I am going to take a page out of Moonsingers book and post my own "idea".
I was thinking that the automatic document generator could be modified to give some really neat results as well as possibly providing alternate ways of looking at a games strengths (and weaknesses)
The idea would be that we keep track of several statistics each game turn. A new program created from the ideas used in the auto-documentor would record certain key values for every turn of the game that is played while the program is running.
This would not in any way affect the official scoring of games, though it could lead to alternate "good game" indicators such as "strongest production" (like the "first xxx" win indicator).
One characteristic we could keep is the actual "score" based on territory and population. We would gain this score value for every turn of the game rather than just an average at the end. This could be useful for seeing how players score tends to grow in comparison to the Jason curve and maybe even help in "fine tuning" the curve.
Some people don't like the "territory and pop" model of scoring, because it leads to a single playing style - fast domination, which is not the only way to win the game. However going for fast domination is likely to outscore other methods of play. The current scoring method does not take corruption into account. After a certain point, additional cities do no help the strength of your civilization other than the fact that you are also denying the AI the land. In other words, score is not a perfect measure of a civilizations strength in that it does not take units, infrastructure or corruption into account. It might be possible to look at other methods of "keeping score", that are not very accessable from the final save. As I menton above, these methods wouldn't affect the results, but would give a player that is strong but not necessarily in terms of total land something to show that strength
I am thinking the following would be worth tracking (in addition to turn score). Each would be saved each turn and could then be graphed. It would additionally give feedback on how a civ grows (and perhaps more importantly how a good players civ grows while yours does not).
-----------------------------------------------------------------------------
Uncorrupted production of shields - This would be a simple sum of the uncorrupted shields in all cities. It measures the strength of your civ to grow both infrastructure and units.
The sum of gold per turn net + science per turn - I would sum these because they are closely linked and moving the science slider would have a lesser effect (it will have some effect as you may have more market places than libraries for example). It is a measure of your ability to research and purchase enhancements, luxuries, etc. from other civilizations.
Additionally, unit cost and maintenance cost indicators might also help. It would show players the potential problem of ruining ones economy by too many units or buildings. I think I sometimes suffer from the latter myself.
The beaker sum of all the techs known to a civ. This would measure your tech advance rate. It is related to the previous sum, but also includes your ability to make intelligent trades with the AI.
The sum of the cost of all offensive and defensive units to build in shields. This is your "army" size and thus your ability to take land from rival civs. The warrior upgrade strategy would really stand out as a huge jump and make it clear how the experts are using this (and when) which could be valuable for those that have a less clear idea on the concept and how to execute it. It also helps a player to learn what kind of forces are needed to successfully pursue a war.
These are the major ones I can think of. There might be others (a wonder counter?) (number of cities?).
-------------------------------------------------------------------------
This program is runs while you are playing the game. It uses whatever tricks the automatic documentor uses to record results for every turn of the game. These results are printed to a file for each and every turn the utility is run. Players can then submit the final file when their game ends. Then either a specifically designed reader or a program like excell is used to see the results and compare games. This whole process is optional for players, there would be no requirement that you do any of this. However, if you do run the program and submit the result you can compare your game to an experts and see how his play differed, and perhaps learn to play better. This is actually similar to the QSC results, but is available for every turn of the game, rather than just a single year.
Additionally, the results could give the staff some very interesting data to play with. It could show how different aspects of the game grow and how tradeoffs between various stategies used in the game work. It would also give insight in how things like commerce or production grows over time. For example, it is probably a lot more jagged than civ score, with jumps when markets, factories, etc are discovered and built. Finally, it could provide very cool looking graphs for players to ponder and rewards for games that show brilliance in some particular aspect of the game that has nothing to do with score.
The downside of course is that such a progam takes work. The real question is how close is the above program to the automatic document program that is in developement? I am not sure.
Anyway, I thought I would post this idea and see if it goes anywhere.
I was thinking that the automatic document generator could be modified to give some really neat results as well as possibly providing alternate ways of looking at a games strengths (and weaknesses)
The idea would be that we keep track of several statistics each game turn. A new program created from the ideas used in the auto-documentor would record certain key values for every turn of the game that is played while the program is running.
This would not in any way affect the official scoring of games, though it could lead to alternate "good game" indicators such as "strongest production" (like the "first xxx" win indicator).
One characteristic we could keep is the actual "score" based on territory and population. We would gain this score value for every turn of the game rather than just an average at the end. This could be useful for seeing how players score tends to grow in comparison to the Jason curve and maybe even help in "fine tuning" the curve.
Some people don't like the "territory and pop" model of scoring, because it leads to a single playing style - fast domination, which is not the only way to win the game. However going for fast domination is likely to outscore other methods of play. The current scoring method does not take corruption into account. After a certain point, additional cities do no help the strength of your civilization other than the fact that you are also denying the AI the land. In other words, score is not a perfect measure of a civilizations strength in that it does not take units, infrastructure or corruption into account. It might be possible to look at other methods of "keeping score", that are not very accessable from the final save. As I menton above, these methods wouldn't affect the results, but would give a player that is strong but not necessarily in terms of total land something to show that strength
I am thinking the following would be worth tracking (in addition to turn score). Each would be saved each turn and could then be graphed. It would additionally give feedback on how a civ grows (and perhaps more importantly how a good players civ grows while yours does not).
-----------------------------------------------------------------------------
Uncorrupted production of shields - This would be a simple sum of the uncorrupted shields in all cities. It measures the strength of your civ to grow both infrastructure and units.
The sum of gold per turn net + science per turn - I would sum these because they are closely linked and moving the science slider would have a lesser effect (it will have some effect as you may have more market places than libraries for example). It is a measure of your ability to research and purchase enhancements, luxuries, etc. from other civilizations.
Additionally, unit cost and maintenance cost indicators might also help. It would show players the potential problem of ruining ones economy by too many units or buildings. I think I sometimes suffer from the latter myself.
The beaker sum of all the techs known to a civ. This would measure your tech advance rate. It is related to the previous sum, but also includes your ability to make intelligent trades with the AI.
The sum of the cost of all offensive and defensive units to build in shields. This is your "army" size and thus your ability to take land from rival civs. The warrior upgrade strategy would really stand out as a huge jump and make it clear how the experts are using this (and when) which could be valuable for those that have a less clear idea on the concept and how to execute it. It also helps a player to learn what kind of forces are needed to successfully pursue a war.
These are the major ones I can think of. There might be others (a wonder counter?) (number of cities?).
-------------------------------------------------------------------------
This program is runs while you are playing the game. It uses whatever tricks the automatic documentor uses to record results for every turn of the game. These results are printed to a file for each and every turn the utility is run. Players can then submit the final file when their game ends. Then either a specifically designed reader or a program like excell is used to see the results and compare games. This whole process is optional for players, there would be no requirement that you do any of this. However, if you do run the program and submit the result you can compare your game to an experts and see how his play differed, and perhaps learn to play better. This is actually similar to the QSC results, but is available for every turn of the game, rather than just a single year.
Additionally, the results could give the staff some very interesting data to play with. It could show how different aspects of the game grow and how tradeoffs between various stategies used in the game work. It would also give insight in how things like commerce or production grows over time. For example, it is probably a lot more jagged than civ score, with jumps when markets, factories, etc are discovered and built. Finally, it could provide very cool looking graphs for players to ponder and rewards for games that show brilliance in some particular aspect of the game that has nothing to do with score.
The downside of course is that such a progam takes work. The real question is how close is the above program to the automatic document program that is in developement? I am not sure.
Anyway, I thought I would post this idea and see if it goes anywhere.
