Okay, the gloves are off. Let the fur fly


I would like to hear one single cogent technically accurate explanation of how a multi-way box speaker can be more musically accurate than single drivers or stats. As a speaker designer for more than 25 years, I have yet to hear an argument that holds water, technically. The usual response involves bass or treble extension, as if that is the overriding principle in music reproduction. My position is that any information lost or jumbled in the complex signal path of multi-way box speakers can never be recovered by prodigious bass response, supersonic treble extension, or copious numbers of various drivers. Louder,yes. Deeper,yes. Higher, maybe. More pleasing to certain people,yes. But, more musically revealing and accurate,no. I posted this because I know that it will surely elicit numerous defensive emotional responses. I am prepared to suffer slings and arrows from many directions. But, my question still remains. Can you technically justify your position with facts?
twl
TWL - Of course that descriptive was meant in fun. Your statement at the beginning about slings and arrows - maybe you're trying to catch some where there are none? No need to apologize for having a point of view here (or to ride in on a Trojan horse just to stir the pot!). BTW, I happen to have a fascination with line-source panels, and hope to one day own some as well (my speakers are Thiel CS 2.2's). But in the meantime, do you have any hot suggestions about what to listen to in the way of crossover-less single-driver designs? Cheers yourself!
From a simplistic view, I might ask: if a single driver speaker system is -40db down at 25Hz, could that be considered musically acurate? While it is clear that many trade offs are made using cross-overs, such as the current being out of phase 90 deg. This fact creates a strong argument for electrostatics. Please don't hammer me if my "numbers" are not factual, I am simply trying to present a point.

Cheers, Paul
Zaikesman, I currently use a Fostex FE103 based single driver system in a tubular transmission line configuration, tuned to 41.7Hz. Since the FE103's are only 4" drivers, they will not be everyone's cup of tea. But they are common in high end systems here and in Japan.I drive them with a 1 watt Berning OTL and they give me satisfactory results for my application. With the impedance curve mods I did, and the T-Line loading, the 41.7Hz bass response is achieved. I am not in need of lower bass than that for my listening. SPL levels are low by most audiophile standards, but reasonably loud levels are possible without much compression except on full symphonic and heavy rock. Jazz, vocals, pop, folk,etc. are breathtaking in detail and naturalness, and imaging and soundstage is typical point source perfect(dare I say it?). They won't rock the house down, but I live on a large lot and when I walk the dog, I can hear the music faintly from 30 yards away with the windows closed. They do not suffer the usual HF rolloff of single drivers because of my impedance re-curve mod.(also greatly helps bass response.) So a usable bandwidth from 41.7-20kHZ is not bad for a single driver and SPL's in the upper 90's to boot. The whole ball of wax ran me $250 for the pair. Now, I've worked in high end audio stores and been around, so I've heard plenty of good stuff and these babies are good! If you really need that bass kick, then you could cross in a sub pretty low and you've got it. But when I put on Sade(Best of)I can feel that low bass in my body - with no subs. This type of design can also be done with 6's or 8's too, and you'll get more SPL and lower bass that way. But, the 4" is fast,light, and easy to control at all frequencies. As this whole thread demonstrated, tastes vary. Whether you could live with this sound, I don't know, but I sure can.And I saved alot of jack in the process. I've heard multi-thousand dollar speakers that can't match these within their limitation range. So, if you can live with lower SPL,bass to 40HZ, some compression in the really big music, and stunning, highly detailed and natural sound in all other aspects, these will please you.
Fascinating thread here, Twl - spinning into something more like a tapestry, I'd say...

In my experience, the most fundamental quality a speaker has to get right is tonal balance. This implies adequate bandwidth and a non-irritating midrange. After this, personal preference dictates what priority is placed on clarity, timbral shadings, harmonic richness, coherence, dynamic contrast, soundstaging, bandwidth extension, maximum usable loudness, and so forth.

As an owner of and dealer for both single-driver and multi-driver systems, I think that in the lower price ranges the multi-driver systems are generally the better set of compromises. Then at higher price ranges, single-driver systems begin to overcome their bandwidth limitations, and become much more competitive. As an oversimplified general trend, I'd say below five grand mulitiway systems rule, while above ten grand I'd lean towards single-driver systems, and in between I'm not sure.
Paul, you are correct on both accounts. A speaker down 40 db at 25Hz or any thing like that could not be considered accurate at that frequency. The question is what do you do about it? This has been the discussion above. Read some more above posts, because I can't write a book again on it here. In you crossover statement, basic simple crossovers shift 90 degrees out of phase on a 6db slope. Most audiophile speakers will use 12db because that slope is 180 out, and can be compensated for by switching the polarity on the appropriate driver. 18db slopes are 270, and 24db slopes are 360 or 0. This is simplified theory, but there are other methods used to address phase shift. As for electrostats,the "full range" types are single drivers, albeit a different type than cone drivers. (Some stats use multiple panels though)