Author Topic: Like Chess, A Game Of AI War Has Three Abstract "Phases"  (Read 9402 times)

Offline x4000

  • Chris Park, Arcen Games Founder and Lead Designer
  • Administrator
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,225
This wiki article is filled with so much overall-structural information, plus so many specific smaller tips, that it seems worthy of a sticky here.
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Wingflier

  • Core Member Mark II
  • *****
  • Posts: 2,751
  • To add me on Steam, click the little Steam icon ^
Re: Like Chess, A Game Of AI War Has Three Abstract "Phases"
« Reply #1 on: June 24, 2010, 06:22:45 PM »
This was a very interesting read, and also very enlightening (and insightful) into the nature of the game, and what makes it so unique, interesting, challenging, and balanced.

Some things that I found particularly interesting:

A) That the first "phase" of the game was the aggressive or "land grabbing" phase, wherein you take as much 'territory' as possible before the enemy has time to mount up a decent counter-force to stop your pushing.  What made this extremely interesting to me was the dilemma over whether to take a possibly very rewarding T3 or T4 planet at this stage, gaining some nice bonuses, but also decimating much of your fleet in the process or opting for the smaller, less threatening planets.  The concept of leaving in favor of getting more, smaller planets, or taking a bigger one but not being able to expand as much as quickly was quite exciting and eye-opening to me, and I really like that it's a choice that every player can/will make.

B) I found it interesting that the greatest chance of losing happened during the 2nd or "middle" phase of the game.  While it makes sense once you think about it, I always assumed that the most dangerous time in the game was near the end, since if you fail to attack an AI homeworld, they will probably send EVERYTHING at you, leaving you defenseless and crushed.  This is because I am a very aggressive player, and use as little defense as possible; but I can see how an average player would have enough defense around his planets to be prepared for a failure like this.

C) The "extraordinary" players you mentioned, ones, for example, that would only take planets for what they need and abandon them, playing a very aggressive/knowledge gaining type style and then finishing off the AIs quickly, and the type of player that likes to take over the whole galaxy, but build such a massive force of defense as to withstand any attack the AI sends, were very interesting to me as well.  What I find most appealing is that the game incorporates/allows this kind of deviation from the obviously intended strategy, of picking and choosing which planets are worth taking, and leaving the rest so as not to provoke the AI any more than necessary to accomplish your goals.  It's amazing to me that the game is capable of such extremes, and with that said, I'm sure there will be many more interesting and exciting strategies to come.  The mark of a great game is when incredible things happen that the developer(s) didn't even plan for, but the game is so dynamic and diverse that these unexpected and wonderful things can occur.

D) The final thing that really piqued my interest was written under the "Chance of Losing" caption in the last phase category of the article.  You are completely right:  In most games, when you are ready to deal the final blow to your opponent, your chance of losing or being defeated is almost nonexistent.  I love that in AI War, until it says "Victory", you always have a chance to lose.  Games that truly end after the mid-game can get boring extremely quickly, since the end-game almost always dissolves into practiced pattern, which only takes time to execute; and since there is almost no chance of failure, become stale very quickly.  A great game should not adhere to those characteristics, as even near victory, you should be forced to stay vigilant, and work hard (maybe the hardest) for success.

Thank you for the article, it was very insightful and I enjoyed reading it.
« Last Edit: June 24, 2010, 06:39:39 PM by Wingflier »
"Inner peace is the void of expectation. It is the absence of our shared desperation to feel a certain way."

Offline x4000

  • Chris Park, Arcen Games Founder and Lead Designer
  • Administrator
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,225
Re: Like Chess, A Game Of AI War Has Three Abstract "Phases"
« Reply #2 on: June 24, 2010, 09:01:55 PM »
Glad you enjoyed it!  I enjoyed your response comments, as well. I'd write more, but I'm really pressed for time and I don't have much to add, anyway.  But very cool response. :)
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Colonel J

  • Newbie Mark III
  • *
  • Posts: 36
Re: Like Chess, A Game Of AI War Has Three Abstract "Phases"
« Reply #3 on: February 27, 2011, 11:45:14 AM »
I was thinking about X's chess analogy last night when I conceded a game at early to middle phase because the AI had effectively 'pinned' all my key pieces like a good chess player and more or less shut down my ability to develop my position.

The map - and my inexperience - meant that I'd allowed too many bordering worlds to reinforce too much, split my forces on too many fronts and got myself pinned down everywhere. If I moved ships in to try to neuter one of the key threat planets, it left one of my planets at reduced strength and created so much aggro I immediately got rushed by the AI from another direction - and on they rolled to my homeworld.

Tough map and good learning experience - I had a few Captive Human Settlements (Entrenched Homeworlder) and tough nut MkIII AI Eye / Ion Cannoned planets a couple of steps from my homeworld and I didn't manage my early expansion smartly enough around them. Tag Teamer on worlds with 1000+ Threat build up didn't help either  :'(

I'm learning.  :) Brilliant game.