|
Author |
File Description |
St_Aquinas |
Posted on 02/16/11 @ 02:40 AM (updated 03/25/11)
File Details |
Version: |
The Conquerors 1.0c |
This is my first AI! Written for the Britons, currently has a rudimentary adapting military, working on a variable army that can rush. Plays both RM and DM. Does not use cheats. *Update* v1.7.4: Fixed Building rules. Added early dark age villager scout
v1.7.4.1: Fixed errors in the britmilitary file
v2.0: Complete Rebuild of the James I script. After looking at and playing against other scripts I decided to simplify and change my overall structure. Thanks to all other scripters, you most likely have directly impacted the script, or at least inspired a change. |
Pages: [1] 2 3 » Last » | Author | Comments & Reviews ( All | Comments Only | Reviews Only ) |
---|
Mash (id: Mashek)
Staff
Official Reviewer |
Posted on 02/16/11 @ 06:04 PM
I can't seem to detect the "James_I" AI in standard game, even though I've extracted the AI into the AI folder. Any ideas? |
St_Aquinas
File Author |
Posted on 02/16/11 @ 09:28 PM
The AI name itself does not have an underscore in it, try renaming the ai file if it doesn't say James I. |
Campidoctoris |
Posted on 02/17/11 @ 07:55 AM
You have a bug which causes you don't research feudal age despite to have +500 food and necessary buildings.
Your AI instantly builds a barracks playing Random Map. Is it other bug and you only want it for DM? Or, is it your strategy also for RM? If it's the second case delay it until transition to feudal.
Read the advices that offwoo2000 has received for his Noob AI. |
St_Aquinas
File Author |
Posted on 02/17/11 @ 05:39 PM
AS I also posted in the description I have updated the files. Work on an adaptive military is in place. Also hopefully a faster time to feudal age. |
Campidoctoris |
Posted on 02/17/11 @ 06:48 PM
I suggest you to test your own script by yourself before to upload it to make we'll test it for you.
For example, I've tried to test your AI, but it has an error because one of your rules is too long. If you had test it you could know it and save for you the time to upload it and for me the time to download and test it anyway. I often, after a few failed "improvements" I have more than ten errors, upload it ten times should be quite exhausting for myself.
Other reason is that so you can see if you have too many food or few gold and adjust your gatherer percentages, if number of build order is adequate to your strategy, etc. Your own eyes and your understanding of the game will be your best teachers.
Of course, it's possible that you aren't able to correct a bug, find a bug or improve your strategy despite to make a few test. Then it's time time to request for help. Anyway, you won't forget the information that you have discovered by yourself, while whatever you can read could be.
You also can read other threads, learning from advices between different scripters.
For example, in aiscripters you can find in recent threads advices for a scripter who wanted a castle time even faster, another (me) who wanted a system to take a more compact town, sometimes it's a escrow system for a faster age advancing or take important and expensive technologies earlier,etc. |
St_Aquinas
File Author |
Posted on 02/19/11 @ 12:50 PM
Updated! Version 1.4 now available
|
jdstankosky |
Posted on 02/19/11 @ 02:47 PM
You have a typo that prevents a file from loading...
Other comments:
This AI looks like it could have really really good potential, but there are some things you need to do with your economy before that can happen.
First, some low notes:
Your villagers walk way too far for resources. I saw some walk clear across the map, THROUGH the enemy town just to get at some fish. :-/
Second, your very first minutes are kind of shaky with the decision to go for wood before going for food. As a general rule of thumb, don't try thinking about wood until you have 9-10 villagers. This ways you can get enough food right at the get go to have a constant stream of villagers being produced without much down time at the town center.
Other things that I didn't quite understand, you had built a barracks, 2 archery ranges, and a stable... but you only used the one barracks. The others weren't even utilized until much later. Why didn't you start training archers from the archery ranges? or at least Skirms to defend against enemy archers while waiting for your castle? If you aren't using your stable until castle age, I would hold off building it until you start to advance to it.
Ok, now some high notes:
I liked (I haven't read your code, so I'm assuming here) how you started training counter units for the enemy's (standard computer on Hard) knights and archers once you actually WERE in castle age. That was very smart.
I really liked when I saw you build a second Town Center.
I liked the fact that you built farms in Dark Age, this is very helpful for AI's.
Conclusion:
I really want to see you develop James_I further, it really makes me think it can become a great script, however, you need to do some work on your economy before that can happen!
If you want expert advice and constant friendly help, check us out on aiscripters or post questions in the AI Scripting section of this site.
Good luck and happy frustrations! :D |
St_Aquinas
File Author |
Posted on 02/19/11 @ 05:04 PM
Updated! Version 1.5 now available![Edited on 02/20/11 @ 03:13 AM]
|
Campidoctoris |
Posted on 02/20/11 @ 07:57 AM
St_Aquinas, your version 1.5 has other error in the line 23 of the BritBasics.per, it doesn’t identify the word “defrule”. I’ve corrected this error adding a closing parenthesis en the previex rule because James I 1.4 has improved enough to intrigue me. Anyway, I suggest you again a last test before upload your script, because we often have errors and bugs after our “last” change in our strategy. If you test Deathmatch (DM) you’ll find other error, but I’m only interested in Random Map (RM), so youl’ll have to correct it yourself. Obviously, my comment are about your RM strategy.
Your Dark Age economy is quite good now. The only mistake I’ve seen this time is that you still continue building the lumber camp before the mill. Your berries are further to the town center than the near trees, so the mill should have the highest priority if you don’t want foragers wasting a lot of time walking instead of gathering.
Next problem. Your gatherer percentages. In transition to feudal and early feudal they are not bad, you have made your military building fast, but later you need a higher percentage of villagers in food. You were winning against the Standard AI, but your unbalanced economy avoid that you arrived to Castle Age at time. You had a huge excess of wood and gold, and very little food. Of course, finding a good percentage gatherers is difficult, more if you, like me, are changing your units. For example, when my Mayan script is training only archers in feudal it arrives to castle in 24’, but when it trains skirms against archery or m@a against skirms, I have an excess of gold and a lack of food which delays my castle time to +30’.
There are two ways to avoid an unbalanced economy:
- Readjusting the gatherer percentages if you detect you have unbalanced resources and your unit selection of the moment is unbalancing it even more. For example, with an excess of wood and gold, if your script turns to archers again you could not need readjust nothing, because archers will eat that excess and will allow you stock a lot of food. Anyway, this is very complicate, in fact I want to do this for my next update, but it requires work and my AI have more urgent matters.
- Market. The only reason to build a market in a flushing strategy is buy and sell resources to get a more balanced economy. For example, if you have your feudal buildings finished and have more than 200 wood you can sell them. If you have more than 900 food and less than 200 gold sell them, and if you have more than 400 gold and less 800 food you can buy food to get a faster castle time. Of course this numbers are only suggestions, but if you don’t use the market NEVER until +2000 resources, your trading rules will be only useful to DM games.
Jd is right about not construct buildings that you are not going to use, in general. But there are exceptions. For example, my next update will change its market for a second barracks, although my Mayans often don’t train infantry in the Feudal Age, anyway, if enemy starts with cavalry, I’ll have enough spears to protect my cheap archers in the half of time. This suggestion that Marathon gave me in aiscripters really works well in my script. Anyway, until your eco became enough well balanced to get very few buy/selling I suggest to keep your market in yours.
Other small thing is that if you train infantry you should research scail mail, except if you really need to advance to Castle Age NOW, in which case you also should stop military training. This tech really improves your infantry, and if you aren’t training cavalry this should have a higher priority than forging, because it improves your infantry even more. Although both techs are very good.
Finally. Study Training AI, it’s great to learn.
EDIT:
I forgot other mistake. James_I stops villager production in early feudal. It's wrong, you should train much more if you pretend spam soldiers from 4 military buildings, research blacksmith techs and save resources to advance to feudal.
My scrip only stops villagers production when I have 100 villagers (if you train more, the rest will be idle) or if my population is near 200. Anyway, other scripts have a lower limit in feudal to advance to castle age quicker. But every serious script will try to get 100, at least in the Imperial Age.
P.S: When I arrive to 60-70 villagers, I get so many resources that I "can't" avoid to advance to castle age anyway.[Edited on 02/21/11 @ 01:26 AM]
|
St_Aquinas
File Author |
Posted on 02/20/11 @ 12:37 PM
Work on the new version happened a little ahead of schedule. Version 1.6 is now available! has some bugs I can't figure out that are explained in the description. Any ideas on fixes are greatly appreciated!
Edit:
Updated! Version 1.7 now available![Edited on 02/21/11 @ 11:16 PM]
|
Pages: [1] 2 3 » Last » |
HGDL v0.8.2 |
Statistics |
Downloads: | 293 |
Favorites: [] | 0 |
Size: | 37.90 KB |
Added: | 02/16/11 |
Updated: | 03/25/11 |
|