This is the official forum for the free, fan-made Age of Empires III modification, Wars of Liberty.

Welcome to the New Forum

We hope you enjoy the official new forum of the Wars of Liberty mod for Age of Empires III! Visit this thread to get acquainted with the new forum, and let us know if there's anything we can fix. Feel free to dismiss this message at any time by clicking the "X" in the top right corner.

Thanks!
The Wars of Liberty Team

Forum rules

  • Welcome to The Town Hall! This forum is for news and discussion about anything related to WoL. You can ask any WoL-related questions here.
  • The official language in this forum is English.
  • Be nice to other members, don't spam comments or smilies, and don't do anything illegal.
  • Posts that don't fit this forum will be moved or deleted.
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Fri May 17, 2019 10:53 pm

Finally, I’ve found a dead moment to write! (not really, it’s just that I didn’t plan a lot for today) I had a field trip yesterday that lasted 14 hours, so I could use some light like this before preparing myself for the exams.
This will be the start of the second round matchups. For those who don’t know what this is about, or for those who forgot who fights who, go to the first two posts of this thread and you will be answered.
 
Portugal VS Argentina
I bet on Portugal that they will win. Why? Spam Ordonenza and the Argentinean vils are dead. The cavalry tag on them does more harm than good, but at least they benefit from the Riding School, giving Argentina nice booming potential. With that, I’m afraid that’s going to be history, so the cavalry tag will get reduced to being a nuisance for the civ without any other advantages.
After 2:44 minutes, we have our winner. Argentina has gathered a treasure and obtained Our Savior Tibs. GG.
Now for real, Tibs isn’t that strong ingame. After a bit more than 3 minutes, Portugal goes II. That’s very fast. Argentina is the first to get III, in 8 minutes. The first fight happens a bit later, at the site where Argentina is going to put its fort. The fort wagon approaches as the fight is nearing its end. But the Portuguese win the fight, and the wagon was unable to build the fort as the Ports were standing on the foundation, so the fort wagon got destroyed. The Ports go III soon after this, and build their fort on the exact some location Argentina planned to build theirs. The second fight takes place here again. This time several Monteneras or so storm to the line infantry, while the Portuguese fort wagon is approaching the foundation. Will we see the same thing again? No, as the Portuguese decided to attack with the few line infantry, causing those horsemen to return back and kill them. The fort had been built safely. The Portuguese perform the attack on a base, and they have a lot of Ordonenzas, and Argentina has nearly no army and no towers. This looks bad. But why are they so weak? Answer: they have made an army to attack the fort. That army is even much stronger than the army that is attacking them! The Ports are greatly damaging the Argentinean base meanwhile. Ordonenzas are decent units, they are weak and have low HP, but their siege is great, and they tank hits for the other units. Argentina on the other hand failed at destroying the fort. 15 minutes in, and the Ports have been killing vils for like 2 minutes in the enemy base. That small army is only getting damage from the TC. Minute 17, and Argentina finally has a tower, while Ports are still in their base. A new Portuguese army approaches. They attack a few buildings in the back of the base, and Argentina send the Voluntarios shipment (hint: waste of coin). The Ports noticed those units, and started attacking them, and defeat them without much problem. Somehow, they task their army back to the fort instead of continuing attacking the Argentineans. Minute 23, Portugal starts a new attack, and defeats the Argentineans. Argentina lost largely because of my prediction: their vils died to the Ordonenzas. I want to mention that AI BUG: Portugal doesn’t upgrade Ordonenzas, even if they get a substantial amount of them.
 
Belgium VS Mexico
This time, Belgium doesn’t start with the extra TC wagon. The AI is still the same as the last one, so it won’t use the free frontier wagons. Minute 9:40, and Mexico has quite a large army heading towards the Belgians. The army couldn’t beat the defending Belgian forces. Two minutes later, and a new Mexican army is approaching. Note how the Mexicans have thrown a lot of gold to those shipments to send the Apache and Navajo allies. Meanwhile, the Belgians are also sending a small force. Just when Mexico went III, the Belgians arrived, but the Mexicans had enough army there to fend them off. 14:33, and a trade routes passes through a gold mine. Somehow that mine spawned on the trade route. Both civs are also III. At minute 15, the Mexicans attack the army that was meant to defend the Belgian fort wagon, but the Mexicans proved to be stronger and defeat all Belgians, including the fort wagon. I have the feeling that the insane amount of HP on the Insurgente greatly helps the Mexican AI. The Mexican army reaches the Belgian base, and starts doing a lot of damage as the Belgians have almost no troops and the towers are not upgraded. A bit later, a new Mexican army arrives, albeit a small one. A shipment of rangers didn’t quite accomplish a lot. Minute 20, and more Mexican troops arrive. Mexico is putting a lot of pressure on my home country! A minute later, and Belgium finally has upgraded its towers. Now the Mexicans will lose troops a lot quicker. And indeed, at minute 23 Belgium has no Mexicans in its base. But two armies approach the base two minutes later. The Belgians provide a surprise to the Mexicans as they have shipped the Pavisier shipment (which gives them 12 of them!). Minute 27, and Mexico is IV. Somehow Belgium thinks it is a good time to attack with the remaining pavisiers and two hussars. So few units won’t help you defeat someone who has triple your score. An enormous Mexican army intercepts them, and they storm into the Belgian base, which prompts Leopold II to accept his defeat. This was an easy victory for Mexico. I fear that Mexico will lose next time, as these games will be run on the new patch, and in that patch the Mexicans will have to use a wagon to build the Presidio. And AIs have trouble using wagons.
 
Serbia VS Italy
Serbia had a poor showing against Bulgaria in the previous round. I hope that they will perform better now, but somehow I have my doubts. Serbia is the first to reach II. Before minute 3, which is very spectacular! Italy follows 30 seconds later, making the advantage rather minor for Serbia. Minute 9, and Italy is age III. Serbia has an attacking army a minute later, which ignores the fort foundation (they might also have not seen it). The army starts attacking some Italian vils. Another army of Serbia had noticed the fort, destroyed the troops that were protecting it and during the building phase they damaged the fort. The fort however wasn’t destroyed. Serbia went III during that attack. Minute 13, and no fighting is taking place. Italy had lost quite some vils, but they do have two TCs and a fort. Two minutes later, and a small Italian force is raiding the Serbian vils that are hunting. The AI hasn’t been programmed to make use of crops, so hunting is very important for the Serbians. Without it, they would have food issues. A small Italian army, accompanied by those raiders, attacks the empty Serbian base. Serbian caserns don’t shoot unless upgraded, and that’s a big weakness. A bit later, when that small force had been defeated by TC fire, a larger Italian force arrives. It kills nearly all villagers, but a macemen shipment helped defeat those forces. Somehow Serbia thinks that it is a good idea to put 5 vils on praying, when they have nearly no vils. Minute 24, Italy is IV, while a new Italian army is sightseeing in the Serbian base. They find it rather dark, so they decide to place some torches for light. Good idea, as they prevent creepers and zombies form spawning there. So nice of Italy. King Petar, greatly moved by this kind act, wanted to send a letter of gratitude, but accidently send a letter of defeat.
 
Japan VS Austria
Minute 1:10, and a BUG happened. A Wild Boar treasure guardian got attracted by a shrine, and started attacking the shrine. I had to intervene to stop this bug, as I considered it to be critical enough. Because of that, Japan has a shrine with 2/3 of its HP left. After 3 minutes, Japan is III. Funny detail: Japan prefers gathering the wine berries instead of gathering from the cherry orchards. Austria does a rather late age up, at 4:14, but that’s not too bad. The first fighting of the game happens at minute 7:40. Welp!  That is early. Austria has literally 0 troops to fight those units. However, Austria goes III soon during this attack, and a shipment+tower saved them, but they did lose quite some villagers and gathering time. I think Austria took the English princess, so the villager count may recover quickly. Minute 12, a new Japanese army attacks, and Austria no army. A Schloss wagon gets destroyed and most buildings too. The Austrian AI also prefers to have villagers walking close to the Japanese troops than hiding them in the town center. A shipment saves Austria again, and it also defeats some raiding troops. Note: Japan is still age II, and has fewer points than Austria. I also made a mistake apparently: the Schloss wagon didn’t get destroyed. But it did get attacked, and somehow it is just standing still. At 17:40, Japan is finally age III. Definitely one of the slowest age 3 I have seen. To make things worse, a few dopplesoldners are raiding the Japanese vils. Multiple samourai were needed to defeat it. At minute 20, Austria performs its own attack against the Japanese. It is a small army, but it is mainly composed of the very strong dopplesoldners. AI BUG: it is almost like the Austrian AI only builds these. Austria made use of this distraction to build the Schloss near the Japanese. However, the Japanese mustered a nice attacking force a bit later, and demolished the fort. Minute 23, and Japan is IV. Minute 25, and two Japanese armies are attacking Austria: one at the front of the base, and one at some hunting villagers. Two minutes later, a new force arrives. The dopples are having trouble, I think that there might’ve been a few ninja’s in the Japanese army. Minute 29, and Austria got defeated. Poor showing of Austria if you ask me. They refused diversifying their army, and had constantly a small or no army. They did not suffer from a shortage on resources, as they had almost always more vils than the Japanese as they must’ve taken the English princess.


Summary: Portugal, Mexico, Italy and Japan go to the next round.
I got everything right, so +8.
David has 3 games right, so +6 
 
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Sun Jun 02, 2019 11:29 am

Sorry for the long time that I didn’t post anything. But I have exams. And I also have a task I have to finish. I’ve found some time to write, so here we go.
 
Ethiopia VS England
I thought that Ethiopia would win. Both civs have strong AI’s, but the Ethiopian buildings get so tanky that it is almost impossible to beat. Before you think I spoiled everything with this: I’ve also found a major ai bug, and it’s Ethiopia who’s affected by it.
Minute 3:04, and both civs are still age 1. For a European civ, that’s not that unusual, but for an African civ it is. You wonder why Ethiopia is so slow? Well: they didn’t pick up their food crates. 600 food is just rotting near the TC. 4:29, and the Brits are the first to go II. That’s late, but Ethiopia is performing even worse, only aging at 4:39. There are at least 11 food crates near their TC! No worries, they will eventually pick them up. 8:34, and Ethiopia is III. Their economy found some new breath after collecting those crates. A small fight occurs in the Ethiopian base few seconds later. Minute 12, the Brits attack again, and age up to III at bit after. Note how at this moment the Ethiopian Garrison has already 5k HP, the tower building has 6k HP and 48 attack. Ethiopian defense is really monstrous. Minute 14, Ethiopia goes for its first attack. This was also the very moment that England decided to send a fort wagon, and even worse, they also decided to send in a lot of villagers forward for hunting or mining. The fort gets destroyed, and many villagers died. But Ethiopia couldn’t achieve much more with that attack. Minute 20, and Ethiopia attacks again. The Brits have an enormous amount of Highlanders, who were a bit OP in the previous patch. The attack went nowhere, but at least Ethiopia went IV. Three minutes later, new Ethiopian attack that goes nowhere. At the same moment, the Brits too were performing a futile attack. They are also age IV.
AI BUG: Ethiopia doesn’t use rock crates from the Quarry. Yeah, that one is major. Any African AI should just max out rock crates when it is IV, no planning needed. But that bug will have major consequences for Ethiopia! Even worse, they are behind in score. It’s only minute 27 though. This time, Ethiopia goes for a different angle and dangerously enters the British base while going Imperial. The assault didn’t go further than killing a few villagers and destroying a few buildings. Even the reinforcements didn’t help that much. Minute 31, another futile attack. The Brits, who demolished the Ethiopian army, decide to go for a counter attack. Guess what? The Ethiopian base got shredded!
Ok, I’m kidding, futile is the most fitting adjective in this whole match. The Brits tried attacking every 3 minutes with a very large army for like 21 minutes, but you guess it: futile. Minute 50, and the scores are: 1493 for Ethiopia, 1246 for the Brits. Minute 50, for the first time in IDK 25 minutes, Ethiopia performs and attack. But for what? To lose points because you’ve spend resources (economic score) and units (military score)? Good job, Menelik! Minute 57, and the Brits are performing a very nice attack. I think more time was needed to deplete the Ethiopian gold reserves. Anyway, the game lasted 1h, so let’s compare the scores!
Ethiopia: 1877
Brits: 1548
Congratulations I guess? -_-

Aztecs VS Romania
Sometimes some of the objects start on weird locations. This time, some cypresses are growing on the trade route. Anyway: AI BUG: Mandos’s best friend, Vlad the Vampire, has been defeated by some renegade Janissaries. Reason: the AI thinks that is has quite an army, as it has an explorer + a bear + two ravens. But the AI has no crackshot, is already bad at fighting treasures, but more importantly, ravens aren’t particularly useful in battle. 4:08, Aztecs are II. At 5:48, Romania goes II. That must be the slowest II so far. GG Romania? Minute 10, and the Aztecs, who are apparently already age III,  perform a minor raid. However, they raid the bears in the circus, who provide no benefit as the dance is bugged. The Romanians had some difficulty defeating those few troops, but that’s purely because they are performing a nice attack themselves! A shipment saved the Aztecs. Minute 12, both are III, and if score would be a good indicator, then you would say the Romanians are winning. Minute 16, and the Aztecs invaded the Romanian base. When this attack was nearly stopped, the Romanians received the Mameluk shipment and decided to do a new attack. But the Aztecs are good spammers, and Romanians didn’t achieve anything. Romania builds the Star Fort in his own base. AI BUG: at minute 22, the Romanian market has not been upgraded yet! Major bug! Minute 23, both civs encounter each other in the middle of the map, which resulted in a narrow victory for Romania. Three minutes later, a new Romanian army attacks, but their infantry are just much weaker than the Aztec troops. AI BUG: Balkan AI’s don’t know how to use crops or geese, which means that when hunts are gone, they will run out of food. Now, Tuscany has quite some berries, but even if the AI is getting into food troubles, only 3 villagers are gathering food from them now. Minute 30, and both civs attack each other. The Aztecs kept quite an army in their base, so the Romanians didn’t do as much damage as the Aztecs did to them. Note how both civs are still III. They are putting a lot of pressure on each other! Minute 32, Aztecs finally go IV. From this moment onwards, Romania has no food at all, and couldn’t do much more than doing TC fire and replacing some buildings. Romania struggled for 15 minutes before being defeated. Romania lost this game purely because of lack of food. Maybe they could’ve won if they had built artillery to counter the Aztec infantry, and if they had gathered more from those berries.
 
Paraguay VS Russia
Minute 8:30, and the first attack happens. The guanco riders or so, one of the native shipments of Paraguay, destroy the Russian army and start attacking villagers. Too few of them had survived to really be dangerous to them. Two minutes later, Paraguay attacks again, but the Grenadiers are too effective against them. I have to note that 16 ranged attack on a 3 pop unit is atrocious. Minute 12, and both civs are III. The Russians escort their  fort wagon, while the Paraguayan escort is quite far behind their fort wagon. Eventually a clash happens, and both fort wagons are destroyed, and few Paraguayan military units survived. A new, large army invades the Russian base. AI BUG: the Mestre sometimes starts building a building, and then goes away for quite some time, to then finish the building minutes later. I don’t see any use in that tactic. Anyway, in this case it is their first stable that got delayed. And that’s a major disadvantage for them! Minute 20, and Paraguay is IV. A minute later, and somehow the Paraguayans are dealing a lot of damage against the Russians. I wonder if the Russians know how to build troops. Minute 27, the Russians find some men stupid enough to attack the Paraguayans. Meanwhile, the Paraguayans find a new tactic: artillery raids! With seven Cohetes, they attack the Russians. Hint: don’t do that. Minute 28, more Paraguayans enter the Russian base, while Team Paraguay also advances one age. Criolle canon incoming! The new troops that are coming will probably have an easy time: there are about 0 troops in the Russian base. Many vils, no troops. Another major thing is that the Blockhouses haven’t been upgraded. In most of the previous AI games, towers played the major role, but if you don’t upgrade them, then you’re doomed. Poor Russians. The aforementioned army with the Criolle cannon obliterated the Russians and destroyed their impressive base easily.


Summary: Ethiopia, the Aztecs and Paraguay win
That's +5 for me and David.

Don't expect a new part any time soon. But I will finish this, don't worry.
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
DavidIbragimov
Independentista
Independentista
Posts: 147
Joined: Tue Apr 09, 2019 6:26 pm
Location: South Carolina, USA
Favorite Civilization: Romanians

Re: AI game (or at least a try)

Wed Jun 05, 2019 1:49 am

Alright, while you're away, I will do some total sum recounts. Goodluck to whatever you're doing Ossus and come back soon!
Breadfruit.
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Sun Jun 30, 2019 9:46 pm

DavidIbragimov wrote: Alright, while you're away, I will do some total sum recounts. Goodluck to whatever you're doing Ossus and come back soon!
I'm back! After my most stressful uni exams I've had for now.

Korea VS Incas
Korea surprisingly got to the next round, and now they face the scary Incas. I think Korea is doomed to lose, as their cavalry (somehow they spam a lot of cavalry) will die against the Incan Huamincas. Important note: this was recorded on the previous patch.
Things start out normally, with both civs aging before 4:20. I think Korea took the whirlpool wonder, I haven’t played with Korea enough to know it well. If it is the case, then Korea made quite a big mistake this early. AI BUG: the Incas killed some treasure guardians but tried picking it up with the Incan king, which is impossible. Maybe that unit should be programmed like the envoy instead like the explorer? After 10 ingame minutes, the Incas perform the first attack, and they have a nice looking army. The Koreans on the other hand have nearly no troops, and didn’t retreat their vils in time, which means they lost quite some of them. Meanwhile, they are just starting to make a new wonder. Minute 11, and the Koreans are III. If I recall correctly, this is the point that they will start constructing cavalry. Pls be quick Korea, win before the Incas unleash the spam! Minute 14, and the Incas are sending a new army. The previous army had only just been defeated with very few troops left on the Korean side. Even worse for the Koreans, the Incas are III too. A Tartar archer (the mercenary cav archer) shipment saved the Koreans though. At minute 18, the Koreans are advancing with an army, while their base is under pressure. Sure, the pressure isn’t menacing, problem is, the Korean AI doesn’t know to build towers, so it has to deal with the Huamincas with cavalry. The Korean attack was a total disaster btw, it did nothing. Minute 22, and things don’t look good for the Koreans. Quiet an army is now attacking them, while they have nearly no buildings or villagers left. And indeed, after 23:14 minutes, it’s gg. The statistics show what was to be expected: the Koreans have only had a small number of military units, and the KD ratio is terrible. Add to that the constant attacks that damage the economy, it’s a guaranteed failure for them.

 
Tupi VS Germany
Germany goes II in 3:17. That’s quiet fast! Uhlans incoming, the Tupi are doomed! Oh wait, the Tupi aged up a few seconds later. Nevermind. Minute 7:30, and the Tupi perform a first raid. Not a lot of units are involved on both sides, but it does set the pace of the game: rush rush rush against defend defend defend. But it appears that the Tupi send some reinforcements. They destroy the barracks as a result. Then it becomes clear why Germany is starting so badly: they age up at 8:30. A shipment saves the Germans from losing any more vils or their TC. But a bit later, at 9:30, a new small group of Tupi soldiers attack. The Germans meanwhile have nothing guarding the base, as they are sending their own army to the Tupi. Both civs had quite some casualties because of those attacks. Then, at 12:30, I spot Voluntarios. Really? If I were leading the Tupi now, I would spend my resources on aging up instead of paying for a shipment that gives rather weak troops. A few seconds later, and the Tupi go for another attack. While that army manages to kill the German fort wagon, a German army went for a sneak attack, and actually manages to kill quite some vils, mostly because the Tupi have absolutely no defenses. The Tupi army seems equally effective at killing villagers, and it took the Germans several minutes to clear all of them. 20 minutes in, and the Tupi have worryingly few gatherers on the Ocas. Even worse: AI BUG: still no longhouses, like last time, since it’s the same version. 21 minutes in, and the Tupi raid the Germans again. I wonder how the Germans can even replenish their villagers, they are losing them à volonté. However, the units little little to the German base. Actually, I find that the biggest weakness of the Tupi: their siege is atrocious early. That’s one of the reasons why good Tupi players tend to go fast fortress, to get the pombero for sieging. Minute 23, and the Tupi finally go III. Is this the slowest III we’ve had? IDK. Anyway, at the same minute the Germans went IV! I think this will be unwinnable for the Tupi. An attack a minute later did not help, and the counter-attack of the Germans did quite some damage. The Germans are now much more aggressive. AI BUG: minute 32, the Tupi AI has many idle gatherers. AI BUG: the Tupi AI is sending gatherers to kill deer. Or watch them? Anyway, it doesn’t give them resources. Minute 36, after several small and insignificant attacks of the Tupi, the Germans send a massive army to the Tupi. Even if that army retreated after some minutes, the army came back and demolished the Tupi after 43 minutes. What a surprising end! The Tupi were very close to defeating the Germans, they had lost so many villagers, their economy must’ve been atrocious during most of the game. However, they never managed to defeat the upgraded towers, which eventually caused them to give free kills to the Germans. Advancing that late also made things worse. Poor performance from the Germans too if you ask me, they had not made a lot of military, and they had built the first Krupp factory after about 30 minutes.
 
Malta VS Colombia
Weird matchup, I had no idea who would win. Both civs age up at normal times. Minute 8, a tiny army gets defeated by Malta. Minute 10, both civs are age 3. Not a lot has happened for now. Since both civs age upped at around the same time, they also place forts around the same time. And the AI thought it would be good idea to place the fort in the middle. But if two enemies to this strategy, then both forts will be built at the same location! This is a race now! The Colombian prefer to send an army to attack the Maltese, instead of guarding the fort wagon. Smart, that way the fort wagon won’t be in the middle of the fight. And it appears that Malta is doing the same thing. To make things even more chaotic, the two armies passed each other because of not being in line-of-sight. The Colombians start sieging the Maltese base, while the Maltese troops have intercepted the Colombian fort wagon. The Maltese have won the site, and their reward is a fort. The Colombian army meanwhile got defeated. Minute 13, and a Maltese army storms to the Colombian base. Not to destroy, not to raid it, nor to sightsee. They only want to kill the two carabineers. That’s it. Three minutes later, and it seems that the Colombians have trained some troops. That’s not tolerated! So the Anti-Aggression Squad (AAS) goes back to the rebellious city and kills the traitors and burns down their hideouts. However, the rebels that get out of there seemed quite strong and defeated the AAS. Minute 23, a Maltese army decides to burn down the city, instead of sending AASs. They also go IV.
But then. They came. Like a miracle. The miracle that will save the Colombian people. The Ancient Colombian Gods decided to send the box of Pandora. Trebuchets. 12 pop wasted guarding the TC.
Minute 27, and now the Colombians are IV too. They didn’t suffer much villager losses from the Maltese pressure. Minute 32, and Malta is Imperial. There had been a lot of fighting between the two, with the Colombians constantly defending. Three minutes later, and the reduced population space is having a lot of effects on the Maltese: they lost their fort. Three minutes later, while two big armies are clashing, the Colombians go V. AI BUG: seems like the Maltese are still fighting with unupgraded infantry (not even veteran). That’s probably the reason why all their attacks died out so quickly. Minute 41, and the Colombians launch their first assault. The attack wasn’t exactly a huge success, but, it showed that the Maltese are struggling. Minute 45, two large armies clash again in the middle of the map. After the Colombian victory, they went to the Maltese base. More and more troops are assaulting the Maltese. I think that the Maltese are actually having villager troubles. The Maltese keep on surviving thanks to the Trebs, who are behind a wall of buildings that protect them. The Colombians are pressing very hard, a Maltese army is non-existent. But, eventually the timer decided that after 59:59, 1:00:00 follows, which means: points victory for someone. Spoiler: I am the happiest of all, because this match was painful to watch and write about. Anyway.
Malta: 1920
Colombia: 1683

Summary: Incas, Germany and Malta go to the next round. Now that I think about it: wasn't the previous Maltese win a points win too? Pls Malta, don't win the tournament that way can you?
That's +5 for me (points wins... sigh) and only +3 for David.
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Sun Jun 30, 2019 10:09 pm

That finishes the second round! Next rounds to be recorded will be the second chance rounds!
For the reference, I will try to type a small diagram of how it works.
Round X                                --> Losers of Round X  -->                                    Second Chance of Round X
 1
Everyone                               --->          France, Mapuche, Peru, Ottomans, Zulu, India, Sioux, Haiti, Canada, Spain, Greece, USA, Brazil, Iroquois, Egypt, Bulgaria, Australia, China, Netherlands, Chile
|                                                                                                                                             |
|                                                                                                                                             |
2
England, Japan, Portugal, Austria, Aztec, Colombia, Inca ---->        Winners of the second chances + Argentina, Belgium, Serbia, Austria, England, Romania, Russia
Paraguay, Tupi, Korea, Argentina, Italy, Belgium, Mexico                  Korea, Tupi, Colombia
Romania, Serbia, Germany, Ethiopia, Malta, Russia
|                                                                                                                                             |
|                                                                                                                                             |

3
Portugal, Mexico, Italy, Japan, Ethiopia, Aztecs, Paraguay     --->          Winners of the second chances + losers of the third round
Incas, Germany, Malta
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Sun Jun 30, 2019 10:14 pm

First round of the second chances matchups!
Map: Hispaniola
1. Brazil VS Iroquois                       -- Iroquois
2. Australia VS Chile                      -- Chile
3. Haiti VS France                          -- Haiti
4. Ottomans VS China                    -- China
5. Zulu VS Mapuche                       -- Zulu
6. USA VS Sioux                            -- Sioux
7. Spain VS Peru                            -- Spain
8. India VS Greece                         -- Greece
9. Egypt VS Bulgaria                      -- Bulgaria
10. Netherlands VS Canada           -- Netherlands
Last edited by Circle of Ossus on Thu Jul 11, 2019 12:24 pm, edited 1 time in total.
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
DavidIbragimov
Independentista
Independentista
Posts: 147
Joined: Tue Apr 09, 2019 6:26 pm
Location: South Carolina, USA
Favorite Civilization: Romanians

Re: AI game (or at least a try)

Mon Jul 01, 2019 7:19 pm

Alright! New matchups! 

Lets get started.

#1- Iroquois: I say this because of the way the Brazilians like to spam with the voluntarios and aren't so good at working with the slave increase in the capitol.
#2- Chile: Their economy is insane, but also the fact that the Australian AI is having a hard time lately.
#3- France: Their Cavalry is (i believe) one of the best in the game. (WARNING!, will be very painful to get all those pesky sevites)
#4- Ottomans: Ottomans usually get quick armies and have many buffs on units. Weakness in number will probably not do for China this time)
#5- Mapuche: From personal experience, Mapuche win, but I'm not sure about this one, lets see how the computers play.  
#6- Sioux: They will win, if they finish before Age 4
#7- Spain: Really good at early raiding
#8- India: Though the cavalry is weak, they are hecka annoying to get rid of.
#9- Bulgaria: (My Ex-Favorite civ) "If they Survive"
#10- Netherlands: Money can buy you anything, even a victory.
Breadfruit.
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Thu Jul 11, 2019 12:18 pm

First round of the second chances! I will keep this shorter, but fortunately, most games weren’t that interesting (or short) to write long detailed texts about. AI’s that lose in the second chances are eliminated. Since this will be shorter, I will do all 10 games in this one posts. Here we go.
 
Brazil VS Iroquois
Brazil started out decently, and even managed to use the agricultural wagon. The AI is notorious for not using wagons (hence the Italian AI not using them), so this is a great feat of our AI coder Alistairjah! Brazil went much earlier than the Iroquois to age III. Both parties raided each other multiple times, but the Iroquois eventually won the game. The reason is that the Brazilians hadn’t really build many villagers. I suppose that they didn’t research the techs. Proposal: force the Brazilian AI to go III with the first option, as it grants 3 techs that enhance the slave limit. Game lasted 33 minutes.
 
Australia VS Chile
I discovered some weird code in the AI files that prevented Australia from building a good economy. I changed that. I wonder how they’ll perform now. Hint: they tried, but were still horrible. Chile wins. AI BUG: the AI loves attacking the enemy units. If they don’t spot any more units, they tend to retreat from the attack instead of finishing off their enemy. Is this due to the AI attacking with defense plans? Game lasted 32 minutes.
 
Haiti VS France
Haiti the Unraidable versus its old colonial master. Interesting. Haiti performed decently in the previous round, and I expected them to hold out long enough to push the French back. Did that happen?
Both civs start out well. The French perform the first attack, but Haiti survives it. Haiti goes for the counter-offense, while a second army from France goes towards the Haitian town. They don’t cross each other. That’s because Hispaniola has a plateau in the center. The Haitian attack failed to kill a significant number of Pied Noirs. Somehow they had 423 HP!!! Apparently France also bears the adjective “the Unraidable”. The French attack was much more successful, Haiti had nothing to defend himself, and so Haiti lost the game after 17 minutes. I think this is the shortest game we’ve seen so far.

 
Ottomans VS China
China’s second chance. This AI became very fearsome since Alistairjah started coding the AI. In the early days of Alistairjah, the Ottoman AI was known for mercenary spamming, but he reduced that a bit, so I’ve got no idea what the AI produces instead.
After 3 minutes, both civs are II. The AI does things a bit quicker because this map, Hispaniola, contains more starting crates. After 8 minutes, China has already a decent army with horsemen and a Flamethrower. The Ottomans on the other hand have towers and 8 janissaries with three deli and one cavalry archer. That may not look like a lot, but remember, those units are strong. Now that I think about it, the Ottomans received a very minor buff to villager production last patch. Not that it matters though. At 9:30, both civs attack each other. Yes, the Ottomans are attacking with that tiny army. At minute 10, Cixi asks me “Are you not entertained?”. I hope you are. The Ottoman troops, who arrived first in the opponent’s base, are doing quite some damage as they have no opposition (China has only one castle somewhere in the back of their base). But China advances to age III, and gets several skirmisher-like troops, who counter all of the remaining Ottoman troops. Just in time for Cixi! But IDK what happened, when I looked back, I couldn’t see those units again, and the Ottomans were still there. The Ottomans on the other hand didn’t suffer a lot from the large Chinese attack. After about 4 minutes, the Chinese managed to defeat the first wave of Ottomans. At least they are now ahead in age? Well, no, the Ottomans aged up after the Chinese attack had been defeated. At minute 16, the Ottomans have made a fort at the foot of the Chinese base, and a few moments later, they start attacking the Chinese again, and eliminate the Chinese after 24 minutes. Very impressive game from the Ottomans, but more importantly, a very shocking elimination of a notoriously strong AI. Or is the Chinese still good? This claims otherwise!

 
Zulu VS Mapuche
For me, the game would be decided by the Mapuche AI spamming the anti-infantry unit or not. That would be the best choice against the Zulu, who mostly rely on infantry, but if they don’t, then the Zulu might run over the Mapuche as they aren’t too good at spamming units.
While the Zulu aged up early, the Mapuche somehow went II after 4:36. That’s very slow on a map where you start with extra crates. At 6:41, the Zulu perform a first attack. The Mapuche seem to only have cavalry to defend with. Against spearmen. Yeah, RIP those cav. The attack didn’t do much TBH, but the Mapuche did lose more resources than the Zulu on units. 10:36, and the Zulu are Industrial. OMG. That also means they can ship their super strong Ishikulu against a civ that is still age 2. XD XD XD It didn’t stop the Mapuche from performing an attack at 11:30 though, but that attack got stopped quite easily. At 12:44, the Zulu go on the offensive again. With Ishikulu. Yeah, the Mapuche aren’t training the unit they should train like at all. Somehow the Mapuche survived the Zulu attack, albeit barely. They may thank their war huts for that feat. Two minutes later, the Zulu come back, with more Ishikulu. This is over I think. Guess what? It isn’t over. With barely 8 buildings left, the Mapuche survived that attack too. No worries Zulu fans, more is coming! The Zulu AI thought that it was good to run through the base, under a TC, and to then attack all buildings except for the TC. The Mapuche survived the attack again, with 6 buildings left. Oh look, more Zulu units are coming. They have been sending small armies like every two minutes now! This is insane! At minute 19, the incredible happens: about the only remaining building is the TC, so the Zulu AI decided to retreat to their base. 21:15, the Zulu go Imperial while the Mapuche are still age 2. Minute 23, a gaint Zulu army with Revolving guns are attacking the hunting Mapuche villagers. I notice that many of the garrison units have no upgrades at all. At this time, the Mapuche are III too. At 24:33, the Mapuche base is practically destroyed, with the TC having lost 1/3 of its HP. But the AI decides to retreat again. AI BUG: African AIs don’t build rocks at the Quarry. Minute 30, the Mapuche place a fort near the Zulu base. Eh? You’re supposed to defend your base, not annoy the Zulu who have an army that can run over your base ten times! Anyway, the Zulu don’t react to the presence of the fort, somehow. Minute 37, and the Mapuche are Industrial. AI BUG: I don’t see any farm in the Mapuche base. I’ve got now no idea why the Zulu didn’t attack the Mapuche again in this 10 minutes period. They have the troops. But somehow the AI acts like it has been completely disabled. The Mapuche meanwhile keep sending armies of white horse dudes and native blowgunners. The Zulu AI does exactly nothing the whole time. They don’t even defend. If the troops get into range of the Zulu  units, the units attack of themselves, but no military unit of the Zulu gets moved. The only way I can see the Zulu win this matchup (which they should already have if they hadn’t retreated multiple times) is by forcing a score victory. But that will be difficult. At minute 55, the Mapuche AI is still pressing a lot, the Zulu build barely any troops, and have about no towers. They do have a lead. 680 compared to 623. But both are extremely low scores for a game that lasted almost an hour. But alas, the Mapuche quickly jumps over the Zulu in score and even manage to destroy the TC two minutes before the points victory. This game was the most WTF I’ve ever seen. And that just after the already OMG game where China got eliminated! At the endgame graphs, I spot something strange: the Zulu never went over 35 villagers. This must have been partially the reason why they lost, at least if they had more villagers, then they could force the points victory. Also: AI BUG: Zulu AI doesn’t build matriarchs. That’s a big one!

 
The States VS the Sioux
The USA AI is quite docile, like all Anglo civs. They did nothing worth commenting about in this matchup, and lost after 26 minutes. Could’ve been earlier if the Sioux didn’t retreat their army in the attack preceding the last attack. I had high expectations of the Sioux, and they did do well this game. But I am worried though: they hadn’t built a stable this game!!
 
Spain VS Peru
Both civs performed a first attack at the same time, at minute 10, and came across each other, resulting in the expected defeat of Peru. Breneros suck if you don’t kite. But it seems that the Spanish had another army, and that army is harassing the Peruvian base. Minute 12, and Spain is III, the Peruvian base meanwhile feels empty, even if it has two town centers. A minute later, the Peruvians age up too, and some breneros got shipped with the age up. A bit later, the AI ships the grenadiers+decoys card, finishing off the lone Spanish soldier that was still attacking a dock. But then they discover something: the Spanish had set up a fort near their dock! That costed the lives of multiple fishermen. Minute 16, Spain attacks again, with a small army The Spanish continued pressuring from their base, and destroyed the Peruvian original TC after an unknow time, as the timer disappeared. Probably minute 26. That was a very poor game from Peru, who usually does decent in the early game.
 
India VS Greece
Greece did nothing at all and lost. Reason: they didn’t build a dock. I don’t even know if they had built any troops. The game lasted 20 minutes, and India could’ve ended it sooner if they didn’t retreat multiple times. Greece did not go over 20 villagers. Ouch.
 
Egypt VS Bulgaria
I can summarize this in a few lines:
- Before recording this, I’ve tried a few games to see if Bulgaria does train soldiers. They do. So when they go IV, they might become fearsome. I did see that they weren’t too keen to attack with the trained troops though.
- Bulgaria had access to the native settlements, and did use them to train the Cariboo Blowgunners.
- From the previous AI game, I’ve seen that Egypt doesn’t train troops from the caravans. In the past they did, so no idea why not now. And in this game, it wasn’t different. This means that the AI has to rely on shipments and fellahs from age ups. AI BUG!
- But fellahs are strong against villagers, and received a major siege buff (finally). Stalling out will probably not work for Bulgaria.
- Bulgaria did indeed lose a lot of vils early so many that they went III too slow and got overrun by the Fellahs after 27 minutes.
- Bulgaria had send 2 ship shipments instead of other shipments, which slowed them down a lot and is probably partially the cause of them losing so much in age II. If possible, pls remove the water deck from this AI.

 
Netherlands VS Canada
The Dutch went II after 5:13. Excuse me, WTF? You have extra crates dude! Well, that’s rip the the Dutch I guess? Or not. Anglo AIs suck. The Canadians started out decently aggressive, but the Dutch made an incredibly beautiful base, well defended by towers. The Dutch attacks encountered little opposition, and defeated the Canadians after 25 minutes. The most impressive thing the Canadians did, was using a botanist to gather treasures in age 2. That’s actually a very good thing!


Summary: Iroquois, Chile, France, Ottomans, Mapuche, Sioux, Spain, India, Egypt and the Netherlands go to the second round of the second chances, and will be joined there by the losers of the second round, namely: Argentina, Belgium, Serbia, Austria, England, Romania, Russia, Korea, Tupi and Colombia.

Regarding the guessing game: this was a very bad round for me!
I had only 5 of them correct, which is +10 points for me. David had 9 of them correct, granting him +18 points. This means that David is leading O_o
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
Circle of Ossus
Topic Author
Volunteer
Volunteer
Posts: 461
Joined: Fri Mar 31, 2017 9:48 am
Location: Belgium
Favorite Civilization: Colombians

Re: AI game (or at least a try)

Thu Jul 11, 2019 12:35 pm

Second round of the second chances duels:
Map: I haven't decided yet. I hope Red Beach, if it is fair. EDIT: It is Red Beach
Duels:
1. France VS Belgium     --- France
2. Ottomans VS Egypt    --- Ottomans
3. Tupi VS Netherlands   --- Netherlands
4. Chile VS Serbia          --- Chile
5. Colombia VS Argentina --- Colombia
6. Spain VS Korea         --- Yes, this is exactly the same matchup is in the first round. So I guess Korea wins this?
7. Sioux VS Iroquois      --- Sioux
8. Russia VS Austria     --- Austria
9. England VS Romania --- England
10. India VS Mapuche   --- India 
Last edited by Circle of Ossus on Mon Jul 29, 2019 9:18 am, edited 1 time in total.
AI game on patch 1.0.11b (and onwards): viewtopic.php?f=4&t=1981
Swiss UA proposal (obsolete?): viewtopic.php?f=8&t=768&p=6612#p6612
 
User avatar
DavidIbragimov
Independentista
Independentista
Posts: 147
Joined: Tue Apr 09, 2019 6:26 pm
Location: South Carolina, USA
Favorite Civilization: Romanians

Re: AI game (or at least a try)

Mon Jul 15, 2019 7:36 am

An almost clean guess streak for me! (Almost)

Interesting results, but alot of bugs since its the AI. So I will put more time into predictions this round....

1.Belgium
-From experience against this civ, the AI really does love spamming their elephants late game, and with the recent buff for them, i think they are very well equipped to fight off the French Cavalry
2. Ottomans
-Same story as with the Chinese. Their basic stats for all units simply overpower the weak armies of Egypt.
3. Netherlands
-Sadly, unless the Tupi get struck by lightning and develope Flash speed, they won't stand a chance.
4. Chile
-At this moment, Serbia is the only civ i didn't play with at all and can't say anything, but their AI has constantly dissappointed me.

(At this moment I realized all my predictions past this point would be identical to Ossus.)
Therefore, for the rest of the civs I will try playing the underdog for the sake of diversity.

5.Argentina
 I believe they really do have a potential, with their cattle and Horse Grenadiers. They will need to be very vigilante with their workers and save their cavalry from the Caribineers (i think how they are called). Though their Cavalry is at a disadvantage, I think if the majority of the army is Infantry and maybe 10 horse grenadiers, they could get some results.
6.Spain
Will need to review previous match, but Spain is one of the get-on early raid civs. Even if they fail to accomplish that, they can still feed off factories and cavalry.
7. Iroquois
If I am correct, the tomahawks have a great bonus against cavalry, and with the AI same patterns of in general spawning Cav., I think the I's are in for a treat.
8. Russia
With experience playing against Austria, they biggest weakness is getting overwhelmed early game. The soldiers have too much impact on the economy, and until late game cant really militarize. If Russia overwhelms them with an intense overflow of Privates, they will probably win not because of dominance, but just purely leeching Austria of Economical strength.
9. Romania
-Romanians are truly going to be the underdogs of this round. Again, in another condition would vote for the Brits, but maybe if they make decent amounts of Skirts might they win.
10, Mapuche
-God how easier would all these matches be if it were real people battling out. But since we are here and can only have predictions based off AI algorithm, I can only go off these facts....
-Mapuche like spawning Infantry, if they choose correctly, they will foil India's Cavalry system.
- They are good at siege and India has wonders it relies on for different supports. If they get the worker Aura booster building I call that game over for Indians...
- "They have the power of God and Hoop Thrower on their side!"

Again, trying to give perspective on all these civs. Some are at a slight disadvantage while others will only win with a miracle. The main thing I am concerned for are the in-built Bonuses (Columbia and Mapuche) and the older AI civs, which are known to be more stable and tactically aggresive than the newcomers. Also if the Latest Update hits (soon), the AI might get improved or the stats altered, so it depends on how long it will take our man Ossus to finish the results.

Again Ossus, random but thanks for putting in and sacrificing your time for these tournaments, really appreciate what you do.

Get them underdogs!
Breadfruit.

Who is online

Users browsing this forum: No registered users and 26 guests