First, Let's Clear Up Some Confusion
Bi-wiring and bi-amping are not the same thing. Also, what you probably think of as bi-amping is probably not actual bi-amping.
Multi-amp: All frequencies are run through the pre-amp and amplifier and are presented at the speaker terminal
Bi-amp: LF and MF/HF (or LF/MF and HF) are split into separate signal paths prior to the amplifier
Tri-amp: LF, MF and HF are split into three signal paths prior to the amplifer before presenting three separate signals (LF,MF,HF) at the specified speaker terminal
An active subwoofer being sourced by a receiver's sub out is a basic form of bi-amping.
The scope of the information in this article does not apply (for the most part) to digital audio systems.
Bi-amping is mostly about power consumption with some benefits in reduced IM - if the system is set up properly. .
Bi-amping (and tri-amping) is pretty much a necessity in the pro-audio world. This is because the amount of power needed to make all of those lip-synching artists sound so loud in your local hockey arena would (literally) make your head explode.
Without bi- and tri-amping that insane amount of power would be exponentially greater. This is because in multi-amped systems passive crossovers are used and passive crossovers suck up an amazing amount of power. Another factor is the fact that live sound is not compressed the way recorded music is, so power levels tend to swing wildly throughout the preformance. Add that all together and the only feasible way to make an act sound decent without shutting down the local power grid is to tri-amp or bi-amp the audio.
This is accomplished through the use of active crossovers placed between the source and the amplifier.
If you aren't using an active crossover between the source and the amplifier, you are not actually bi-amping anything, but to make you feel better the audio gods have coined the term "passive bi-amping." To truly bi-amp your home system, you'd have to remove the passive crossovers from the speaker cabinets and use an active crossover in the signal chain above the pre-amps. But if you've spent a fortune on a good pair of speakers with well-designed passive crossovers, it seems kind of dumb to pull the crossovers out in order to save a few watts of power or reduce some theoretical intermodulation distortion (IM).
Also, keep this in mind: Unless you are pulling out the passive crossovers and replacing them with a well-tuned and highly stable active crossover network, and then tuning that active crossover properly with the proper test equipment, the power benefits you gain probably isn't worth the extra effort and money.
In terms of intermodulation distortion, you are basically replacing one circuit that may or may not cause IM with another circuit that may or may not cause IM. If you've bought a decent pair of speakers with a well-designed passive crossover network you may not realize any noticeable gains, and in fact, unless you know how to set an active crossover, you are more than likely causing more harm to your overall sound quality than if you had just left your system alone.
What About Power?
A passive crossover uses inductors, resistors and capacitors. Capacitors not so much, but inductors and resistors take a lot of the power they receive and turn it into heat before it gets turned into sound. This is why systems in commercial venues need to use active (electronic) crossovers to get the job done. So, if you're "bi-amping" your home system but still using the factory installed passive crossovers, the gain in power efficiency you were looking for when you went to bi-amp in the first place will be negated because of all of those energy-to-heat-transferring resistors and inductors. Granted, the MF/HF may be powered with a lower power amp, therefore creating less chance for harmonic distortion, but if you're going to run that low power through a passive network, you're going to need to increase the output current to keep up with your low frequencies. The Catch-22 of bi-amping.
Of course, then you have to concern yourself with matching the volume of the HF with the volume of the LF so you can listen to your music as it was intended to sound. This is often overlooked but it's a pretty major thing.
Also, if you are not splitting the frequencies before the power amp you are not reducing the stress of frequency-related loading within the amplifier circuit and power supply as the amps are still receiving the full frequency range. This means the amp is producing the full frequency range and presenting it to the speakers regardless of what you want it to do.
Horizontal and Vertical Bi-Amping
In spite of all of this, you say you are still willing to see if a bi-amped system is right for you. Okay. Are you going to horizontally or vertically bi-amp your system? And does it matter?
In a horizontally bi-amped system, one stereo amp powers both (L & R) bass drivers and one stereo amp powers both (L & R) MF/HF drivers.
In a vertically bi-amped system, one stereo amp is used for the MF/HF and LF (L channel) and another stereo amp is used for the MF/HF and LF (R channel).
In both cases, you can think of it in terms of having four separate amplifiers.
Note that monoblock systems are not necessarily bi-amped systems and aren't covered by this article.
Theoretically, a vertical system will be more efficient because the heavier loading of the lower frequencies is split between two amps (and subsequently two power supplies). Meanwhile, some people like the horizontal system because they can use one type of amp that may work better with high frequencies (like a nice low-power tube amp) and one type of amp that works better with low frequencies (like a meatier high-power solid state amp). Unfortunately, like most things in life, this sounds really neat on paper but in the real world there are unintended consequences that may cause you more aural grief than sonic joy.
Getting the signal levels of the two different-type amps to match is extremely hard to do without the right measurement equipment. Therefore, you may not be listening to your music with the intended balance between the highs and lows
It is very difficult to get two different amplifiers with two separate sonic characteristics to play nice with each other in the midrange area (where most of the musical information resides)
So once again, if you are considering bi-amping your system, tread carefully and do as much research as you possibly can before taking the plunge. You may be better off spending your money on a really good, clean, 500WPC integrated amp, than buying 2 250WPC amps.
After all of the above is taken into consideration there are still some benefits to bi-amping.
Transient signals are less likely to be present within a frequency range and are therefore less likely to cause amp overloading and clipping
Reduced interference from CEMF (counter electro-motive force). CEMF is developed by a hard-working LF driver that "pushes" electrical energy back to the crossover. Basically, as a driver moves back and forth, it produces a current in the voice coil. During a quiet passage immediately following a loud passage some of this CEMF may potentially interfere with other frequencies
A Subjective Conclusion
Unless you're willing to spend seriously large amounts of money and you're then willing to take out the crossovers that were designed for - and installed in - your speakers and you're then willing to take the time to properly set your active crossovers, your money may be better spent on a really good pair of speakers and a really good amplification/pre-amplification system.
Can you benefit somewhat from even a passively bi-amped system? To an extent yes, but the return on your investment may not be enough to offset the benefits of spending your money on a decent to really good amp/pre-amp system.
Another thing to consider: Our engineers spend an enormous amount of time and money getting our passive crossovers set precisely for the cabinet volume and driver configuration of our speakers so you don't have to.
There will be some who swear this piece is dead-wrong and that's okay. A lot of this comes down to what we think we hear and there is nothing wrong with that, but those are the technical reasons for going one way or the other.