Spying mechanics?

General discussion. Please use this area for any off-topic discussion.
creativename
Posts:9
Joined:Sat Apr 14, 2007 2:12 am
Spying mechanics?

Postby creativename » Sat Apr 14, 2007 2:15 am

Has anyone ever found out how the spying system works in this game? e.g., if you have X spies against Y enemy spies and both are equal at spying, what are their chances of successful espionage/sabotage on a per turn basis, and vice versa?

Basically I want to know how many defensive spies are "enough". And also if one wants to maximize the odds of stealing tech, is it better to have 10 spies split across 5 races, or to place them all against 1 race (assuming all races have equal spy bonus and tech level).

User avatar
Cabman
Posts:235
Joined:Fri Jul 29, 2005 5:58 pm
Location:Zabrze, Poland

Postby Cabman » Sat Apr 14, 2007 6:05 am

This case remainst still the biggest Moo mistery. It is very random. Sometimes you can succed with just one spy against 10, but sometimes 10 spies do no harm to just one defending Democratic technologies. It is obvious that the bigger spy bonus and more spies the bigger probability of succes, but the question how does it pariculary work?...is unaswered.

User avatar
Matthew
Posts:186
Joined:Mon Mar 13, 2006 10:06 am

Postby Matthew » Sat Apr 14, 2007 11:27 am

I have seen some of the assembly code for spying. My ability to decipher assembly still needs work but here is what I make of it.

It first checks leaders for the assassin ability. They get their roll first. Any spies killed by leaders are automatically removed. Next, they calculate the odds of a spying success based on leaders, race picks and technology. This is checked against the defending agents ability. Then each offensive spy gets to roll based on this chance against the defending agent(s) chance.

The higher the spying percentage you have and the more spies you have, the better the odds of success. It's still a random chance to succeed or fail. Most of the time, spies fail because defensive agents have better odds. By the time you meet your rivals you should try to have two agents. Towards the middle game, you should have 5-9 and in the late game 10-20 is fine. If you only have one rival then you can decide to go on the offensive. If you have many rivals then it’s safer to defend.

marhawkman
Posts:67
Joined:Sat Mar 10, 2007 10:30 pm

Postby marhawkman » Mon Apr 16, 2007 10:47 pm

Is the chance to steal technology affected by the amount of technology to steal?

User avatar
Matthew
Posts:186
Joined:Mon Mar 13, 2006 10:06 am

Postby Matthew » Tue Apr 17, 2007 1:25 am

Is the chance to steal technology affected by the amount of technology to steal?
The only time that I can see if is if the enemy has nothing for you to steal.

marhawkman
Posts:67
Joined:Sat Mar 10, 2007 10:30 pm

Postby marhawkman » Tue Apr 17, 2007 11:54 am

Ah. this explains how I'll go 20 turns with no spying, and then as soon as I research something new.... Bam! almost immediately somebody steals it.

creativename
Posts:9
Joined:Sat Apr 14, 2007 2:12 am

Postby creativename » Wed Apr 18, 2007 9:04 pm

Interesting replies, thanks. So it's still a mystery after all this time.

From my own experience it's better to split spies across multiple races. This probably does give you better odds because that way you're more likely to have something to steal, even if it otherwise doesn't help with spying odds.

marhawkman
Posts:67
Joined:Sat Mar 10, 2007 10:30 pm

Postby marhawkman » Wed Apr 18, 2007 11:31 pm

How does it calculate the odds of spies dying?

User avatar
Matthew
Posts:186
Joined:Mon Mar 13, 2006 10:06 am

Postby Matthew » Thu Apr 19, 2007 8:40 am

How does it calculate the odds of spies dying?
Here's what appears to happen in the code from what I understand.

Check leaders for assassin ability. It's random chance based on the skill level of the leader. It looks like that can only kill one per turn.

For each player, it determines what their spying ability is and agent defense ability. Each offensive spy rolls to see if they succeed or not then each defensive agent rolls to see if they succeed. If a spy succeeds, it does a second roll to determine the outcome. This can be stealing a tech, sabotage or killing an enemy agent. If the defensive agent succeeds it rolls to see if it countered an offensive spying attempt and/or kills an enemy spy. It looks like you can only kill one spy or one defending agent per turn. Since it's totally random, it's impossible to predict the outcome, except when the spying ability is zero or if there are no spies. The best way to improve the odds is with spying tech and more spies.

marhawkman
Posts:67
Joined:Sat Mar 10, 2007 10:30 pm

Postby marhawkman » Thu Apr 19, 2007 7:15 pm

In other words if your spies suck they'll get killed off left and right?

User avatar
Matthew
Posts:186
Joined:Mon Mar 13, 2006 10:06 am

Postby Matthew » Thu Apr 19, 2007 10:46 pm

In other words if your spies suck they'll get killed off left and right?
If your race has little or no spying ability and your opponent has good spying through race picks and tech, your spies will probably die a lot more often than theirs. Sometimes they do defy the odds but it's rare.


Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 16 guests