When I started reading this thread, I was going to suggest the Elliot articles, but I see that's been done. Active crossovers are definitely the better way, as passive crossovers introduce compromises as the Elliot articles reveal. For one thing, they dissipate a substantial amount of power, which is one reason you gain more output with active. One point I'll add to this discussion is that audio power distribution is not linear, and more power is needed at lower frequencies than at high frequencies.The 50% power distribution point occurs at about 325 Hz (from a graph I obtained from a National Semiconductor databook some years ago). What this means is that to equal a single 100 watt amp driving a lossless passive crossover (which doesn't exist), two 50 watt amplifiers would need to be crossed over at 325 Hz. For a more typical midwoofer to tweeter crossover of 2 KHz, the lower frequencies would require approximately 85 watts and the higher frequencies only 15 watts to equal a single 100 watt amp. From a practical standpoint, this is a good thing, because most reasonably priced tweeters will not handle a lot of power while woofers and midwoofers will. Beyond the power distribution, eliminating the passive crossover also eliminates the power loss it causes. It's not uncommon to have a passive crossover dissipate half the power delivered to it. So, it might take 200 watts delivered across a passive crossover to equal the sound output of a total 100 watts in an active crossover arrangement. Amps with suitable wattage for the lower frequencies are readily available. Low wattage amps for the upper frequencies are not, although would not be difficult for DIY. Using even a 7075 for the upper frequency split would be overkill for a typical system.