Hi Lucas,
Have you tried the time domain approach? Excite the cavity with a plane source or line source with a short pulse. Then using the observation point look at the decay of the time signal, measuring the decay at a time after the input has “turned off”.
You only need enough time steps for the signal to drop to a insignificant level. Try using a couple observation points in the cavity at different points to see how long it takes for the power to drop.
This should be possible in the 32 bit version, its limitation only being amount of memory that it can use.
Let me know if you are still having trouble!