Michael Emrys Posted February 21, 2014 Share Posted February 21, 2014 Damn, wish i had known. I just gave my last $10,000,000 to Farmville. Sucker! Michael 0 Quote Link to comment Share on other sites More sharing options...
womble Posted February 21, 2014 Share Posted February 21, 2014 However, I somehow think the AI faces an exceptionally tough challenge in CM, for there're rarely other games where an AI controlled entity faces so many environmental factors like terrain types, paths, LOS/LOF considerations etc etc. IMO it wouldn't be a bad idea to pump it up a notch to give the game a bit more re-playability... Chicken, meet egg. That very complexity that challenges the AI is even more of a challenge for the AI creators. I'm sure it'll gradually get improved, but I don't think Steve is going to take a punt on getting a return from n hours of dedictated AI programming that will match the return from the same time spent giving us more bells and whistles and room in which to ring and blow them. Personally, I can't consume BFC's prodcut at the rate it's being produced at the moment, and would welcome a pause in new theatres/timespans if it meant the AI for the existing games got a significant upgrade. I'd welcome it with open wallet... All we gotta do is persuade Steve that enough people will pay for a "big upgrade" (to all their existing theatres) to make up for not getting the income from a Bulge or Barbarossa or some such release. I wonder if there's a "tipping point" where the number of engine upgrades people would need to buy makes engine upgrades the largest income stream (ahead of modules and new family-base-games). 0 Quote Link to comment Share on other sites More sharing options...
Skwabie Posted February 21, 2014 Share Posted February 21, 2014 Chicken, meet egg. That very complexity that challenges the AI is even more of a challenge for the AI creators. I'm sure it'll gradually get improved, but I don't think Steve is going to take a punt on getting a return from n hours of dedictated AI programming that will match the return from the same time spent giving us more bells and whistles and room in which to ring and blow them. Personally, I can't consume BFC's prodcut at the rate it's being produced at the moment, and would welcome a pause in new theatres/timespans if it meant the AI for the existing games got a significant upgrade. I'd welcome it with open wallet... All we gotta do is persuade Steve that enough people will pay for a "big upgrade" (to all their existing theatres) to make up for not getting the income from a Bulge or Barbarossa or some such release. I wonder if there's a "tipping point" where the number of engine upgrades people would need to buy makes engine upgrades the largest income stream (ahead of modules and new family-base-games). Well Womble, at the end of the day I think it comes down to the economy. With the current state many ppl are just trying to make a living and play some games to blow off steam. With a good economy going, you'd have more ppl pursuing quality entertainment, that'd be when they'll want a better AI to provide a tougher challenge. There can be a few of us screaming (typing) hard on the forum, but it'd still be the masses out there that provide the $$s. Atm things seem to be turning for the better, and the mentioned triggers seem to be a nice start. Which leads me to a second point. It might not be a good idea to do an entire pack just for the AI. That would contain quite a level of risk. Maybe incremental changes, comes bit by bit with addon modules seems more sensible. I mean BFC has been around for more than 10 years outliving most of the gaming company/studios, they must be highly risk averse. 0 Quote Link to comment Share on other sites More sharing options...
Tordenskjold Posted February 21, 2014 Author Share Posted February 21, 2014 No doubt that a better AI makes for a more challenging game. The problem is the disproportional amount of work to get that part of the game improved vs. what can be improved in other parts of the game using the same amount of effort. Steve For me, the biggest improvement will be a better AI !!! 0 Quote Link to comment Share on other sites More sharing options...
Dadekster Posted February 21, 2014 Share Posted February 21, 2014 For me, the biggest improvement will be a better AI !!! I look forward to seeing how the new trigger points work out but ultimately I think CMRT is going to be the point where I make a real effort to find some live opponents to play. It'll never get better than that imo or at least until they make Skynet to play against. 0 Quote Link to comment Share on other sites More sharing options...
Collingwood Posted February 21, 2014 Share Posted February 21, 2014 I'm being optimistic here and hoping that triggers are a good compromise to getting better AI. Eventually when we have nested and conditional triggers, it will allow the scenario designer to assess the terrain and suitable plans, and anticipate the human player's plans and possible deployments and movements. The triggers are then the framework for ensuring the computer opponent reasonably maneuvers and responds. AI after all is programmed by humans who try to impart human knowledge about a problem into algorithms. You get a new appreciation for what a wonderful computer our brain is when you try to break down this knowledge into algorithms and code that is both effective and efficient. For a completely dynamic AI - one where you could just throw some terrain at it, with some objectives, and expect it to deploy and maneuver in a canny way, like Bil for example... honestly I can't see that happening. I've got experience in AI systems and every time I think about different ways I could approach the "Combat Mission dynamic AI challenge" I think I could do it as long as it was only expected to be competent rather than expert. But when I consider how much time it would take - I end up measuring it in years. Perhaps someone smart could get it done faster, but these things always turn out to be more complex and tricky than we imagine. A simple (and ultimately doomed) approach would be the brute force method: get the AI to simulate every possible permutation of move it could make, and every possible counter move etc (like the tactical side of chess engines) and assess which one seems to lead to the best possible outcome. The problem here is the number of choices quickly become astronomical, so you need to know how to prune the search tree. Otherwise you might have to wait several days for the computer to choose its move Brute force requires a lot of computing power, and chess positions are way easier to assess than CM situations, IMO. Especially since if a Queen takes a Rook, the rook is taken. But in CM when a tank shoots at a tank - several different outcomes are possible. So yeah, maybe Skynet if we go the brute force way. 0 Quote Link to comment Share on other sites More sharing options...
Baneman Posted February 21, 2014 Share Posted February 21, 2014 I'm being optimistic here and hoping that triggers are a good compromise to getting better AI. . By definition, they are better AI, since the AI we're talking about here is "AI plans", not TacAI. Eventually when we have nested and conditional triggers, it will allow the scenario designer to assess the terrain and suitable plans, and anticipate the human player's plans and possible deployments and movements. And a large queue of Scenario Designers at the mental health clinic... The problem here is the number of choices quickly become astronomical, so you need to know how to prune the search tree. I'm just guessing, but I'd imagine that the TacAI is a good example of a search-tree already pruned effectively After all, for every example we find where we're yelling at our pixeltroops for being morons, there are literally hundreds where it does pretty much what we expect. 0 Quote Link to comment Share on other sites More sharing options...
Collingwood Posted February 22, 2014 Share Posted February 22, 2014 I'm just guessing, but I'd imagine that the TacAI is a good example of a search-tree already pruned effectively I agree the TacAI works mostly well. And besides if we want to play a truly crafty opponent H2H is always there. And a large queue of Scenario Designers at the mental health clinic... I'll see you there! 0 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.