All information has been moved here:
http://forums.civfanatics.com/forumdisplay.php?s=&forumid=63
http://forums.civfanatics.com/forumdisplay.php?s=&forumid=63
Civ 3 Tournament Proposal
- Thread starter Grey Fox
- Start date
Wow GreyFox! This has been up for almost 10 hrs and no responses yet? 
Could it be that the half-dozen of us discussing it on chat last night are the only ones interested?
(This will at least bump it up on the page again.)
C'mon people, we'd like some feedback here!
PS: Again GreyFox:
Could it be that the half-dozen of us discussing it on chat last night are the only ones interested?
(This will at least bump it up on the page again.)
C'mon people, we'd like some feedback here!
PS: Again GreyFox:
First let me qualify this by saying I am in the process of completing my first GOTM. I may have other thoughts in a month or two. I like the idea of the different categories for the levels. I am not really enjoying the current GOTM because it is too easy. I would prefer to play at monarch or higher. I also think it is important to get others to play the game though and the lower levels are needed to get them going. Your plan allows for all levels of players to play.
The only problem I have with your suggestion is the map size. I don't particulay care for the tiny world games. The games are too short and don't really allow for building up your empire.
Other than that one comment I think it is a great idea!!
The only problem I have with your suggestion is the map size. I don't particulay care for the tiny world games. The games are too short and don't really allow for building up your empire.
Other than that one comment I think it is a great idea!!
Falcon02
General
While talking it over last night we came to the general consencious that smaller maps would be better specifically to make sure the tournament wouldn't take too long. And so each season of 4 games wouldn't last for 4 months.
However, you do have a point, we may want to see if we can work out a practical plan that would allow for small to large, instead of tiny to standard. What do you people think of that?
However, you do have a point, we may want to see if we can work out a practical plan that would allow for small to large, instead of tiny to standard. What do you people think of that?
- Joined
- Nov 16, 2001
- Messages
- 2,686
Proposed Scoring Formula:
OverallPercentage = ((MapSize * TurnsPercentage) + ScorePercentage) / (MapSize + 1)
TurnsPercentage = FewestTurns / PlayerTurns
ScorePercentage = PlayerScore / HighestScore
FewestTurns = Fewest number of turns needed by any player in the division to achieve the specified victory condition.
PlayerTurns = Number of turns taken by the player to achieve the specified victory condition.
PlayerScore = Score by the player, regardless of how the game was ended.
HighestScore = Highest score submitted by any player in the division.
MapSize
at 80% WaterLevel
at 70% WaterLevel
at 60% WaterLevel
Tiny
1.00
1.50
2.00
Small
1.33
2.00
2.66
Standard
2.77
4.16
5.55
Explaination
The idea is to reward both early finishes and high scores without requiring the game to be milked to do well. The two main parts of the equation are percentages of turns played and score. These are represented by the TurnsPercentage and ScorePercentage equations. The fastest finish becomes the standard, as does the highest score. The standards are used to deterimine a percentage representing the efficiency that each player achieved in both categories. These 2 percentages are weighted, added together, and then divided by 1 + the weight given to determine the OverallPercentage. All percentages will be 3 digits for more accuracy.
MapSize and WaterLevel are taken into account when determining how the percentages are weighted. The amount of land tiles directly affects the maximum score attainable by milking. To keep milking in check, the more land tiles on the map the higher the modifier.
Tiles per map size
width * height / 2
Tiny: 60 * 60 / 2 = 1800 tiles
Small: 80 * 80 / 2 = 2400 tiles
Standard: 100 * 100 / 2 = 5000 tiles
The tile numbers are reduced to 18, 24, and 50 respectively, then each multiplied by the land percentages for each given water level. 80% water means 20% land, ect. So the multipliers are 0.2, 0.3, and 0.4.
18 * 0.2 = 3.6
18 * 0.3 = 5.4
18 * 0.4 = 7.2
24 * 0.2 = 4.8
24 * 0.3 = 7.2
24 * 0.4 = 9.6
50 * 0.2 = 10
50 * 0.3 = 15
50 * 0.4 = 20
All of these numbers are then divided by the lowest result, 3.6, to give us the MapSize multipliers. These multipliers weight the TurnsPercentage in relation to the ScorePercentage, making milking less viable.
Missing the target victory condition
When a player doesn't achieve the given victory condition, their TurnsPercentage will be 0%. In this case, the player's OverallPercentage will be entirely dependant on their score.
If a player is to lose the game, then it is dealt with the same way. By losing, the player won't have achieved the target victory condition, and so will only recieve a percentage based on their score.
OverallPercentage = ((MapSize * TurnsPercentage) + ScorePercentage) / (MapSize + 1)
TurnsPercentage = FewestTurns / PlayerTurns
ScorePercentage = PlayerScore / HighestScore
FewestTurns = Fewest number of turns needed by any player in the division to achieve the specified victory condition.
PlayerTurns = Number of turns taken by the player to achieve the specified victory condition.
PlayerScore = Score by the player, regardless of how the game was ended.
HighestScore = Highest score submitted by any player in the division.
MapSize
at 80% WaterLevel
at 70% WaterLevel
at 60% WaterLevel
Tiny
1.00
1.50
2.00
Small
1.33
2.00
2.66
Standard
2.77
4.16
5.55
Explaination
The idea is to reward both early finishes and high scores without requiring the game to be milked to do well. The two main parts of the equation are percentages of turns played and score. These are represented by the TurnsPercentage and ScorePercentage equations. The fastest finish becomes the standard, as does the highest score. The standards are used to deterimine a percentage representing the efficiency that each player achieved in both categories. These 2 percentages are weighted, added together, and then divided by 1 + the weight given to determine the OverallPercentage. All percentages will be 3 digits for more accuracy.
MapSize and WaterLevel are taken into account when determining how the percentages are weighted. The amount of land tiles directly affects the maximum score attainable by milking. To keep milking in check, the more land tiles on the map the higher the modifier.
Tiles per map size
width * height / 2
Tiny: 60 * 60 / 2 = 1800 tiles
Small: 80 * 80 / 2 = 2400 tiles
Standard: 100 * 100 / 2 = 5000 tiles
The tile numbers are reduced to 18, 24, and 50 respectively, then each multiplied by the land percentages for each given water level. 80% water means 20% land, ect. So the multipliers are 0.2, 0.3, and 0.4.
18 * 0.2 = 3.6
18 * 0.3 = 5.4
18 * 0.4 = 7.2
24 * 0.2 = 4.8
24 * 0.3 = 7.2
24 * 0.4 = 9.6
50 * 0.2 = 10
50 * 0.3 = 15
50 * 0.4 = 20
All of these numbers are then divided by the lowest result, 3.6, to give us the MapSize multipliers. These multipliers weight the TurnsPercentage in relation to the ScorePercentage, making milking less viable.
Missing the target victory condition
When a player doesn't achieve the given victory condition, their TurnsPercentage will be 0%. In this case, the player's OverallPercentage will be entirely dependant on their score.
If a player is to lose the game, then it is dealt with the same way. By losing, the player won't have achieved the target victory condition, and so will only recieve a percentage based on their score.
- Joined
- Nov 16, 2001
- Messages
- 2,686
There are a couple wrinkles in the scoring system that still need to be ironed out. As I see them:
Should a Tiny/80% Water map be considered the "normal" map type, getting the modifier of 1? This is an area that I'm not too sure of, having more experience on larger maps. The "normal" map type should be the one where milking and early finishes are evenly balanced.
Now I seem to have forgotten the other problem that jumped out at me... I know it's there!
Should a Tiny/80% Water map be considered the "normal" map type, getting the modifier of 1? This is an area that I'm not too sure of, having more experience on larger maps. The "normal" map type should be the one where milking and early finishes are evenly balanced.
Now I seem to have forgotten the other problem that jumped out at me... I know it's there!
Falcon02
General
I think I like your scoring method, Aeson. While it is a bit confusing, from what I understand it should be pretty fair.
Cartouche Bee
Appropriations Consultant
Salute Aeson,
OK, first of all I would have to look at this at sometime other than Friday Night!!!
But, since we know that domination victory includes costal and fresh water squares in the calculation as land , we should consider this in the way we determine how many land and water squares are on a map. Costal squares should be included in the land total for calculation purposes as proposed.
As a side issue, I also think that the number of entries should be considered. The number of entries implies the degree of difficulty. Fewer entries implies a situation that restricted submissions, therefore a higher general difficulty. This would be a factor applied between different competions, from the past and the current.
Cartouche Bee
OK, first of all I would have to look at this at sometime other than Friday Night!!!
But, since we know that domination victory includes costal and fresh water squares in the calculation as land , we should consider this in the way we determine how many land and water squares are on a map. Costal squares should be included in the land total for calculation purposes as proposed.
As a side issue, I also think that the number of entries should be considered. The number of entries implies the degree of difficulty. Fewer entries implies a situation that restricted submissions, therefore a higher general difficulty. This would be a factor applied between different competions, from the past and the current.
Cartouche Bee
Black Fluffy Lion
Qxria
- Joined
- Oct 13, 2001
- Messages
- 977
I love it! easy levels, small maps, sounds great. Only thing is 12 days?! seems a bit short to me even on a tiny map
If you all like it, set it up. 
But it's not going to replace the current GOTM. 
That's why I'm moving it to General Discussions, since succession games are there as well. And you'll get more attention there as well I think.
But it's not going to replace the current GOTM. That's why I'm moving it to General Discussions, since succession games are there as well. And you'll get more attention there as well I think.dikwhit
Warlord
standard, small,tiny. lame,lamer,and lamest IMO. Tournaments on the three smallest map sizes?????? Where is the intrest for any1 that plays the bigger maps. When i build my empire, I want the empire feel and look.. not the feeling constantly of thats it?????? Not to mention that qwick conquest victories will still outscore anything else regardless of modifiers.As an example...... lets say monarch diff. tiny map 60 water. a fast conquest 5-10000 pts is not too difficult with some luck.But an extended say histograph victory or cultural with say 3000 pts is a lot of work. with a 2.0 modifier its only 6k,,,,, got to do something about balancing the high end scoring on quick conquest to make it viable for those of us that arent playing conquerization. Im not trying to start a flame war just trying to point out that there are those of us that dont play kill em all asap on itty bitty maps
- Joined
- Nov 16, 2001
- Messages
- 2,686
Well each map will have a specific victory condition that the players need to achieve to "score" well. Any game which didn't achieve that victory condition could at most score .500, and only then on a Tiny/80% where their score was the highest in the division.
Example:
Small/70% Water map with a Space Launch victory condition. All 3 players in the Veteran Division, Monarch difficulty. Just using these settings because GOTM05 was a similar map and gives us some easy references.
Player 1 (highest score) conquers the world at 500BC. I am being very generous here, Lucky's 840AD conquest was the earliest in GOTM05. (2050 + 500) * 4 = 10200 score bonus for an early conquest. A base score of around 500. Total Score is 10700. Victory condition wasn't met.
Player 2 (fastest launch) launches their spaceship in 1810AD. In GOTM05, Nathan Barclay got a score of 3027 for a 1810AD launch. Victory condition was met in 370 turns.
Player 3 conquers early and milks their score until a 2050AD launch. In GOTM05 SirPleb had the highest score with 7412 from a 2050 milked game. Victory condition was met in 540 turns.
Player 1's OverallAverage = ((2 * 0) + 10700/10700) / (2 + 1) = .333
Player 2's OverallAverage = ((2 * 370/370) + 3027/10700) / (2 + 1) = .761
Player 3's OverallAverage = ((2 * 370/540) + 7412/10700) / (2 + 1) = .688
As you can see, meeting the specified victory condition will be critical to getting a good percentage, especially on the larger landmass games. Milking would usually be closer with the average, but the "high" early conquest score hurt Player 3 more than Player 2. The players who meet the victory condition earliest and score well will be the ones who have the highest percentage. With a MapSize multiplier of 2, the highest any player can "score" without meeting the specified victory condition is .333
Example:
Small/70% Water map with a Space Launch victory condition. All 3 players in the Veteran Division, Monarch difficulty. Just using these settings because GOTM05 was a similar map and gives us some easy references.
Player 1 (highest score) conquers the world at 500BC. I am being very generous here, Lucky's 840AD conquest was the earliest in GOTM05. (2050 + 500) * 4 = 10200 score bonus for an early conquest. A base score of around 500. Total Score is 10700. Victory condition wasn't met.
Player 2 (fastest launch) launches their spaceship in 1810AD. In GOTM05, Nathan Barclay got a score of 3027 for a 1810AD launch. Victory condition was met in 370 turns.
Player 3 conquers early and milks their score until a 2050AD launch. In GOTM05 SirPleb had the highest score with 7412 from a 2050 milked game. Victory condition was met in 540 turns.
Player 1's OverallAverage = ((2 * 0) + 10700/10700) / (2 + 1) = .333
Player 2's OverallAverage = ((2 * 370/370) + 3027/10700) / (2 + 1) = .761
Player 3's OverallAverage = ((2 * 370/540) + 7412/10700) / (2 + 1) = .688
As you can see, meeting the specified victory condition will be critical to getting a good percentage, especially on the larger landmass games. Milking would usually be closer with the average, but the "high" early conquest score hurt Player 3 more than Player 2. The players who meet the victory condition earliest and score well will be the ones who have the highest percentage. With a MapSize multiplier of 2, the highest any player can "score" without meeting the specified victory condition is .333
- Joined
- Nov 16, 2001
- Messages
- 2,686
Example: (using 840AD as the early conquest victory)
Small/70% Water map with a Space Launch victory condition. All 3 players in the Veteran Division, Monarch difficulty. Just using these settings because GOTM05 was a similar map and gives us some easy references.
Player 1 conquers the world at 840AD. Lucky's 840AD conquest was the earliest in GOTM05. Total Score is 6017. Victory condition wasn't met.
Player 2 (fastest launch) launches their spaceship in 1810AD. In GOTM05, Nathan Barclay got a score of 3027 for a 1810AD launch. Victory condition was met in 370 turns.
Player 3 (highest score) conquers early and milks their score until a 2050AD launch. In GOTM05 SirPleb had the highest score with 7412 from a 2050 milked game. Victory condition was met in 540 turns.
Player 1's OverallAverage = ((2 * 0) + 6017/7412) / (2 + 1) = .271
Player 2's OverallAverage = ((2 * 370/370) + 3027/7412) / (2 + 1) = .803
Player 3's OverallAverage = ((2 * 370/540) + 7412/7412) / (2 + 1) = .790
Without the unaturally high conquest score, milking and early victory are much closer.
Small/70% Water map with a Space Launch victory condition. All 3 players in the Veteran Division, Monarch difficulty. Just using these settings because GOTM05 was a similar map and gives us some easy references.
Player 1 conquers the world at 840AD. Lucky's 840AD conquest was the earliest in GOTM05. Total Score is 6017. Victory condition wasn't met.
Player 2 (fastest launch) launches their spaceship in 1810AD. In GOTM05, Nathan Barclay got a score of 3027 for a 1810AD launch. Victory condition was met in 370 turns.
Player 3 (highest score) conquers early and milks their score until a 2050AD launch. In GOTM05 SirPleb had the highest score with 7412 from a 2050 milked game. Victory condition was met in 540 turns.
Player 1's OverallAverage = ((2 * 0) + 6017/7412) / (2 + 1) = .271
Player 2's OverallAverage = ((2 * 370/370) + 3027/7412) / (2 + 1) = .803
Player 3's OverallAverage = ((2 * 370/540) + 7412/7412) / (2 + 1) = .790
Without the unaturally high conquest score, milking and early victory are much closer.
- Joined
- Nov 16, 2001
- Messages
- 2,686
I agree that a perfect formula would include coastal tiles and "free" sea tiles that can be claimed. Also it would include modifiers for Pangaea, Continents, Archipelago; Warm, Temperate, Cool; Arid, Normal, Wet; 3, 4, and 5 billion year settings. Various other modifiers depending on interactions between map settings, difficulty level, and specified victory conditions would be needed as well.
Without actually analyzing each individual map though, it would be impossible to formulate an accurate algorithm that includes all these factors. In the chat room discussion about this formula, it was decided that it wasn't worth all the extra math just to include something that had so much variation from map to map. Hopefully in practice, this more simplified version will hold up. When it is applied to GOTM05 it seems to achieve it's purpose pretty well.
SirPleb
Shaken, not stirred.
Nice work on the scoring formula Aeson! Your examples suggest to me that the initial MapSize values might work fairly well. It sure seems good enough to start with, can be adjusted later based on experience.
sounds good. I like the smaller map sizes and the anti milking bias of most of the victory conditions. Moves towards a quick game that i might be able to finish in 12 days. Might be a push though 
Also would there be a kind of score redundancy like the GOTM so it's like best 3/5 so people that do get busy or go on holiday or something don't end up being religated. Or would that negate the 'point' of the tourney to find the best player in ALL the conditions?
Also would there be a kind of score redundancy like the GOTM so it's like best 3/5 so people that do get busy or go on holiday or something don't end up being religated. Or would that negate the 'point' of the tourney to find the best player in ALL the conditions?
Originally posted by Matrix
If you all like it, set it up.
Of course not. The idea was never to replace the GOTM.
And moving the thread to general... I thought I created the thread in the General Discussion...
Anyone that is Interested, mail me @ this adress: frst01@student.bth.se Use the Word: "Tournament" in the Subject.
Try to add wich difficulty you usually play, and any Ideas you have about the Tournament.
new to this forum, but not new to civ3..
I think that if you would like to do a hard settings with Large/huge map settings..
1> Give everyone same map, same start position, same empire. same settings
2> Every 50 - 100 years you post your; scores.. expansions, etc.. compare it that way
3> Everyone that falls behind each turn in period, is Eliminated Period.. OR you do a percent of the leader.. Ex. 50 people start the tourney; the first turn in leader has score 2000 the top 50% stay in, so score sub 1000's are eliminated.
This is just a variation, the actual rules could be much less or more demanding.
This would work with a long tournament using the hardest settings and using a huge map..
I hope this helps
Mord
I think that if you would like to do a hard settings with Large/huge map settings..
1> Give everyone same map, same start position, same empire. same settings
2> Every 50 - 100 years you post your; scores.. expansions, etc.. compare it that way
3> Everyone that falls behind each turn in period, is Eliminated Period.. OR you do a percent of the leader.. Ex. 50 people start the tourney; the first turn in leader has score 2000 the top 50% stay in, so score sub 1000's are eliminated.
This is just a variation, the actual rules could be much less or more demanding.
This would work with a long tournament using the hardest settings and using a huge map..
I hope this helps
Mord
Similar threads
