and to back up what Chengar Qordath just said here is aprof's reply to that exact question that I put to him in the Mexican AAR thread.
Cheers, Ice
Cheers, Ice
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
[EDIT]: (forgot to comment on the above)
Reading aprof's AAR doen't support that claim, for example the United States of Central America seems to suffer from chronic rebellions.
[/EDIT]
Anyway, the player can always just press through unpopular but necessary reforms (industry etc.) since he will never lose power. It doesn't matter to the player if the government in toppled by a coup or loses an election - he still controls the policies.
Thus it will be much more easy fot the player to industrialize than in real life where governments had to take popularity into considerations. Coupled with the 20/20 hindsight of knowing what to really aim for makes the player extremely strong compared to the AI, if the AI attempts to play `historically'.
I like this `side effect' since it enables us to create economic miracles in unexpected countries, if the playing field is otherwise even
I love to turn destabilized, declining countries into first rate powers.
Last edited:
In my experience with paradox, the AI steady improves from patch to patch and from game to game. Keep the good work paradox.
Agreed. I meant it in the sense that in real life one particular reform will have undesirable effects on you country, as a direct result of the reform. For example, In EUII terms, moving a notch to dp for aristocracy will cause the dp for descentralization to move 2 notchs, etc.
Agree also with the'historical' theory. In paradox games the AIs are required to act somewhat historical, and the player not. For example, a harder HOI game as germany will be if suddenly the Soviet AI will turn onto the human player when you he is attacking Poland. The USA AI ignoring all the historicity will build some 100 division inmdetially while you try to contain the Soviet onslaught, and land them in denmark. And the french will mass troops and attack you on the west. Italy won join your loosing cause. And you can go on and on. No many other games are required to follow history,.
Those games neither have to follow the balancing of power. IE. many players complain when AI austria absorb the HRE as ridiculous, but will also find ridiculous having to restrict their own tiny german minor to one province
And those are only some the practical issues facing the AI design, without considering the technical issues, like adaptivity, learnability, sharing of knowledge, etc.
All the EU-engine games I've played so far have an AI that is far superior to any other game I've played, with the exception of chess and Othello (but those games are of a mathematical nature and have a total lack of randomness). This is especially true when you factor in the complexity of the games as compared to others.
The Paradox team must have learnt from mistakes in EU and HoI when developing Victoria. The Victoria AI will not be perfect, but I'm sure it'll be the best one yet.
The Paradox team must have learnt from mistakes in EU and HoI when developing Victoria. The Victoria AI will not be perfect, but I'm sure it'll be the best one yet.
I totally agree...which is why I'm soooo hopeful that Victoria will kick ass.
Cheers, Ice
I concurr .... I wish someday (may be victoria itself) we will have a game with this depth, where most of the development focus will be pun into creating a superb AI. I can only dream. Paradox games, however, are good enough for me
no, thats not what happened in RL. Germany did not win on the western front, and that was because they had so many troops on the eastern front. by the time they could move their troops from the now peaceful eastern front, the americans had already moved in to boost the numbers to the point where the final german offensive was ineffective.
in sytass's aar, he easily handles all three fronts at once, with enough men in every place to continue to make advances because he outnumbers the enemy in every direction. this is not how it happened, more or less (and incase you forgot, Germany did lose the war).
In my view, the only way mexico should or could avoid a war with america is three possible ways
1. mexico does not surive long enough to be a concern to america.
2. mexico gives up any territory america demands
3. america is being totally pacifist (should only happen maybe 1 out of 1,000 times game is played)
this wasnt an extreme circumstance like WW1 was, it was simply one continent with only two major powers. history tells up that these two nations must come into conflict, its just the way things work. the question of "IF mexico has a strong military, THEN we must build ours up too, incase they attack us" will creep into their minds and eventually create anti-mexican sentimate amongst their people, with the same effect happening in mexico against americans. it wont take long for these fealings to cause a war.
No, as I read it Sytass had to put operations in the west on hold while he bailed out A-H and was even forced back at one point. And just look at the trouble he had to go to to erase Serbia.
You are forgetting that in Sytass's AAR, a number of other things also did not happen like they did IRL. Italy and Sweden joining the Central Power's cause for example, and Turkey making strong progress against the British. Had these things happened IRL, the Germans might have won the war.
And AFAIK the final German offensive was not ineffective... it was one of the most brilliant offensives of them all, due to new tactics and a high concentration of artillery... but like almost all the others it failed to achieve it's main objective.
4. Mexico backs the CSA and you have a 3 way power struggle that results in stagnation.
5. Attacking Mexico would be Suicide for the USA because she's more powerful economically and Military.
5. Attacking Mexico would be Suicide for the USA because she's more powerful economically and Military.
I count 3 major powers in NA during the greater part of the 19th century...
I'm pretty sure that if both US and Mexico are AI-led you will get your precious war. But if Mexico is played by a human who takes special measures to keep on US's good side, should he not be able to avoid war?
Come on lads, there will be plenty of time to slaughter the AI after the game comes out
Take AARs with a pinch of salt, back in the pre-badboy days of EU1 there were cautious AARs and tales of lost wars.
Take AARs with a pinch of salt, back in the pre-badboy days of EU1 there were cautious AARs and tales of lost wars.
regarding ais
EU, HoI and Victoria are very complex games. I am not wise enough to define how complex they are, but I would claim that they are way more complex then Chess for instance.
Millions of man hours has been spent on trying to create an ai that can play chess efficiently against a human opponent. It has partly succeded. The success part of the "partly" is that today the best chess computers can give the best human players a tough game and sometimes even beat them. The not so succesful part is that when playing an ai it often "feels" like it is not a human opponent.
If we would spend the same amounts of research on an ai for paradox games the first consequence would be that the cost for us consumers to buy a box would be astronomical in comparison to today. The second consequence would be that we would still be unsatisfied with it.
I think that it is a waste of money even trying to create an ai for such a complex game that should be anyway near a human opponent.
Instead the solution is to be found in these options:
1. Create a gaming environment where the ai can play like an ai and be accepted as such. As long as the ai is used to do stuff like the ai is good at doing it is fine.
2. Play multiplayer.
These two options can be combined. Fine examples are found in many traditional board games. For instance in Advanced Third Reich where non-player states are handled by a real player as soon as it gets involved in the game, and meanwhile it lives in peace follows a given pattern. This is one example of how to work around the problem.
The conlusion of my post is that if you expect the ai in Victoria to be substanstiallly better than previous ais, and in particular if you expect it to play like a human, you will be dissapointed.
If you however accept the limitations that are put an ai-creation by time and money, you might find ways to enjoy the game anyway.
EU, HoI and Victoria are very complex games. I am not wise enough to define how complex they are, but I would claim that they are way more complex then Chess for instance.
Millions of man hours has been spent on trying to create an ai that can play chess efficiently against a human opponent. It has partly succeded. The success part of the "partly" is that today the best chess computers can give the best human players a tough game and sometimes even beat them. The not so succesful part is that when playing an ai it often "feels" like it is not a human opponent.
If we would spend the same amounts of research on an ai for paradox games the first consequence would be that the cost for us consumers to buy a box would be astronomical in comparison to today. The second consequence would be that we would still be unsatisfied with it.
I think that it is a waste of money even trying to create an ai for such a complex game that should be anyway near a human opponent.
Instead the solution is to be found in these options:
1. Create a gaming environment where the ai can play like an ai and be accepted as such. As long as the ai is used to do stuff like the ai is good at doing it is fine.
2. Play multiplayer.
These two options can be combined. Fine examples are found in many traditional board games. For instance in Advanced Third Reich where non-player states are handled by a real player as soon as it gets involved in the game, and meanwhile it lives in peace follows a given pattern. This is one example of how to work around the problem.
The conlusion of my post is that if you expect the ai in Victoria to be substanstiallly better than previous ais, and in particular if you expect it to play like a human, you will be dissapointed.
If you however accept the limitations that are put an ai-creation by time and money, you might find ways to enjoy the game anyway.
I'm becoming more and more tired of people saying "It is impossible, it could not happen", and their only explanation is "It could not happen, because it did not happen".
Some people seemingly don't have any imagination.
In the AARs I saw one or two things, that made me wonder whether such an outcome could be possible in RL. But I don't think there is something wrong with Germany victorious in WWI, Mexican colonization, or tech trading.
1. German victory:
Sytass achieved more in then Germans did in RL, but it was not a walk in the park. There were two important factors, that did not occured in reality: Sweden & Italy joining Central Powers.
Especially the latter is important: Germans weren't that far from victory on the western front in RL, and if they could count on Italians help, the outcome would have be obvious.
2. Mexican colonization:
Take a rest. What's wrong with fairly rich country, wisely guided, taking one small colony? The UK would never agree, blah blah. They did accept Belgium colonies, didn't they?
3. Tech trading:
At first, I thougt that maybe there is something wrong with the fast pace whith which one can gain techs by trading. But it is actually the thing that happened in RL. The inventions weren't made independently in diffrent countries, but purchased or just copied from the inventor. And it did happen rather fast. Think about, say, positivism - how long did it take to spread that idea.
Some people seemingly don't have any imagination.
In the AARs I saw one or two things, that made me wonder whether such an outcome could be possible in RL. But I don't think there is something wrong with Germany victorious in WWI, Mexican colonization, or tech trading.
1. German victory:
Sytass achieved more in then Germans did in RL, but it was not a walk in the park. There were two important factors, that did not occured in reality: Sweden & Italy joining Central Powers.
Especially the latter is important: Germans weren't that far from victory on the western front in RL, and if they could count on Italians help, the outcome would have be obvious.
2. Mexican colonization:
Take a rest. What's wrong with fairly rich country, wisely guided, taking one small colony? The UK would never agree, blah blah. They did accept Belgium colonies, didn't they?
3. Tech trading:
At first, I thougt that maybe there is something wrong with the fast pace whith which one can gain techs by trading. But it is actually the thing that happened in RL. The inventions weren't made independently in diffrent countries, but purchased or just copied from the inventor. And it did happen rather fast. Think about, say, positivism - how long did it take to spread that idea.
Agreed with creature.
What are you talking about?
Aproof plays Mexico and he chooses to colonise som provinces in Africa, so what if it didnt happen, he choosed to do that because he plays the game. If the game just runs like in reality, how fun would that be?
Sytass winning WW1 and? whats the problem? you shouldent be able to win with a country that historically lost? thats ridiculous.
You need to take a break guys, its a game.
If Sytass plays smart of course he should be able to win.
I dont get you guys, how do you really think a game should simulate alternate history?What do you want from this game?
Please explain that, im really curious.
What are you talking about?
Aproof plays Mexico and he chooses to colonise som provinces in Africa, so what if it didnt happen, he choosed to do that because he plays the game. If the game just runs like in reality, how fun would that be?
Sytass winning WW1 and? whats the problem? you shouldent be able to win with a country that historically lost? thats ridiculous.
You need to take a break guys, its a game.
If Sytass plays smart of course he should be able to win.
I dont get you guys, how do you really think a game should simulate alternate history?What do you want from this game?
Please explain that, im really curious.
Well my 2c regarding this problem;
If an event such as WW1's outcome is reversed by the player, then either the event should have been close to a 50:50 IRL, or the player should have had to work damned hard to achieve the result.
The game shouldn't be a replay of history, but it shouldn't be Risk with more options either.
well I hope they got rid of the ai going to sleep. Not sure if I would really call it sleep but it would seem like the longer a gaming session went the ai units would sometimes be stuck and upon restarting the game they would start moving again.