I'm trying to cache a relatively light particle sim. Maybe 40,000 particles at most. The emitters are on motion paths, and in earlier versions of the file the cache would take maybe five minutes to calculate.
As soon as I set the motion paths to "follow," though, the calculation time jumped immensely. (I'm using follow so that another emitter that lags shortly behind the first one trails it in the right way.) The particles aren't inheriting motion or anything. What would cause this huge spike in calculation time? I didn't think motion path evaluation was all that processor-intensive. Has anybody heard of this problem before?