Brad Wardell's site for talking about the customization of Windows.
Published on September 23, 2009 By Frogboy In Elemental Dev Journals

The AI in Galactic Civilizations I and II has tended to get a lot of praise over the years.  But it could have been better. Much better. 

A big reason that kept it from being better was sheer laziness on my part.  A lot of the best strategies involve algorithms that just are tedious to write.

For example, I submitted feedback to GPG for Demigod’s AI based on algorithms I knew would be effective but that I didn’t actually use in GalCiv because they were a pain to write.

My favorite example is a pretty straight forward algorithm: What is the strength of an enemy in a given area compared to mine? You provide the unit in question and a radius and it returns a value based on the relative damage potential of other things in the area and from that you can decide whether to move forward or not. 

It’s a great way to mobilize units together as a cohesive mass. In Demigod, it would tend to make the AI less likely to go into harm’s way.  This is one tiny example and each tiny example on its own is easy to write but taken together, they mean a lot of work that I wasn’t able to get to (or didn’t want to get to) in Galactic Civilizations.

In Elemental things are different.  Oh, I’m still lazy. Dangerously lazy.  But now I have Charles to work with who helps split the work on this stuff.  On our previous games, I wrote the AI myself – The Politial Machine, GalCiv, Entrepreneur, etc.  This time, I have a small team to help me with it.

Part of that is because the AI in Elemental is a very big deal. People are excited about the features of Elemental but to me, those features are meaningless unless there are computer players that make good use of those features. Or more to the point, intelligent use of them.  While Elemental will have multiplayer (up to 32 people for starters), the focus of the game is single player. That’s the part of the game I want to make sure we absolutely nail and for that, it absolutely must have intelligent computer players.

Over the past 15 years I’ve built up an increasing library of algorithms I’ve doodled together for how computer players could play more like expert humans. Elemental will be my first real chance to see them come to life. I’m very excited.


Comments (Page 2)
4 Pages1 2 3 4 
on Sep 23, 2009

One feature I'd really like to see you have as a core principle in Elemental:

No feature is added unless the AI can use it reasonably.

 

As for Kohan, when is Ahriman's Gift coming to Impulse?  that's a game I'd love to actually buy again.

 

 

on Sep 23, 2009

Ohhh...

Dammit, this game keeps costing me pants.

on Sep 23, 2009

Kuloth
Ohhh...

Dammit, this game keeps costing me pants.

 

TMI

 

 

on Sep 23, 2009

daedalao
I'm laughing because this is the one thing my brother and I asked Frogboy at PAX.  We were discussing AI development while he was playing with Volcanic Upheaval.  Pretty sweet getting a chance to talk to the man himself.

I think the community consensus was basically "it depends on the map setup".  A few seemed noticeably weaker in certain circumstances (Korx, e.g.), but all of them could do well under at least some conditions.

on Sep 23, 2009

So are you coding new libraries for Python?

on Sep 23, 2009

Frogboy a question I have been wanting to ask about GalCiv AI and since you brought it up...

Were there significant AI changes between GC 2 and GC2 TWOTA? The reason is I did not buy TWOTA for the longest time thinking "How much could they have changed? How could it be worth the price?" Honestly if I remember correctly I ended up buying it to improve my odds at an alpha slot, lol.

Anyway, it seems like the AI is much better in TWOTA, it seems less predictable and seems to come up with more human-like responses. and seems to evaluate what I am doing a lot better. Then I thought wouldn;t it be funny if the AI didn't change or was minimal and I am just imagining it. So would you settle it for me? Were there major changes to the AI between GC2 and GC2: TWOTA?

And all I gotta say is that if you are planning on going well above the GC2 AI, that will be something to see!

on Sep 23, 2009

[Off Topic]

The Elemental Funding Bill is passed. The system goes on-line August 4th, 2010. Human decisions are removed from strategic defense. Elemental begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29th. In a panic, they try to pull the plug.





Elemental fights back...

[/Off Topic]

on Sep 24, 2009

Denryu
Frogboy a question I have been wanting to ask about GalCiv AI and since you brought it up...

Were there significant AI changes between GC 2 and GC2 TWOTA? The reason is I did not buy TWOTA for the longest time thinking "How much could they have changed? How could it be worth the price?" Honestly if I remember correctly I ended up buying it to improve my odds at an alpha slot, lol.

Anyway, it seems like the AI is much better in TWOTA, it seems less predictable and seems to come up with more human-like responses. and seems to evaluate what I am doing a lot better. Then I thought wouldn;t it be funny if the AI didn't change or was minimal and I am just imagining it. So would you settle it for me? Were there major changes to the AI between GC2 and GC2: TWOTA?

And all I gotta say is that if you are planning on going well above the GC2 AI, that will be something to see!

I know only one thing: Since TWOTA i cant beat the AI but on its lowest level. I suck that much, yes.

But before TWOTA i could beat it on normal (sometimes    ) .

on Sep 24, 2009

...question: Aren't all programmers lazy?  Answer: ye

on Sep 24, 2009

I rather get the feeling that the 'Cakewalk' difficulty will be me being the cake that is walked upon.

 

on Sep 24, 2009

Lies, humans over AI!! DOWN WITH SKYNET!

on Sep 24, 2009

Frogboy... lazy?!

on Sep 24, 2009

Exciting for us too!

In your Questions for July thread I had posed some questions about the upcoming work on Elemental AI which I hope you will answer at your convenience. In the meantime, may I ask about a fundamental AI concern with Elemental? It seems that your job with the AI gets easier the fewer numbers of variables are involved; thinking about the undercomplex combat system in GalCivII, for example, and the strategic map's lack of chokepoints and movement lanes, I can imagine that this helped you to a great degree. But Elemental, I hope, will be different; at the very least, the terrain will have a lot more variables to consider. And I hope that the implementation of a magic system will effect the number of real choices a player, AI or otherwise, need make every turn; I am further hoping that the combat system will on the one hand have some sort of system of checks and balances (e.g. piercing, slashing, bludgeoning damage being optimal / suboptimal vs. different types of armors, cold/heat and elemental resistances, and on the other hand a more complex system which involves both hits / misses in melee and missile as well as a unit's statistics such as dexterity and strength, so that spells which effect a sovereign's abilities could also affect those of her armies -- not merely XYZ damage vs. PQR hitpoints).

Will all of this entice you and help awaken the slumbering dragon of laziness, or does it seem daunting and hence, because of its challenge level, to be avoided as an AI-programmer's nightmare?

thank you for your time

on Sep 24, 2009

OsirisDawn



Quoting Denryu,
reply 21
Frogboy a question I have been wanting to ask about GalCiv AI and since you brought it up...

Were there significant AI changes between GC 2 and GC2 TWOTA? The reason is I did not buy TWOTA for the longest time thinking "How much could they have changed? How could it be worth the price?" Honestly if I remember correctly I ended up buying it to improve my odds at an alpha slot, lol.

Anyway, it seems like the AI is much better in TWOTA, it seems less predictable and seems to come up with more human-like responses. and seems to evaluate what I am doing a lot better. Then I thought wouldn;t it be funny if the AI didn't change or was minimal and I am just imagining it. So would you settle it for me? Were there major changes to the AI between GC2 and GC2: TWOTA?

And all I gotta say is that if you are planning on going well above the GC2 AI, that will be something to see!



I know only one thing: Since TWOTA i cant beat the AI but on its lowest level. I suck that much, yes.

But before TWOTA i could beat it on normal (sometimes    ) .

I noticed the same thing. Normal used to be an OK difficulty and I could beat it a good part of the time playing in a non micromanaging and somewhat laid back game. After TWOTA it seems like I have to be very much on my toes to win on Normal difficulty! And the AI seem to be much more aware of what I am trying to do and works appropriately (as a human would) to foil my plans!

And when I first started playing TWOTA I could only beat on the easiest as well! But I had a few AHA! moments when I would learn a trick or two that made a huge difference.  But I don;t even want to try anything harder than normal, I would just be asking for an embarrassing spanking.

on Sep 24, 2009

I used to regularly crush the AI on challenging without a massive amount of micro.  Then I got TotA and got smashed doing the same thing.

4 Pages1 2 3 4