Map size and calculation load

[sburke]

I realize this subject has come up a few times about what is managable for the game, but have recently had a game suddenly hit some sort of threshold that is not allowing us to continue after 30+ turns. Am hoping to get some clarification of what numbers really drive the computer processing in order to get a better feel for what is a reasonable size map and scenario.

Broadsword and I are testing out a map and got to about turn 30 and at this point the game locks up within the next few turns no matter what we do. The game has mostly been a low scale conflict up until now, but at this particular moment a couple platoons are engaged on either side along with a few tanks and AT guns and a lot of smoke. Not a large battle itself by any stretch.

I have tried cutting down the map size by a bit to see the impact on file size then compared it to the existing file along with files from another large scenario (Die Ammis Kommen) and a huge scenario (Fire Brigade). I think as was pointed out somewhere on the forum, there is a geometric increase of file size related to map size and that does seem to be true. What I still don't understand is why it ran for 30+ turns and then exhibited issues. What expectation is the processor load when say 2 companies of Infantry are firing upon one another? My PBEM of Die Ammis Kommen has roughly a comparable number of units engaged right now (Comparable to the "Nicole" battle - the actual fighting forces are measured in platoons rather than companies).

For a map reference I took the map size divided each side by 8 to get the number of action squares then multiplied to get the total number of action squares per map (hoping I have my math right there). La Nicole is significantly larger in number of action points, but the file size is only double. If there is any conclusion to draw from this it could be that the increase in map size isn't necessarily reflected in the file size increase so much as the number of calculations required based on a larger percentage increase in action sq.

Die Ammis Kommen - Map 1040x1520 24,700 action sq file size 11,349k

La Nicole - Map 1888x2420 71,390 action sq file size 22,655k

La Nicole lite - Map 1808x2256 63,743 action sq file size 18,468k

Fire Brigade - Map 3232x3360 169,680 action sq file size 46,992k

Though the number of units involved in Die Ammis Kommen is larger than La Nicole and more are due to arrive, the map is significantly smaller and the number of action squares is 1/3. Am currently in a PBEM at about turn 15 or so and it isn't exhibiting any issues.

I am assuming that there are a lot of issues involved and it isn't a simple calculation of X number of units x Y number of action squares x Z number of projectile calculations to perform. The game is also calculating LOS to determine who can even fire, concealment factors from smoke and terrain etc. What I guess I am looking for is some proportional expectation. If you have a large number of units, you can only expect to get to maybe a 1x1 map, but you can go to a 4x4 map if you have one M8 vs 1 PSW.

Any thoughts from the playtesters? I realize you also have to throw in what rig you are running on, but looking for a best case scenario here and can work down from there.

[RockinHarry]

Good luck for getting any responses from BFC here. Has been pointed to in the Tech Support forum numerous times, but we´ll likely need to wait for patches to deal with it, without it beeing explained furtherly. However, before I spend any more time working on larger scenario projects, or Euros for oncoming modules, I like to have the issues cleared.

[sburke]

LOL yeah I know. I wasn't looking to get an "official" response. I honestly think there isn't a set answer as the range of systems out there makes any attempt to do so pointless. I am more curious to see if in the limited time we have had to actually learn and experience where the boundaries might be if folks have some idea. I have no problems putting in the effort to make a large map, just wondering when it comes to slicing and dicing it for individual scenarios what scale we need to be thinking about.