time and effort). And/or endlessly delayed (again due to the lengthy development time). [8|]
Campaign Series Legion https://cslegion.com/
Campaign Series Lead Coder https://www.matrixgames.com/forums/view ... hp?f=10167
Panzer Campaigns, Panzer Battles Lead Coder https://wargameds.com
User avatar
rico21
Posts: 3034
Joined: Fri Mar 11, 2016 8:05 am

RE: In Development

Post by rico21 »

ORIGINAL: RichG

I think Rico's last response might be a pointer to the way things could progress.

With the current rise in machine learning technology it's not unfeasible to imagine a computer opponent that learns to play a good game just by allowing it to observe real humans playing for a learning period. The problem would be making sure it didn't become too "gamey", so the programmers effort might end up having to deal with 'house rules' and the like to stop a machine opponent from taking advantage of the system in ways real players wouldn't.
To be exact, I was thinking of the current economic race on AI and I am not sure that the historic leaders are the winners of tomorrow.[:)]
User avatar
rsallen64
Posts: 203
Joined: Mon Jun 15, 2009 6:20 pm
Location: Olympia, WA

RE: In Development

Post by rsallen64 »

AI is important to me, and it's not because I don't want to invest the time to find a human opponent. I started playing board war-games in the 1970's. I've been at this awhile. It's not easy to find compatible players. It's not easy to find the time to play against an opponent when both people work and schedules vary. I know PBEM makes it easier. But look at some of the forums and you can see the complaints about people dropping out (or off the grid) on longer PBEM games and leaving their opponent hanging. I don't know if my schedule would allow me to commit to another player. That's why I need a halfway decent AI. I enjoy computer games for that reason. I don't think I should lose out just because work and life takes its toll. And online play holds no interest for me whatsoever. Just my opinion...
Desert War 1940-1942 Beta Tester
Agressors: Ancient Rome Beta Tester
Flashpoint Campaigns: Southern Storm Beta Tester
Post Reply

Return to “General Discussion”