ZOOMIE1980 -> RE: 1 Pilot - 1 Mission - 10 Kills? (8/9/2004 5:12:28 PM)
|
A lot has to do, I'm sure with how often the Random Number generator is 1) used in a computation sequence, 2) seeded. Consider the the Random number generator call being a roll of the dice and seeding being using a new die. Is every air-to-air shot taken attempted a new roll of the die or is it rolled only every so often in an outside loop of some sort. What about seeding. A "bad seed" can result in some fairly screwy number outputs. What it is seeded with can have a lot to do with it as well (the seed, itself, should be somewhat random as well, like the milli/micro second part of the system clock). And we will HOPE they are not using rand(), but something homegrown and a bit more "robust". And then we get what I call the "agent queue" issue. This refers to how, in a Telemarketing call center, for instance, you keep agent 1 from getting inundated while agent five gets almost nothing (i.e. how to evenly disperse work across the queue). This problem here seems to be the same sort of problem we see when one AK takes 50 hits and the others in the TF barely get scratched. In this case, it appears Pilot 1 may be getting "picked to fight" more often than the others
|
|
|
|