Wartron V0.95 (UP/1.0c)

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.
Post Reply
cakemaphoneige
Guru
Posts: 807
Joined: Wed Feb 13, 2008 6:50 pm

Wartron V0.95 (UP/1.0c)

Post by cakemaphoneige »

I decided its about time i release Wartron, so here it is.

Wartron v0.9
*Plays goths ONLY. Try and play it with another civ and it will resign
*Plays basic random map settings.
*Built around Farertrons script and alot of stuff from brititrons script.
*Has brititrons TC bug, but only like 20% of the time. Most the time it shouldnt do it.
*Can beat ALOT of ai's out there. But still fails epicly against some. I dont wanna make any claims about which ones it does beat so find out yourself. Mainly loses to really good flushing ai's.
*Once it gets perfusion i haven't seen it lose.
*Chooses between a Huskarl/Elite huskarl flood and a Longswordsman/Champion flood.
- Taunt 15 to it(only when it is player 1/your viewing it) before castle age or before it chooses itself to make it go for a Husk/EHusk flood
- Taunt 16 to it(only when it is player 1/your viewing it) before castle age or before it chooses itself to make it go for a Champ/Lsm flood
- Taunt 20 to it while its in castle age to ask which flood it is going for.
- Default flood is Champ flood if Husk flood isnt chosen by the time it starts researching imperial.
*Will attack in castle age with either LSM/Husk flood if it has military superiority.
*Will wait till it gets perfusion in Imperial age if it doesnt have superiority
*Doesnt Flush
*Its dark age and feudal age eco hasnt been fully refined yet so castle times vary between 18.20 - 19.45.

Any hints/tips welcome.
A few things have been added in the last few days that havent been tested fully so this may return some bugs. But in the few tests i have done they have worked really well. Especially using attack groups instead of attack now. Let me know if you spot anything.

Version 0.91 Uploaded
*Better Flush defense
*A few tweaks done here and there
*Now has compatibility for Aztecs, Vikings and Japanese but is best with Goths. (will do some more work on these other civs soon, they have only been given a few rules for compatibility)
*Can play PIDM with goths, not the best at it though, i made these rules in a couple of hours.
*Counter rules for certain strategies added

Version 0.92 Uploaded
*few fixes with defense and getting certain researches
*other stuff

Version 0.93
*General fixes and tweaking all round.
*Should do a few things better and bug less(the bug is it turns wood spending and food spending off for researches/some buildings and sometimes doesnt turn it back on resulting in epic failtacularness) but still far away from where i want it to be.

Version 0.95BETA
*This is the overhaul version ive been working on(and off) for the last month or so. What i did with this version was strip Wartron down of all its complicatedness and rules that were failing it and have begun building it back up.
*This version does not have the usual Huskarl strategy nor its anti-monk strategy, im yet to get around to re-instating those strategies. A few other counter strats are missing too i think.
*Much stronger eco in early-mid castle age.
*Handles attacks very well. Ai's that used to beat it in there 1st attack wave will either have to fight alot harder to beat Wartron now or will now lose to wartron.
*Has a slowish imp time(due to stronger castle eco) as im about to refine/balance that.
*Still able to play aztec/japs/vikings on top of goths, however for some strange reason thinks its too good to research champion with these civs.
*Will ask you how you like it now when its winning
*Highly advise testing this script.
*Now has UserPatch compatibility, a few UP things put in there.

Version 0.95ALPHA
*Updates and tweaks allround
*Better eco
*stronger

Version 0.95
*Huskarl switch is back and better than ever :devil (still needs work though)
*Trains more vills in imperial age on UP
*Anti monk switch is back in
*General fine tuning
*Attack group lag not as bad (still not great though)
*UP has a "in progress" boar hunting code, due to its incompletion, unexpected things may happen in dark age, but oh well.
*Is very strong :devil :devil :devil

Enjoy
Last edited by cakemaphoneige on Fri May 11, 2012 12:28 am, edited 11 times in total.
rip ai scripters

User avatar
Campeador
Waheguru
Posts: 1340
Joined: Mon Nov 01, 2010 8:11 am
Location: Spain

Re: Wartron V0.9

Post by Campeador »

Congratulations partner. You have beaten my Koreans 3.1.

I've detected only two problems:
a) Unbalanced castle eco. Too little food and too much gold. You could change gatherer percentages or perhaps consider av archers (although they should change your strat and require expensive upgrades, so they are only an option, not a needing).
b) M@a in castle age without long swordmen upgrade during a good while. I avoid it forbidding castle swords until that upgrade, but I recognize that it's risky when you are being trushed so hard. Mainly if you make mostly infantry. Anyway, if you resolve problem a) you'll be able to do this much faster.

You know that my next version will use a FC strategy in walled maps, but it's got a worse castle time than you. Well, you have given me an idea to improve it. Delaying loom. Easy. I think wolves are less problematic in walled maps. I'll have to test it.

cakemaphoneige
Guru
Posts: 807
Joined: Wed Feb 13, 2008 6:50 pm

Re: Wartron V0.9

Post by cakemaphoneige »

Sweet :). I havent actually tested wartron against crusade.
They sound like good ideas, ill look into them. Although its hard to get enough food in castle because of ai's farming capabilities and plus every unit requiring food, even villagers.
I just had a thought. Getting LSM first instead of forging + scale mail, what you reckon? Maybe prioritize scale mail as first smithy upgrade afterwards.
rip ai scripters

User avatar
offwo200
Guru
Posts: 529
Joined: Mon Feb 21, 2011 9:31 am

Re: Wartron V0.9

Post by offwo200 »

Two things:

1) Loosen your drop-distances. To 99% of other scripters I would say the opposite. Your wood/hunt drop distances are way too low imo. I've seen your script bug on non-standard maps even though other scripts play them fine. Hunt should be 22, wood should be 18 imo.

2) Force 30-32 vills before you click feudal. Will this slow your fastest time down? Probably. But this allows you to mass more resources during feudal transistion and click castle overall quicker.
those hackers are just idiots, i spit on them :lol: - Mabuse

cakemaphoneige
Guru
Posts: 807
Joined: Wed Feb 13, 2008 6:50 pm

Re: Wartron V0.9

Post by cakemaphoneige »

Hmm ill look into those things, thanks offwo :)
rip ai scripters

User avatar
jdstankosky
Guru
Posts: 908
Joined: Thu Nov 25, 2010 12:30 pm
Location: Florida, USA
Contact:

Re: Wartron V0.9

Post by jdstankosky »

cakemaphoneige wrote:Sweet :). I havent actually tested wartron against crusade.
They sound like good ideas, ill look into them. Although its hard to get enough food in castle because of ai's farming capabilities and plus every unit requiring food, even villagers.
I just had a thought. Getting LSM first instead of forging + scale mail, what you reckon? Maybe prioritize scale mail as first smithy upgrade afterwards.
Here's my thoughts on this:

If you get the blacksmith upgrades, you can continue to train M@A at the same time from ALL barracks. If you are flushed in this time, you will have maximum soldiers with far superior upgrades than your flushers.

If you get LSM before the blacksmith upgrades, you are limiting your available M@A for a minute or two. Crucial minutes for flush defense. I know that SRN's LSM fast castle strat for Aztecs manages to get ALL available infantry upgrades done within minutes of arriving at castle. You can modify the "predefined index.per" to only select the LSM aztec strat every time if you want to see and study my feudal-->castle process. I'm sure you can port the same ideas into Goths. I've done it to my own before with great success. Not sure if I uploaded those goths though.

[EDIT]
SRN Goths are indeed a clone of the Aztec ones. Because of the differences between civs, results might be slightly different from eachother, but overall, similar.
Last edited by jdstankosky on Wed Sep 07, 2011 9:04 am, edited 1 time in total.
Do not try to script a counter for everything. That is reactionary, and always slower than taking the initiative in battle. Decide from the start what you are going to do, and do not waiver from it. Own it.

User avatar
Campeador
Waheguru
Posts: 1340
Joined: Mon Nov 01, 2010 8:11 am
Location: Spain

Re: Wartron V0.9

Post by Campeador »

Dark villagers to FC. If you need more than three villagers (which can be replaced by WB) in the feudal age to avoid idle TC, it means that you train to little dark villagers.
I think that I've watched The Horde training only 29 with a good result. But Archon is Archon, and he has optimized his FC to an incredible level. In general 30-32 is more recommend. I started with 30 dark villies for my new FC strat (for walled maps) but it didn't work because I need tons of villager training in feudal. So I finally made a 33+3 villagers for my FC, with a 20:XX castle time. I know it's very slow, but I don't build market+blacksmith, I prefer barracks+blacksmith+archery which I consider more adequate for walled maps (and because I'm a noob making FC strategies). At least it arrives to castle with a strong and competitive eco.

Jd has given me an idea. I agree him about research the first attack and armor upgrades of the blacksmith. If you are flushed they will be useful. About LSM tech, I stop swordmen production once I arrive to castle, but I can do it without any fear because I arrive to castle with many villagers and I train other soldiers in other military buildings. For Wartron it could be dangerous, but... what about do it once you detect that you have "almost" the needed resouces to take LSM. It should avoid to have 299F and +65G and wait more for LSM because you train m@a from two barracks, arriving again to 179F. Of course, one possible problem could be if your villagers are garrisoned because an attack and can't farm. I do something similar to stop feudal military training and accelerate my castle time.

User avatar
marathon
Waheguru
Posts: 1976
Joined: Fri May 14, 2010 2:30 pm

Re: Wartron V0.9

Post by marathon »

Just thought of an idea:
To use a timer for upgrades. I mean, you enter castle, you stop training in order to take bow saw and LSM asap, then you start the timer. For like 90" you train units like mad, then you take a break for chain mail and start again, and so on.
Just an idea.

cakemaphoneige
Guru
Posts: 807
Joined: Wed Feb 13, 2008 6:50 pm

Re: Wartron V0.9

Post by cakemaphoneige »

jdstankosky wrote:
cakemaphoneige wrote:Sweet :). I havent actually tested wartron against crusade.
They sound like good ideas, ill look into them. Although its hard to get enough food in castle because of ai's farming capabilities and plus every unit requiring food, even villagers.
I just had a thought. Getting LSM first instead of forging + scale mail, what you reckon? Maybe prioritize scale mail as first smithy upgrade afterwards.
Here's my thoughts on this:

If you get the blacksmith upgrades, you can continue to train M@A at the same time from ALL barracks. If you are flushed in this time, you will have maximum soldiers with far superior upgrades than your flushers.

If you get LSM before the blacksmith upgrades, you are limiting your available M@A for a minute or two. Crucial minutes for flush defense. I know that SRN's LSM fast castle strat for Aztecs manages to get ALL available infantry upgrades done within minutes of arriving at castle. You can modify the "predefined index.per" to only select the LSM aztec strat every time if you want to see and study my feudal-->castle process. I'm sure you can port the same ideas into Goths. I've done it to my own before with great success. Not sure if I uploaded those goths though.

[EDIT]
SRN Goths are indeed a clone of the Aztec ones. Because of the differences between civs, results might be slightly different from eachother, but overall, similar.
This is a very good point. Ill look into SRN's strat.
When i started out making wartron he was meant to simply do a better job of what farmertron does(basically a fast imp and fast champ) but with a tiny bit of military during castle to avoid getting wiped out before it gets to imp which is what fails farmertron most the time. I soon realised that it wasnt going to work cause most ai's will attack in castle and another good percentage will attack in feudal with very strong armies, so i had to make wartron a bit more flexible. At one stage he was getting champion + perfusion at 37 min and launching a massive imp flood at that time (which btw is almost unstoppable, i hold strong in my argument that once he gets perfusion not much can beat him). But this time kept getting pushed back a bit cause i had to keep adding more and more stuff into early castle and mid castle for defense. So thats where he is at now, if he detects a flush, he will get castle researching and then begin training defense immediately (which is very insecure i know), if he hits castle and detects a enemy military bigger than his he will go into a castle attack/defense mode and train a full army in the hope of doing a LSM/Husky flood, but if not will wait normally till imp.

So the whole idea when no threat is detected is to save as much resources as possible(dont research things such as LSM, archer upgrades some infantry smithy upgrades(i might have made him get them anyways i cant remember)) to try and do a fast imp and fast champ/Ehusk. But against 80%+ of ai's these days this will never happen as they all train castle military and trigger him to go into his castle attack/defense mode.

I just wanted to say that so that you all know what wartron is meant to be trying to do.
Im getting to the point where i think i should just stop him from trying to do a fast imp and focus more on his castle age stuff, as the fast imp rarely triggers.

About the fast castle, wartron does 30 dark vills most everytime, wheelbarrow, barracks+blacksmith+archery in feudal.
At one stage i had him doing archery + stable and got him to create a few knights in castle, but they drained too many resources. But come to think of it, they added a big advantage in flush defense too, so maybe ill have to look back into that.

The one thing i want wartron to able to do is beat horde huns, which he has never done. He beat horde celts once(the only other horde civ ive tested against), but im obsessed with beating huns. I did a test match between is machine goths and horde huns and to my amazement machine beat him!, so im probably gonna start studying machine alot. But then it also comes down to should i make wartron so that when he detects a flush, start training feudal defense and try and flush back instead of going for fast castle, which is most the reason he loses to hurde huns anyways, the flush usually always does heavy damage so the next attack is fatal. The idea with trying to get to castle is to try and dish out the huskys to the cav archers, but they usually dont start coming out till 30 min anyways, and the few matches wartron has stood his ground to the point where he has a big husky army, he still lost for some stupid reason anyways!
rip ai scripters

User avatar
jdstankosky
Guru
Posts: 908
Joined: Thu Nov 25, 2010 12:30 pm
Location: Florida, USA
Contact:

Re: Wartron V0.9

Post by jdstankosky »

My best all-around advice is to focus your script on accomplishing one thing, but make it accomplish that one thing extremely well. As much as I'd like to be able to combat in every situation, the fact of the matter is, I'm never going to be able to. That's why my scripts are so weird. I focus on one idea, one strategy, and I go ALL OUT on it. I don't care if it beats 60% or 20% of the enemy AI's, I just want it to be the BEST at what it DOES do.

If Wartron is supposed to be a Fast-Imp Champ rush, then think of how you can make it the BEST at that one particular thing. You'll be the guy to beat when it comes to Fast-Imp Champ rushes, much like Tlama has now become somewhat well-known for being a monk script, or The Horde is known for its Huns. You can make this the best script EVER, but not for every instance. B)
Do not try to script a counter for everything. That is reactionary, and always slower than taking the initiative in battle. Decide from the start what you are going to do, and do not waiver from it. Own it.

Post Reply