Listening to Horns: The Classic Audio Loudspeakers T3.3
(This is part 3 of a series on designing a speaker with electrostatic transparency and superb dynamics. You may wish to read part 1 and part 2 first.)
With the goals and history of the lost technology of horns behind us, it is time to discuss the sound of John Wolff’s speakers. I mainly listened to the Classic Audio T3.3. These retail for $32,500 per pair in the field coil version that I heard ($14,500 with permanent magnet drivers). John’s system uses Atma-Sphere tube electronics, with both analog and CD sources.
John played a variety of music for me. The quality of what I heard was very consistent disc to disc.
The first thing you notice is a high sense of transparency. Midrange sounds – vocals, piano, violin – are very clear. At the same time, the sound isn’t edgy or harsh so it remains quite listenable. Low-level sounds are very well reproduced.
The sense of transparency extends into the bass. I listened to some percussion, and the kick drum sound was outstanding. Jazz ensemble string bass sounded pretty accurate too.
I kept looking for a tonal balance problem, but they aren’t there in any gross way. I did feel that there was a little hardness to the sound at times that exceeds what one hears in live music, but at the same time we were listening at levels a bit above live acoustic music (or levels appropriate to closer seats than I frequent).
That small degree of hardness, though, seems a reasonable tradeoff for the explosive dynamics that the T3.3 is capable of. I’ve been to a lot of CES demos (over 100) and I can say that the T3.3 I heard outside of Detroit is in the top few percent of what I’ve heard dynamically.
Of course, listening to an unknown system in an unknown room has its limitations. For example, I would say the T3.3 sounds slightly rolled off at the top and bottom, and I’d guess there is a bump in the upper midrange. But any of these things could be due to the room or associated equipment. More listening would also be required to better assess the presence of any anomolies.
Anyway, I was impressed. Just for fun, before I left, John played a few cuts through his version of the JBL Hartsfield ($38,500, shown at left). This is a speaker that the mind says is completely wrong. It has a corner loaded bass horn, so in John’s room the speakers were about 25 ft. apart, and the listening chair was perhaps 15 ft. from the wall they were on. The cabinetry was on full display, with lots of sharp corners for diffraction. The Classic Audio Hartsfield uses two supertweeters arranged on the left and right sides of the speaker.
John lowered the stylus onto an LP and -- son of a gun -- the image appeared dead center, with a solid though wide center image about where you’d expect in a normal stereo arrangement. I happened to be at the opera yesterday and I sat in Row J, near the center of the hall. I couldn't help but think back to the Hartsfields because from Row J the stage width is surprisingly like what I heard in John's room. The dynamics were reminiscent of these speakers too.
The tonal balance didn’t seem quite as accurate as with the T3.3, but the sound was quite enjoyable, and I certainly didn't listen long enough to be even moderately sure of the sound. I could see why these speakers sell in Japan (though how they fit there, I don’t know).
So, is it possible to make electrostatics that are ultra-dynamic? I’d have to say that it looks like it is, as long as you’re not too literal about how you do it. There are some tradeoffs involved, but that is always true.
Certainly this first visit made me want to continue the quest.
Comments
Please add me to your user list.
Thank you,
Ralph
I have to say this. I've held it in for too long now. $38,000.??? YOU'RE NUTS!!!
Grab yourself a pair of Newform Research ribbons at less than 1/10th the price and learn.
"...with lots of sharp corners for diffraction..."
This proves that what's being taken for granted isn't alsways true. Diffraction exist only in relation to the physical length of the sound wave. And contrarely to what's being published by so-called experts, diffraction is a way of increasing dispertion of waves.
$38,000 is more than some people's annual incomes and is a ridiculous amount for speakers. This brings to mind the old PT Barnum phrase, "a fool and his money are soon parted"... Congratulations on an irrelevant article.
For $38,000, I think any hardness is inexcuseable. The fatigue factor of this ungainly "neither fish nor fowl" loudspeaker would soon overshadow the big-grin dynamics.
Jim
$38,000 and still looks like a fireplace! :-)
I have been an "audiophile" since 1968, and started selling high fidelity audio equipment shortly after. In 1972, I realized that most people auditioned loudspeakers incorrectly and instituted the "listening to how real they sound" program for my customers.
Also, with much more sophistication, I used that approach to select which brands I sold.
Every once in awhile, I would run across an outstanding product.
My greatest find is the “Mitsubishi DS-505” – a far superior product than any other loudspeaker Mitsubishi marketed including some very expensive, very impressive looking, floor standing monitors.
What struck me the most about these speakers were their transparency AND dynamics throughout the ENTIRE frequency range.
Have you ever had a chance to listen to these specific Mitsubishi loudspeakers? They are also called the DRMs, Digital Reference Monitors. And, they were the only “Mitsubishi” speaker that was sold in the U.S. with only the name Diatone on them, no Mitsubishi anywhere.
I have been an "audiophile" since 1968, and started selling high fidelity audio equipment shortly after. In 1972, I realized that most people auditioed loudspeakers incorrectly and instituted "listening to how real they sound" program for my customers.
Is the DS-505 a current production item in the US?
It would seem that the true "golden ears" also happen to be the richest folks around. I guess I didn't realize that wealth was a genetically inspired attribute. I love music, but could no more afford these monsters than a new BMW, so I assume that designers building stuff like this, at prices like that, are doing so for their own amusement, with no real intention of allowing the general public to enjoy the fruits of their labors. How utterly selfish! I suppose John must have attended the "Wilson School of Audio for the Rich and Famous Only"...., and, I suppose, we poor slobs don't deserve to hear what's good.
Stop complaning and pick up a book. These are here for a reason. And you can build these yourself if IF you wanted to. And with the same drivers for less. Or you could use better ones, for EVEN LESS.
Don't let the high price bother you. Some can affored it. SOme can't That is why every man use to have a garge and a shop. Apartment dwellers don;t need this veacse they can just BUY everything that they need.
My advise, Look and learn what you want and build what and build what YOU need.
THere are plenty of great speakers out there in the world that just need TWEEKING to get them to be $2,000 speakers.
And It will always be cheaper to just by some mass market bolk enclosure or even store brand take out the woffer and go to town with it.
Enjoy! And DUDE, Have a better day!
My question is: How can they be "transparent" as well as "rolled off at the top and bottom"?!? Weird article... How about reviewing something the rest of us can get into such as the Magnepan MMG speakers with a sub of your choice? Placed properly in a room and matched with a decent sub and decent electronics, these provide the closest to the real thing for the price ($599/pair).
Three reasons that a speaker can be transparent and rolled off at the top and bottom:
1. The vast majority of sounds do not occur in the top or bottom octave
2. I pointed out that the T3.3 sounded slightly rolled off; every speaker has some frequency domain anomolies
3. Distortion is a key source of the lack of transparency
Hope that helps. You will find many reviews of speakers from $100 up on AVGuide, including the MMGs:
http://www.avguide.com/comparative-review/magnepan-mg12-and-mmg-loudspea...
Transparency is an audiophile artefact. It is not present in live music.
Transparency can only be produced with an uneven frequency response.
Electrostatics have a mortgage on 'transparency' because their response is not only unevenly rolled off top and bottom, but is uneven in between too!
The problem with non-horn, non-electrostatic i.e. normal speakers, is that their bass is extended and this will cause a loss of naturalness to the music because the room response will make the bass reponse lumpy with modal resonances.
Electrostatics and small horns (i.e. non-huge horns) are so deficient in bass that the lumpy bass issue is circumvented, and 'transparency' appears.
The (only) proper solution is to equalize the in-room response below 200-300Hz, preferably after treating the room with bass traps. I am quirky and use large high quality home built hand tuned bass horns, whose in-room response is as lumpy as any other extended bass speaker, but I have them fully equalized.
If you are using speakers with deficient or unequalised bass, your lust for 'transparency' will continue, but if you successfully implement equalized extended bass, you will be feeling better real soon! ......Arg
Tom, thanks for your comments on these speakers - I found them useful. Can you speak to image specificity and soundstage depth/height/width ? With regard to the 'roll-off' and 'bump' as possibly room related, did John have any comment on that assessment?
John hasn't commented on my write-up. Thank you for pointing out that some of what I heard could be room related. Some could be ancillary equipment. After all, I'm commenting on an entire system, room included, that I'd never heard before. To some degree pinning characteristics on any one component is risky business.
That said, I should add that I tried to focus on the qualities of the sound that can't be there if the speaker can't do them.
I should expand on my roll-off comment. given the above. If you read some of my reviews (this blog is not a review for the reasons we're discussing) you will notice that I find it useful to characterize the frequency balance as being u-shaped or n-shaped, or another shape as applicable. What I mean by these terms is that it can matter a lot to the overall sense of a component whether apparent frequency response rises very slightly or falls very slightly at the extreme frequencies. U-shaped indicates a slight rise at the extremes, n-shaped indicates a slight fall at the extremes. This does not say this is how the component measures, but rather the sense it conveys. These differences, for all I know, might be less than 1 db. My rolled-off comment about the T3.3s is in the sense of an n-shaped response.
The bump comment is along the same lines. I might even be projecting here; the T3.3s are so dynamic that a reviewer like me, steeped on conventional speakers, is bound to associate this performance with a deviation from neutrality. In fact, it may be simply accurate to the source. It is fine to say "compare it to the absolute sound, dufus," which is what I'm doing, but a close-miked recording may be a little hard and extra-dynamic, and if the speaker reveals that it shouldn't be counted as a negative.
So, I'd suggest that readers focus on what the T3.3s do well. I think this suggests that horns are a very promising technology. Which is the point of the blog, for all the expensive equipment haters.
Depth seemed good, and I felt that this could be a strong point of the speaker. Micro-dynamics seemed exceptional, and you'd think this would be a horn strength. Since micro-dynamic reproduction is key to depth, I think this bodes well for horns, too.
Image specificity was hard to judge. Even though John has the T3.3 speakers on the long wall, they seemed relatively close together relative to their size (these are big speakers). In those circumstances, I found the visual weight of the speaker to distract a bit from the sense of imaging. That might be me, and I know some audiophiles who have big speakers and listen in the dark to get around this problem. One also gets used to these things in one's own home.
If you want to have horn speakers, try the Magico Ultimates at $300K+ per pair. Team them up with Audio Research Reference series amps (5 or 6 depending on the speaker configuration), pre-amp, and CD transports and you will find religion! If you to listem to LPs, add the Rockport Technologies Sirrus (real seroius) III LP player. (Of course,you will need to use the $1000K/foot speaker cable made on the secret Soviet L3 space station.) BTW - The Utlimates can be put anywhere, even outdoors and still sound the same.
For the rest of us digital guys, just stick with the Magico Model 6s and Spectral Audio gear. For LPs, add the ELP laser turntable; for SACD and home theater, add a Mietner stack and the new Denon DVD-A1UDCI player; for video add a JVC or DPI projector.
It's nice to dream...
Having just had my 47th birthday, I know that my ears are now worth about 25K max in terms of audio gear. Spending more than that would just be a waste of money. In the past I have heard the zillion dollar installs and my ear memory, as bad as that test is in general, reminds me that the days of hearing the highs and the last percentage of detail is long gone... too many years of rock and roll roadie work did them in, even with me being really careful.
I really doubt that I am alone in this sad state of being. I have heard the speakers mentioned above at Rocky Mountain. Great stuff. But my ears where happy about 30K less a pair. And before anyone says it - I am not hating on anyone's equipment or their ability to spend big bucks. More power to them. But I suspect that the reality is that most of the advantages are psychological rather than technical.
Which is why I still read these articles - Just like the reviews in Car and Driver for the Ferrari and McLarens. Or the pictures in Playboy if you like <grin>, Audio porn.
The plight of the over 45 male ears... I think I am right about the $100K range myself - too many years in the subway tunnels with tunnel boring machines.
Your article says:
"...the Classic Audio T3.3. These retail for $32,500 per pair in the field coil version that I heard ($14,500 with permanent magnet drivers)..."
Well, that means that the field coil version extra parts increase the cost in nothing less than 18,000 US dollars above the permanent magnet version!
Even using the most expensive magnet wire coils (say, Hexagonal pure silver wire, Teflon coated etc. etc.) and truly esoteric components in the DC power supply (say, several FARADS of capacitance, HUGE toroidal transformers, multiple stages -like a dozen- cascaded voltage super-regulators, etc.); the cost of those all extra parts can hardly reach a fourthor a fifth of the 18,000 US dollars price difference (honestly, UNLESS the field coils are cooled to below Superconducting temperatures by a liquid Nitrogen -or, why not- liquid Helium system!)
Seriously, somebody has to be kidding, or playing a bad joke. (Or as Mark in Canada rightly said: "a fool and his money are soon parted")...
Nevertheless, the article at least gives us, mere mortals, some food for the thought: If the midrange element has to work from 550 (or 400) Hz up to 13 KHz, "...to avoid power response discontinuities in the vital midrange..." then we must follow that the crossover network (and its interaction with the drivers) is causing those undesirable effects.
What continues to surprise me, is that almost all of the "famous" and traditional speaker integrators or "Designers" still keep thinking inside the same old closed mind approach of using PASSIVE crossovers and only one (albeit usuallly very expensive) power amp faced-up to the very difficult task of trying to drive a usually very complex electro-mechanical impedance. (At one time many years ago, I was trying to measure the impedance curve of a woofer, and by putting my hand 12" in front of the speaker hanging from a cord, the impedance plot was altered!)
I humbly invite you and your readers to take a good breath of fresh air and read the excellent, complete and clear articles by Rod Elliott from Elliot Sound Products from Down Under (Australia) on the advantages of Bi-amping (and Multi-amping if you wish). The difficulties of trying to properly drive any type of Dynamic driver through a passive crossover are truly enormous. All those problems CAN be solved easily by using ACTIVE low level crossovers and power amplifiers direcly connected to the drivers, which can now be much better controlled.
In Rod Elliott's view, what he calls the "intelligence band" (midrange) should extend from 300 to 3600 Hz, and is the most important psycoacoustically, therefore the midrange driver should be capable of at least an octave below 300 Hz (150 Hz), which is quite lower than John's selected frecuency of 550 (or even 400) Hz.
It seems to me (and a few of my amazed friends) that the usual "Big Standard Speakers plus Big Very Expensive Amplifier" is the most stale and futile approach. My inexpensive but effective approach to Home HiFi setup is of course Multiamped with active crossovers.
For those readers with enough curiosity (and patience) to read Rod Elliott's articles on active vs passive crossovers, those can be read (for free!) at his site: http://www.sound.westhost.com/bi-amp.htm for parts 1 and 2, and also http://www.sound.westhost.com/biamp-vs-passive.htm
THAT makes sense!
Alfredo M. Claussen.
Post new comment