Age of Kings Heaven 139
Welcome
Downloads Home
Best Files
Review Guidelines
Main site
Forums
Code of Conduct
Search

Advanced Search
AI Files
Home
New Releases
New Reviews
New Comments
Top-Rated

Submit File
Browse Categories
New Files
Recent Updates

Downloads Home » AI Files » Rampage 0.7

Download File Post Review Post Comment

Rampage 0.7

Author File Description
Alevo
File Details
Version: The Conquerors
This is my relatively new AI, currently it plays Franks, Aztecs and Huns. For Aztecs it will attempt a rush and for the Huns and Franks it will make the Knight line, for Huns it will also create Elite Tarkans and eventually Light Cavalry and Hussars.

I'm completely open to suggestions and comments on how it plays are more than welcome as I'm trying to improve it all the time, give it a go :)

Version 0.3:
- Some small changes to the escrowing systems.
- Constants problems fixed.
- It now advances to the Imperial Age properly!
- Changes to max villager count.
- Now will train Elite Skirmishers eventually.
- Various other bugs fixed.

As always, any problems and let me know!

Version 0.4:
- A few changes for a faster Castle and what should be a more effective Castle age playing style.
- Changes to unit-creating conditions and stuff.

Version 0.5:
- Many changes to Huns civ for the AI, it now uses things like TSA.
- A faster Castle Age. Still needs to be improved.
- Franks and Aztecs changed slightly (although Huns are still the best one).

Version 0.6:
- Messed about a bit with the Huns civ, should improve it slightly.
- Teutonic civilization added. Will fast-castle in 18-20 minutes (usually around 19 mins) and get loads of skirmishers/spearmen/teutonic knights. Will also halt spearmen production if it sees fit. This is probably now its strongest civ.

Version 0.7:
- A Byzantine script similar to the Teutonic one has been added. A few changes from it though to make it fit with the civilization change.
- It will train camels insead of spearmen.
- It will Castle usually about 19 mins 30 seconds. Sometimes more, sometimes less.
Pages: « First « 1 [2] 3 » Last »
AuthorComments & Reviews   ( All | Comments Only | Reviews Only )
Alevo
File Author
Yeah but my escrows are kept for things like Paladins and Elite Tarkans ;)
marathon Nice first AI :) Very nice

Strategically, it is best to completely shut down everything when you want to take expensive upgrades. For example, when aiming at Paladin, you can go without escrow, until you have taken Cavalier, in which case you could try a heavy escrow, like 40% for a short amount of time.

If you feel threatened during this period, you could stop escrow and release it all, train some cavs, then resume. This is my general approach.
Alevo
File Author
Thanks man, I've got an update I'm putting the finishing touches to now which makes it much more effective and hopefully will take about 2 minutes off of the Castle ageing :)
Campidoctoris I mentioned you India Marina, my KLEW script made with only 20 rules. Congratulations, you have defeated it.

Anyway, your script has still a lot of space to improvement. It's natural, it's still very young.

Fast castle strategies have two options:
a) Market+blacksmith. Zero gold miners until you click Castle Age button, why? It's because you sell your stone for gold, so you save 100 wood for an early mining camp, and a few villagers that can gather food and wood instead of gold.
b) Mining_camp+Barracks+Blacksmith+Stable.

With option a) it will be easier for you take a faster castle time. With option b) you'll get easier a stronger eco, and allows a feudal defense if case your are agressively flushed (if you want it).

I'd recommend you option a) if your target was send knights/eagles/monks against your enemy the sooner the better. But you prefer boom and take a huge economy to allow paladin flood, so I suggest you b), which is what makes Demon AI, an aztec script which also makes an early castle drop (another reason to don't build an early market to sell your stone) and a huge boom. Anyway, Demon is an script which could be very interesting to you, for study. You can download it in aiscripters.

Choose the option you want, but you can improve your castle time with a stronger eco. The best Fast Castle scripts are able to castle in 18:XX with mesos (sometimes) and in 19:XX with every civ (usually). Anyway, your 21:XX is quite acceptable for a first script, but you'll have problem against the most agressive flushers in the Ladder.


Your next problem is that you delay too much your blacksmith upgrades. It can be a good idea stop soldier production to accelerate your imperial times and research ri-paladin and ri-elite-tarkan faster. But one thing is stop knight/tarkan training because you want improve them with expensive upgrades, another very different is delay cheap upgrades to spend 1000F and 800G only to imperial jump and later a lot of extra resources to take those expensive upgrades. See these tests with FU post-imperial units:
- 1 Saracen knight can defeat a Korean cavalier, although surviving with only 12 HP.
- 1 Chinese cavalier loses against a Byzantine paladin, but last one survives with only 6 HP.

I mean, don't train more units if you fear have tons of them but unupgraded, but research what you need when you need, not later.


Last problem. Skirms are great against archery units, and don't cost gold. But cavalry are good against archers anyway (except if you let him reach a critical mass), and vulnerable against spears/camels. Keep in mind that skirms are adequate against spears (although nothing spectacular) but horrible against camels.

So I recommend you spears instead of skirms for your Huns. Hunninc halberdier have most upgrades avaliable, so they can be good against most enemy spears, they are good against camels, and they don't cost gold anyway.

You also could consider light cav instead of knights againt enemies with massed monks. Although you should spend more resources in upgrades in such a case. Well, test your script against Demon (do it from the point of view of this interesting and powerful script) and you'll watch the power of the Aztec paladins ;) And don't worry, my Crusade has sometimes problems against mushing scripts.
Alevo
File Author
Thanks :) I'm sure this will be really helpful, at the moment I'm working on a new blood map but I find time for my AI scripting when I get bored of scenario designing, and when I'm bored of AI scripting I go back to designing scenarios so it should be fine :P
Your Crusade script confuses me, I tried to watch it play in some games and for some reason half the time it just packs up and sits tight with its farms and that's all it does for the rest of the game if an opponent has more troops than it :S
Campidoctoris Crusade trains soldiers since feudal age, sometimes since advancing to feudal, with no stop (except sometimes in very late feudal or very late castle to accelerate a little his age advancing).

If it has 5 or more soldiers than its enemy and doesn't detect that the enemy has a superiority in heavy cavalry which compensates its numerical disadvantage, then attack until he smash the enemy or finish with military disadvantage.

Once Crusade reaches 200 population it attacks anyway, even if the enemy also does it, if Crusade has every important upgrades.

Well, my next version also attacks if an ally request him for it and is not at disadvantage. For example, if Crusade has 50 soldiers and his strongest enemy only has 54, Crusade will attack to support an ally.
Alevo
File Author
I meant if it did lose military advantage sometimes it jut decided to pack up and prepare to die :L
marathon I ran a test with Aztecs.

- It dropped its mill before it discovered berries.
I suggest to drop mill if you have found food, or if it's too late, like 3-4' into game.
If you want to make it better on weird maps, you could add the condition:
(dropsite-min-distance wood < 8)
when you decide to drop lumber camp first.

- Its 7th villager was a lumberjack. This is not the safest idea: if your first villagers go to forage then you will have idle TC time.
I suggest first 8-9 to go to food.
But this isn't a big issue anyway.

- It pressed the feudal button, it did nothing special during transition and when it finally entered feudal it decided to start gold mining. Its feudal buildings came too late.
I suggest to use a goal or something for when you have pressed an age advance button, because most strategies include special plans for age transition periods.
So, when you press the button, you increase woodchoppers, decrease farms and farmers, start gold mining, maybe even drop a barracks.
Dropping the gold camp before you press the button is a good idea as well.

- Your exploring was very good. But your camp distances too high, so both first 2 lumber camps were dropped at huge distances. This costs quite a lot to your economy that early, especially since you have neither wheelbarrow nor hand cart.

- Your sn-maximum-town-size was too big, imo. Better reduce it a little and link it to age and population. A value of 30 or higher is for late castle age or imperial.

- You need to use facts in your research rules. You took forging while armless and waiting for the castle age.

- Too many early mining camps, imo.

- You need to build houses earlier when you expect to have a higher production: in castle age, when you are ready to train soldiers, better drop houses when housing-headroom is 12 or even higher.

You are in a good way. Just shape your strategies in your mind, and then script them :)
Alevo
File Author
Haha I know I should really fix them all, but I'm concentrating mainly on finalizing my Huns one at the moment. The Franks still need a bit of work too so I'm going to save that for last. The Huns are its strongest civ by far at the moment...
Campidoctoris I've tested Hunnic Rampage vs Sassanide Empire, and you won, but you were too near to be destroyed in early Castle, because you built your stable too late.

I told you that you must choose between mining gold and build a barracks or build a market. You can get both things, but only after you click castle age research, because if not your castle time will be slower. You don't build your barracks until transition to castle but you mine gold and build a market which you needn't to sell stone because you have gold anyway.

Arrive to castle in 21' (sometimes 20:30) is the reason because Sassanide punished your economy so hard before your fewer knights helped by TC fire repelled Persians. And you win only because your script booms much better than Sassanide, but you were in a worse position in the beginning and a stronger opponent should finish you.

Later I tested Hun Rampage vs and Aztec Novelty (the strongest AI from the Squires League). You lost, smashed by Aztec cavaliers, but don't worry because when that evil script masses enough monks it becomes really dangerous. Anyway, I observed that you could research important blacksmith techs but you didn't to accelerate your imperial time. It's wrong. I told you that if you want better troops, stop their training to save resources to get technologies or advancing to next age, but stop even cheap and effective techs to spend 1000F and 800G and get later tech which are more expensive has no sense.

Spears are more useful than skirms to protect knights. They also improve their attack with the same blacksmith techs than cavalry.

And study also the BTG_Krush AI. It's great using knights.

Anyway, I recognize that your script has a great potential. It also should be great for Death Match games with one only change.

You have a similar condition to this to advance to feudal age:
(unit-type-count villagers >= DARK-VILLAGERS)

You could replace it by:
(or (unit-type-count villagers >= DARK-VILLAGERS)
(food-amount > 1000)
)

Another suggestion. Study BTG_Krush AI. It has a similar style to y
Pages: « First « 1 [2] 3 » Last »

Post Review Post Comment

HGDL v0.8.0

Login
Forum Username:


Password:




Create a new account
Forgot password?
Statistics
Downloads:342
Favorites: [Who?]0
Size:27.11 KB
Added:06/18/11
Updated:07/26/11