I believe combat is rigged in this game and it ruins it for me

[Lansing]

I am an enormous Civ fan, and I am finally going to give up on Civ 4 until they fix the combat engine. I did a little experiment today with random seeding. I put a series of units against one another, bot attacking and defending, to see if the computer had a distinct advantage in the random rolls. Turns out it does. I did these experiments on Monarch difficulty, and put myself into spots where the computer was going to attack me and where I could attack the computer. I used primarily naval units so that no terrain bonus could factor in, and used identical units (xp wise) in all cases.

Destroyer v. Destroyer

Attacking: Computer wins 40/50 times, 80%.
Defending: Computer wins 38/50 times, 76%.

Battleship v. Battleship

Attacking: Computer wins 38/50 times, 76%.
Defending: Computer wins 41/50 times, 82%.

Frigate v. Frigate
Attacking: Computer wins 33/50 times, 66%.
Defending: Computer wins 43/50 times, 86%.

Swordsman v. Swordsman (desert)
Attacking: Computer wins 40/50 times, 80%
Defending: Computer wins 34/50 times 68%.


I don't know if these numbers are accurate, and frankly I don't care. After game after game of frustration at not actually being able to fight anyone, *ever*, I finally got fed up and decided to see how badly the AI was fudging me over. I loved the previous Civs because of the reasonable balance between combat and expansion, where waging war was at least a realistic possibility (without save/reload cheating). It's simply not possible in this game, and I really think it sucks the vitality out of it completely.

Anyways, make of this what you will, and do your own tests if you want. And if I am missing some obvious explanation or fact that you all know, please just point it out and don't flame me too badly. I'm pretty discouraged about my favourite game right now.

[Tebriel]

You know coastline gives a 10% bonus, right?

[Kuciwalker]

nvm

[Lansing]

Originally posted by Tebriel
You know coastline gives a 10% bonus, right?

Yes, all sea battles were open water.

[Hauptman]

Try it on noble... monarch actually cheats in the AI's favor to make it tougher on you.

As it is already i can kill stacks of 15+ AI units with 4 or 5 of my own... just because I'm smarter. So yes Im pretty sure on monarch + the ai gets bonus's in combat as well as production and research.

Noble however is sposed to be the "even" playing field.

[Dis]

I hate it, but it doesn't ruin the game for me (though it comes close sometimes).

Today my cavalry lost against a samuraii. . Yeah he was in forest, but the combat odds showed almost a 2 to 1. And my cav was fully healed as well.

[Urban Ranger]

Originally posted by Lansing
Yes, all sea battles were open water.

If you keep reloading to get better results, you will be disappointed. The random number sequence for a game is fixed, even though there is an option to change that.

[Urban Ranger]

Originally posted by Dis
Today my cavalry lost against a samuraii. . Yeah he was in forest, but the combat odds showed almost a 2 to 1. And my cav was fully healed as well.

If you read one this thread it appears that Vel had great success with his battles even when the odds were against him.

Combat does not appear to be rigged in this game. It just appear this way. People tend to remember odd or unusual happenings and forget ones that are, um, normal.

[Axxaer]

I think the only part of the combat odds which is "rigged" (innacurate) is units with first strike chances. Longbowman lose much less frequently than other units even when the odds are stacked against them.

Apart from that, if you are finding you can't win battles, then obviously you are playing too high a difficulty.

No it's not impossible, there are plenty of people on this forum capable of regularly winning on monarch. Just because it's too hard for YOU doesn't make it's too hard for everyone else.

It certainly is not worth giving up on a game just because you can't beat on one particular difficulty level.

Put it down each game until you are happy that the (fair) odds aren't "cheating" to make you lose.

[alva]

Combat does not appear to be rigged in this game. It just appear this way. People tend to remember odd or unusual happenings and forget ones that are, um, normal.

And the ones they shouldn't have won.

Just as we do with busses.

[Lansing]

Originally posted by Urban Ranger


If you keep reloading to get better results, you will be disappointed. The random number sequence for a game is fixed, even though there is an option to change that.

I don't think you understand what I did. I turned random seed on and tested the battles in totally 50-50 circumstances. It turned out the AI had a MASSIVE advantage. Very dissapointing.

And I did this precisely because I didn't want to just remember the "bad" results. I wanted to be sure my intuition on the AI rigging was correct. It was.

[DougM]

nothing in the handicap info

There's nothing in the handicap info (or any where else in XML) that would in any obvious way impact the combat outcome percentages in the game.

A fairly straightforward experiment, if someone was willing to try:

On at least 3 different difficulties, use the worldbuilder to place a lot of pairs of units for you and the AI. Say at least 10 each of 3 different seagoing ships. Report the percentage victory when you attack per ship and per difficulty.

That would at least give a good idea of whether there is a bias for or against attack. The next step would be to see if you can make the AI attack with ships also granted via the worldbuilder, and compare its attack success rate.

[Fried-Psitalon]

At least 10? I'd say at least 100.

[Solver]

Wow... some people don't know what odds mean.

If your unit has a 75% chance of winning, it doesn't mean that your units would win 75 out of 100 such battles. Winning 60 of 100 would be significantly more probable!

[Shaka II]

Originally posted by Solver
Wow... some people don't know what odds mean.

If your unit has a 75% chance of winning, it doesn't mean that your units would win 75 out of 100 such battles.

It doesn't? Are you sure?