I set up my testing by creating a map with no terrain variables, and only TCs, Villagers and either Wolves, Deer or Relics as objects. This made it easy to pick out and count the placed Wolves, Deer or Relics against a plain green background. I tested this using both single objects and groups of objects over the entire range of map sizes, with the same scaling factor derived each time. The test was run repeatedly with 1, 2, 3, 5 and 6 single objects on each map size, and a smaller number of times for groups of 2 and 3 objects to confirm that groups were scaled the same way, and with a few different objects to assure that different objects were not scaled differently. I then rechecked some of the data with forests and other objects placed to ensure that the data was correct.
Here is the scaling factor table from the rmsg.doc, with the observed, calculated scaling factors added from my testing:
Map label ----- Tiles on a side ---- Total Tiles ---- Official Scaling Factor ---- Observed Scaling Factor
tiny_map ............ 72x72 ................ 5,184 .................... 0.5 ........................ 1.33
small_map .......... 96x96 ................ 9,216 ................... 0.9 ........................ 2.0
medium_map ...... 120x120 ............ 14,400 .................. 1.4 ........................ 2.8
large_map .......... 144x144 ............. 20,736 .................. 2.1 ....................... 4.0
huge_map ........... 200x200 ............ 40,000 ................. 4.0 ........................ 4.8
gigantic_map ....... 255x255 ............. 62,025 ................. 6.2 ........................ 5.67
As you can see, the numbers differ significantly from the predicted value according to the rmsg.doc.. On the largest map size, fewer objects than expected are placed, on the huge (next to largest) size map only a few more objects will be placed, and on the smallest 4 map sizes a significant increase in the number of objects are placed compared with the expected.
I will try to check AOM for this kind of discrepancy a little earlier in the life of the game!
[This message has been edited by RF_Gandalf (edited 10-05-2002 @ 11:38 AM).]