Supercombattant pretenders....what about... - Page 5

Alneyan · #41 November 3rd, 2003, 09:50 PM

Where did you read I asked the devs not to care about the AI, or in Nagot post for that matter? There is quite a difference between saying, like Nagot said, "AI should not be the first concern for the devs" (or somethign along these liens), and saying "Don't touch the AI".

There may be another factor as well. Changing balance in drastic ways may very well alter the AI. Say you are tweaking the balance between light and heavy units. The AI will have to be changed accordingly. Say you are making magic more/less expensive, supercombattants less common, or something else. The AI will once again have to be modified, and perhaps even important changes.

I would believe these important balance changes will be settled, along with bug fixes among other problems, before the AI is completely tweaked. Still, some changes are currently going on for the AI, in particular fields. (The AI doesn't like to build castles, and Kristoffer asked for suggestions about this matter) But I would not expect the developer(s?) involved with the AI to make a much improved AI while handling all the other requests/bugs/balance concerns/add as needed in a short amount of time. (Except if Illwinter had managed to master the flow of time)

As for these scripts, I will let Nagot speak about this topic. (Apart from modifying the starting provinces, that is to say, the things lurking in the map file)

Nagot Gick Fel · #42 November 3rd, 2003, 10:00 PM

Quote:

Originally posted by Alneyan:
I do have a question for you: how would you write a script to make the AI build more heavy troops?

Something like:
#commander "Commander of Ulm"
#units 50 "Guardian"

Repeat a few times, enough for their upkeep to eat most of the AI's initial income. With that you won't see the AI building many light troops for some turns.

Nagot Gick Fel · #43 November 3rd, 2003, 10:11 PM

Quote:

Originally posted by Alneyan:
As for these scripts, I will let Nagot speak about this topic. (Apart from modifying the starting provinces, that is to say, the things lurking in the map file)

I never implied that map file scripting would solve all the AI's shortcomings. But it's available as we discuss, easy enough to do with minimal practice, and it can give the AI enough of a boost to become challenging to the best players. Obviously it won't change the AI's behavior - eg, its tendancy to build peltasts when it has the resources to build hoplites, or to build other castles than the ones you gave him in the map file.

DominionsFan · #44 November 3rd, 2003, 10:19 PM

Quote:

Originally posted by Nagot Gick Fel:

quote:
Originally posted by Alneyan:
I do have a question for you: how would you write a script to make the AI build more heavy troops?

Something like:
#commander "Commander of Ulm"
#units 50 "Guardian"

Repeat a few times, enough for their upkeep to eat most of the AI's initial income. With that you won't see the AI building many light troops for some turns.

"something like"....."for some turns".....yes this made lot of sense.

Gandalf had made a very challenging map. Yes you can script like that, it will be more fun to play, but it is nothing more just a preset startup. The AI is the same, I won't tell you one more time

Sure the devs will need lot of time to update everything, but hey, we can wait.

Nagot Gick Fel · #45 November 3rd, 2003, 10:23 PM

Quote:

Originally posted by licker:
I dunno about the rest of you, but it seems to me that bugs and balance will be the first things addressed, mostly becasue they are easier to tackle from a codeing standpoint (or in the case of difficult bugs, more important to fix).

I never said anything else. Also keep in mind that, although there may be more SP players than MP ones, participating in an MP game is a huge commitment, while firing up an SP game at home is not. Assume there exists a rare, but game-breaking bug that happens only once out of 20 games, and it hits in the middle of a PBEM involving a dozen players - that will ruin several months of efforts by all of them. In SP you'll just mutter "", forget it and start a fresh game in the next minute.

licker · #46 November 3rd, 2003, 10:31 PM

Quote:

Originally posted by Nagot Gick Fel:

quote:
Originally posted by licker:
I dunno about the rest of you, but it seems to me that bugs and balance will be the first things addressed, mostly becasue they are easier to tackle from a codeing standpoint (or in the case of difficult bugs, more important to fix).

I never said anything else. Also keep in mind that, although there may be more SP players than MP ones, participating in an MP game is a huge commitment, while firing up an SP game at home is not. Assume there exists a rare, but game-breaking bug that happens only once out of 20 games, and it hits in the middle of a PBEM involving a dozen players - that will ruin several months of efforts by all of them. In SP you'll just mutter "", forget it and start a fresh game in the next minute.

What do game breaking bugs have to do with an SP v. MP discussion? Bugs are at the top of the 'fix me' list always, no matter their import to either side...

What improving an AI does is lengthen the lifetime of the game, if you can beat the AI too easilly its not going to be that interesting to try to beat the AI again using a different tact. If the AI can jump up on you, well then you'll stick at it longer. The standard (BS) reply to that is 'go play some MP' but it is BS because not everyone can, or cares to, play MP. Especially PBEM where a game takes months. An improved AI benefits everyone, other than the people who *never* play SP and *never* include AIs in their MP, and you'd have to do some work to convince me that those people even exist

Lets agree though that no one said improving the AI was a bad thing.

HJ · #47 November 3rd, 2003, 10:36 PM

Quote:

Originally posted by licker:
What do game breaking bugs have to do with an SP v. MP discussion? Bugs are at the top of the 'fix me' list always, no matter their import to either side...

What improving an AI does is lengthen the lifetime of the game, if you can beat the AI too easilly its not going to be that interesting to try to beat the AI again using a different tact. If the AI can jump up on you, well then you'll stick at it longer. The standard (BS) reply to that is 'go play some MP' but it is BS because not everyone can, or cares to, play MP. Especially PBEM where a game takes months. An improved AI benefits everyone, other than the people who *never* play SP and *never* include AIs in their MP, and you'd have to do some work to convince me that those people even exist

Lets agree though that no one said improving the AI was a bad thing.

Again, you pretty much said it all.

LordArioch · #48 November 4th, 2003, 01:55 AM

Bugs and balance should certainly be top priority, and balance may even help AI.
However...although I do play single player, including an AI in a MP game wouldn't be very fun. AI's and human players should not be mixed...it becomes a game of abusing the AI to get advantage.
No game as complex as dominions is going to have an AI with no weaknesses, and putting one in MP is just asking people to beat it up for its riches.
And if PBEM takes too long for you try a networked game. Mine's past turn 20 and it got set up Last week.

Particle · #49 November 4th, 2003, 02:01 AM

KO has replied in the AI's thread yesterday:

"I got an idea regarding AI dependancy on LI when I was away. Vacations are 'foyson' for the mind.

Regarding fort construction. I'm not sure how the AI works, but where would you build a fortress? Consider fort type, nation, geography, income, resources etc.

What is the most important matter? How should they be weighted? How much shall current wars affect the spending of time and money. What army should build the fortress?

I wouldn't mind a numerical evaluation of this such as:
Castle cost / 2 < Income + res + gem income x 25 + neighbors x 5
Add a couple of other conditions.

Just to make you think. Eventually it might result in something good."

The devs ARE wishing to upgrade the AI., and we MUST HELP TO THEM WITH IDEAS AND REPORTS. Stop TROLLING HERE Nagot Gick Fel!

[ November 03, 2003, 12:10: Message edited by: Particle ]

johan osterman · #50 November 4th, 2003, 02:28 AM

Quote:

Originally posted by Mortifer:

This means that the hella lot of Posts / suggestions about 'how to improve the strategic AI' were totally pointless?

The AI thread isnt pointless, if nothing else it has moved form: the AI sucks, it must be improved, to a list of specific complaints. The list is actually helpfull and much less irritating than the orginal the AI sucks Posts. Without promising anything making the AI favourably wieght sacred or armored troops when recruiting might be a candidate for an improvement, for example.