|
|
Post by John Marston on Sept 3, 2021 3:17:19 GMT
More aggressive conquest AI. Primarly so that you donβt have an unwinnable situation in the later european conquests where a bunch of allied units swarm a passive general on a city but get injured just enough that they donβt want to attack. And then you get a huge ally blob in the center of the map that you canβt move through and blocks you out of reaching this city forever. This is a problem if youβre playing some really crappy nations where you have to spend like 20-30 turns accomplishing very little in the beginning. To make AI more agressive is an interresting idea. Your analysis is really good SolidLight , maybe the weakness of some countries like Spain, comes from different (higher) levels, where troops start to rest. It seems that the more agressive russians troops, in contrast to spain, fight untill a lower health level, what makes them more 'agressive'. I never thought about this - has anybody better info's? So now that we don't really have anything to do in the EW4 boards and nobody is interested in filling out those general discussions, why not discuss some theories? Let's start with the AI behaviour in 1798. |
|
|
|
Post by John Marston on Sept 3, 2021 3:18:31 GMT
|
|
|
|
Post by John Marston on Sept 3, 2021 4:29:20 GMT
Here's my thoughts (1798 Conquest): 1. Portugal : Portugal is a very "Moderate" AI. It defends itself well, but never really Counter attack Spain. They (I mean Beresford) has a huge obsession for Sevilla and in case he comes out of Lisbon, you know where he goes  2. Spain : Spain has the worst AI, worst Geography and worst set of generals (Even Sardinia's Amadeo III is better as he has Architecture). And when you capture their capital, they freeze at their places and don't move a muscle. 3. France : It has an above average AI. They do their part of spam, but usually get overwhelmed by arguably weaker GB. The other powers come in to make the situation worse. Also, they usually engage in fruitless assault like that of Jourdan's invading GB or Dumoriez at Brunswick and Vienna along with Desaix. Even Moreau, if survives long enough, goes and attacks Danzig lol. 4. HRE : Arguably the most aggressive AI. Due to their central location, they appear like everywhere, including Naples, Athens, Poland, Denmark... 5. Russia : HRE's pal. Basically, their job is to assist HRE in whatever they do. Like assisting Hotze in Denmark, assisting HRE at Istanbul, in Poland, in attacking France... 6. Prussia : Least aggressive among these 3 powers. But that doesn't mean they are any less aggressive. 7. Denmark : Very Aggresive, despite it's small size. They just use brute force to wear down and annihilate the Swedes. 8. Sweden : Has a very aggressive start, but rarely profits the momentum after Dobeln dies. 9. Ottoman Empire : It's probably the most unpredictable AI. In one conquest, you see them attacking Buda, Pest, Vienna and Clujnapoca and in the next conquest, you will see them being steamrolled by Sacken and the Russians, later joined by HRE. 10. Poland : The most important country in the conquest. The survival and aggressiveness of the French faction depends on how long Poland can resist. They don't usually resist for more than 25 rounds though and then Ottoman Empire and the French face the brunt of the coalition attack. |
|
|
|
Post by John Marston on Sept 6, 2021 13:21:55 GMT
|
|
|
|
Post by ππ³π°π΅π΄π¬πΊ on Sept 6, 2021 15:32:03 GMT
1798-conquest: Fra 113% Gbr 108% Koi 110% Rus 105% Tur 115% Hre 112% Pol 120% Pru 110% Spa 113% Por 120% Den 115% Swe 110% Ned 110% Nap 110% Pie 120% Che 100% Mar 100% Tun 100% Alg 100% I believe a lot of the generals have a certain round number (or another trigger), when they start to move. Can you explain the list? Probable this reflects agressivity? |
|
|
|
Post by John Marston on Sept 7, 2021 3:11:08 GMT
1798-conquest: Fra 113% Gbr 108% Koi 110% Rus 105% Tur 115% Hre 112% Pol 120% Pru 110% Spa 113% Por 120% Den 115% Swe 110% Ned 110% Nap 110% Pie 120% Che 100% Mar 100% Tun 100% Alg 100% I believe a lot of the generals have a certain round number (or another trigger), when they start to move. Can you explain the list? Probable this reflects agressivity? Yeah, but it looks inaccurate. Portugal has 120 but it isn't more agressive than say Russia. Maybe the lesser the value, the higher the agressivity? Because it looks more accurate that way. It holds true for neutrals too. If you provoke Switzerland, Bachmann will travel long distances and attack. If you provoke Algeria, Agha will hop into Spain. |
|
|
|
Post by ππ³π°π΅π΄π¬πΊ on Sept 7, 2021 11:27:23 GMT
[ ... ] I believe a lot of the generals have a certain round number (or another trigger), when they start to move. Can you explain the list? Probable this reflects agressivity? Yeah, but it looks inaccurate. Portugal has 120 but it isn't more agressive than say Russia. Maybe the lesser the value, the higher the agressivity? Because it looks more accurate that way. It holds true for neutrals too. If you provoke Switzerland, Bachmann will travel long distances and attack. If you provoke Algeria, Agha will hop into Spain. This is from the 1789 conquest and the only number that can have influence on the ai behaviour in 1789. Apart the starting conditions and trigger !!!!! And ET took a lot of care to build the battlefields. Not one fence is random on the map. The percentage will be probable up to which target-strength AI attacs. 1775 are the values higher. |
|
|
|
Post by John Marston on Sept 7, 2021 12:32:26 GMT
[ ... ] I believe a lot of the generals have a certain round number (or another trigger), when they start to move. Can you explain the list? Probable this reflects agressivity? Yeah, but it looks inaccurate. Portugal has 120 but it isn't more agressive than say Russia. Maybe the lesser the value, the higher the agressivity? Because it looks more accurate that way. It holds true for neutrals too. If you provoke Switzerland, Bachmann will travel long distances and attack. If you provoke Algeria, Agha will hop into Spain. This is from the 1789 conquest and the only number that can have influence on the ai behaviour in 1789. Apart the starting conditions and trigger !!!!! And ET took a lot of care to build the battlefields. Not one fence is random on the map. The percentage will be probable up to which target-strength AI attacs. 1775 are the values higher. What about 1806 and 1809? The Ottomans and Italians spam a lot there. In 1775, British is the main spammer, along with Indian. What about their values? |
|
|
|
Post by ππ³π°π΅π΄π¬πΊ on Sept 7, 2021 20:22:26 GMT
This is from the 1789 conquest and the only number that can have influence on the ai behaviour in 1789. Apart the starting conditions and trigger !!!!! And ET took a lot of care to build the battlefields. Not one fence is random on the map. The percentage will be probable up to which target-strength AI attacs. 1775 are the values higher. What about 1806 and 1809? The Ottomans and Italians spam a lot there. In 1775, British is the main spammer, along with Indian. What about their values? I don't know who it said: When you encounter AI-spam, this is the sign, that you are too slow. AI don't has to colect ressources. So, if AI has enough ressources, AI will randomly build additional units - untill the food limit, and later forts. This has nothing to do with an AI-setting, this is the simple game mechanic and depends on the economy of each country. If you don't reduce enemy-troops by the time, the map can become too crowded to move. If you have later, on long conquests, massive problems with too many troops on the battlefield to move, starving out one of the big spammers, even an allied, can help. Or an AC |
|
|
|
Post by John Marston on Sept 8, 2021 3:00:26 GMT
|
|
|
|
Post by ππ³π°π΅π΄π¬πΊ on Sept 8, 2021 9:40:18 GMT
I did not try it untill now. To be real, I even have no Idea for what I have too look for, or how to test it. I will tag you. John Marston if I have news. |
|
|
|
Post by Nobunaga Oda on Sept 8, 2021 10:36:58 GMT
In such threads, if there is no existing explanation based on the game's programming, normally we'd discuss the surface details.  If the whydunnit is unavailable, we'd focus on the howdunnit. I'd blabber on about how much more aggressive and effective the Coalition is in 1806 (especially if you're a French ally / surbordinate), the initially good British advance north of the Hessen territory in 1798 which usually falters unaided (especially if you're a British ally), and do on. However, such general and arguably somewhat specific behaviours should be well documented. I think John Marston might have discussed 1798 and perhaps 1806 by now.
Conversely, @trostsky, if I'm not wrong, some states with less aggressiveness compared to their neighbours or other states in general seemingly hoard their resources while they "passively" exist. I think Switzerland in most, if not all eras display such behaviour. Egypt in 1815 might have done so as well, until it unleashes a drawn-out offensive which is consistently countered. Perhaps states with lower values hoard resources up to a certain point. At the same time, perhaps their aggressiveness increase? Then, at the pre-determined point, their aggressiveness is high enough to launch a scripted attack. |
|
|
|
Post by littlecorporal on Sept 8, 2021 22:47:48 GMT
I think the best way to test aggression is to play as the last country (so Sardinia in 1798)οΌdo nothingοΌand see how the AI plays. You will notice some countries (France) have multiple strategies; Marmot will go different places. Others like HRE have only one.
|
|
|
|
Post by John Marston on Sept 9, 2021 13:29:20 GMT
Nobunaga Oda, what about the Italian and Ottoman spam at Vienna in 1806? I feel that the coalition spams more because they are more stronger than 1798 and more importantly, 1806 has more cities, which means more spam. That gives a contradictory statement of sorts here because though 1775 economy is probably the worst of all, but still the British and more surprisingly, the Indians spam a lot.
|
|
|
|
Post by ππ³π°π΅π΄π¬πΊ on Sept 9, 2021 16:26:28 GMT
Nobunaga Oda, what about the Italian and Ottoman spam at Vienna in 1806? I feel that the coalition spams more because they are more stronger than 1798 and more importantly, 1806 has more cities, which means more spam. If you talk about R30++ - see my fist comment - this spam is a secure sign that you progress much too slow. That gives a contradictory statement of sorts here because though 1775 economy is probably the worst of all, but still the British and more surprisingly, the Indians spam a lot. 1775 worst economy ? - I really don't think so! We play 1775 to get Lan, because this conquest has so much ressources to offer. 1775 GB starts with an economy similar to HRE (far the strongest one 1798), the US economy is almost 50% stronger. On top all countries start 1775 with much more ressources. [source] |
|
If the whydunnit is unavailable, we'd focus on the howdunnit. 
