Brititron 1.8

Upload your latest creations here.
Please note that by posting comments to posts in this forum you forfeit your personal copyright on whatever you post. AI Scripters reserves the right to re-use objective comments in script reviews without requiring the posting author's permission. Credit, where appropriate will be given in such cases.
User avatar
Promiskuitiv
Waheguru
Posts: 1364
Joined: Thu Nov 10, 2011 1:07 pm

Re: Brititron 1.8

Post by Promiskuitiv »

Tested Chinese Brititron against Promi, you attacked at 28 minutes, Promi only had one single archery range for whatever reason (seems like he didn't detect the danger) until you attacked, he trained skirms from this archery range non stop the whole game but that wasn't enough when you attacked => Promi's skirms died but meanwhile Promi built 2 ranges + a stable => E-Skirms out of 3 ranges + knights out of a stable defeated your army, let Promi boom and win.

I put it all in one sentence 'cause i'm still tired. :lol:

Maybe you should add something against skirms if it doesn't destroy Brititron's spirit.

cakemaphoneige
Guru
Posts: 811
Joined: Wed Feb 13, 2008 6:50 pm

Re: Brititron 1.8

Post by cakemaphoneige »

Yea skirmishers are a problem. It sounds like promi got lucky in that game though.
Ive got brititron training a couple of knights and light cavalry to try and deal with them, however sometimes it may not train as many as it should. I have plans to adjust its research order and train order, so alot of things might change.
Can you guys test all 5 of the civs? mayans, britons, chinese, mongols and turks. I made alot of changes to turks yesterday, i will upload sometime today.

User avatar
Promiskuitiv
Waheguru
Posts: 1364
Joined: Thu Nov 10, 2011 1:07 pm

Re: Brititron 1.8

Post by Promiskuitiv »

Well your archers were stuck at a TC of mine (not getting shot, but shooting at it) => gave me enough time to quickly train a defense.

But after that game i loosened the conditions to build a stable / an additional archery range before the third TC anyway (and it worked out better).

Will test the other civs later too (esp. turks).

cakemaphoneige
Guru
Posts: 811
Joined: Wed Feb 13, 2008 6:50 pm

Re: Brititron 1.8

Post by cakemaphoneige »

Just remember promi, you cant counter everything. One day you will find a counter strategy conflicting with another counter strategy.
For example, the last 5 tests i did between wartron and promi, which was about 1-2 weeks ago now, promi won 1 , lost 3 and drew 1 (game went past 2 hours, gets to boring for me to continue watching after that point). That was with a few slight adjustments. I have big plans to force you to completely change your counter strategy, and thats when the counter wars will truly begin, theoretically its all a big game of chess. Atm, our ai's are even, its just a mater of who's strategy counters who's strategy, its almost neverending.

What you want is an idea strat thats able to handle >= 75% of situations, and a few slightly different strats for the really tough enemies. To develop something like this, i suggest you expand your testing pool bigtime.

Thanks for the feedback btw :).
Also, the turk strategy in the beta i released is quite bad, the improved version is much better, even it can be improved alot though. Turks are a weird civ to script for (no arbs, e skirms, pike's, but great gunpowder and good stable, you really have to get creative).

User avatar
Promiskuitiv
Waheguru
Posts: 1364
Joined: Thu Nov 10, 2011 1:07 pm

Re: Brititron 1.8

Post by Promiskuitiv »

Yeah that's true. Even though the byzantines could possibly counter everything. But one / an AI would've to realize the enemy plans very early and adapt from the start. With how it is now Promi couldn't ever defeat a good eagle warrior rush (except the Horde's 35 min castle-age ew attack? a bug? but i'm not even sure anymore if Promi won this), even if Promi would detect it early he couldn't alter his strat enough to counter a 18-25 min ew rush (or even most other kind of rushes). Promi trains some trash in the castle-age transition which is absolutely useless against ew's, then (if Promi really saw the threat) he'd build a stable and pump out some knights, but that would be too late.
So Promi tries to react on everything early, but as it still tries to follow it's boom plan to some part it's too weak early (if the enemy rushes with all he has, a halfhearted defense won't work). This problem exists until like 30 min.
An other very big problem is the scouting, Promi will send a scout once he has finished the core buildings in castle-age. He will then completely try to counter the enemy and stay alive while he also silently tries to boom. Works very well if let alive until/after 30 mins. But yeah to come back to the scouting problem: If the enemy attacks earlier, i have no idea what he is doing as i don't have sight of his base. How can i counter an 25 minutes rush like that. ;(

Aside from that there's the problem you mentioned with the counter-counter, I also thought about that, but as Promi uses trash where it can to counter the enemy it will almost always win those trades. Maybe those 50 Paladins wreck all my skirms (maybe some leftovers) and heck, even the spears, but my spears will kill them off more than efficiently. I just need like 7-8 raxes where i pump out halbs then.
Or your l-cav AI will also wreck my skirms, but again, my spears will wreck your scouts and it will be worth it (even though the l-cav doesn't even cost gold ;( ).

In the game against Barbarian my trash flood literally just stopped his units when he was attacking (an endless fight in the middle of the map) until he hadn't got gold any left and Promi won. Against weaker opponents i don't even need to wait for that because my e-catas just breach the enemy lines.

So what i mean is, if our strats both counter each other (for example scouts and archers against spears and skirms) then i will still win cost efficiently. My trash is also trained pretty fast.

Sounds like i'd only use trash, but hell no. For example against Wartron you won't see Promi train trash, just x-bows and catas. (:

Sorry if my text is confusing, still a bit tired. ^^


And yeah i want to expand my testing pool, but there aren't that many UP AI's out there. ;(
Probably gotta test more against Barbarian, at first it will be frustrating though. ^^

But i'm not even sure if i can make Promi do well against good opponents on Arabia at all. ;( I'd probably have to change the strat.
Btw Cake, i tested an "old" AI of mine (before the BF tourney started) against Wartron and it was awesome. This old AI completely counters yours so i didn't expect to lose anyway, but to see 100 fully upgraded Elite-Cataphracts munching through your lines was fun. ^^


Well your AI's are very good and i've already learned much by testing against them. (:

Sorry for the long text. :D

cakemaphoneige
Guru
Posts: 811
Joined: Wed Feb 13, 2008 6:50 pm

Re: Brititron 1.8

Post by cakemaphoneige »

I have this strange feeling that you test mostly against my ai's.

Ai's dont have to be UP compatible to test against them, my ai's still lose to several non UP compatible ai's on UP.

I suggest these ai's:
Horde random civ - try and test against at least 5 different civs, particularly huns.
ARFFI Degel
ER7
Demon
BOSS 3 - plays random civ
Good flushing ai's, particularly gothic ones - you may out play goths in imp but in feudal they will crush you. Look around, there are alot of old gothic flushers, they had tournaments specifically for them.
VNS ai's - Chris, halen, caeser

Test against these ai's for a month and i bet you change your ai dramatically ;)

User avatar
Promiskuitiv
Waheguru
Posts: 1364
Joined: Thu Nov 10, 2011 1:07 pm

Re: Brititron 1.8

Post by Promiskuitiv »

I always test against your AI's until Promi is able to win most of the time (until you bring out your next version). :P

I tested against Demon once and woooow, felt like the most crushing defeat Promi has ever experienced.

Already testing against The Horde from time to time. (:

Will ignore ER7 and Demon (I would've to completely change my AI to beat ER, but i don't want to Promi's strat that drastically and i'll test against Demon once my AI is able to keep up with other good AI's), but will test against the rest. :)
Thank you for the suggestions. ;)

Edit: Btw, yeah, Promi loses against dedicated flushers, already trying to fix that. (:
I also know why the last version is weaker against flushes as the versions before, so i know exactly what to do there. (:

User avatar
UnfairestEel
Waheguru
Posts: 1835
Joined: Sat Jul 30, 2011 4:31 pm

Re: Brititron 1.8

Post by UnfairestEel »

cakemaphoneige wrote:I have this strange feeling that you test mostly against my ai's.

Ai's dont have to be UP compatible to test against them, my ai's still lose to several non UP compatible ai's on UP.

I suggest these ai's:
Horde random civ - try and test against at least 5 different civs, particularly huns.
ARFFI Degel
ER7
Demon
BOSS 3 - plays random civ
Good flushing ai's, particularly gothic ones - you may out play goths in imp but in feudal they will crush you. Look around, there are alot of old gothic flushers, they had tournaments specifically for them.
VNS ai's - Chris, halen, caeser

Test against these ai's for a month and i bet you change your ai dramatically ;)
No. Not ER7. I don't think it's possible for byz to beat this AI :ph34r: :o :head

Still more than likely to improve your unit output though...
Knowledge is Time.
This simple fact can be derived from these beliefs:
Time is Money.
Money is Power.
Knowledge is Power.
Therefore, Knowledge is Time.
Q.E.D.

Clarty
Forumer
Posts: 34
Joined: Sat Jun 09, 2012 2:52 am

Re: Brititron 1.8

Post by Clarty »

This is an interesting AI, I tested against the Chinese and it castled extremely quickly both times (18:25 and 19:05) without resorting to selling stone. I feel its weakest point is probably 25-30 minutes in, when the opponent has siege to deal with the castle but Brititron hasn't built up sufficient numbers to beat them in a head on fight. Also I suggest getting pikemen earlier if you suspect cavalry (huns, franks etc.), because in my tests it had difficulty dealing with large numbers of them.

User avatar
the1unknown
Yogi
Posts: 344
Joined: Sat Jul 21, 2018 3:27 am

Re: Brititron 1.8

Post by the1unknown »

15 pages and no download link

please someone post it, I need it for the tournament
"you don't have to be great to start, but you have to start to be great" -the villager

Post Reply