None but the most pedantic of commenters would debate the contention that Sony has made more great audio products than any other company in history, but Sony typically hasn't made particularly good speakers. Why? I don't know, but the situation has spawned all sorts of absurd and unsupported speculations. I even heard from an employee of a different Japanese audio company that some Japanese believe it's because the speakers are tuned for Japanese people and Americans' bigger noses change the sound.

In the last few years, though, Sony has put such nonsense to rest with a series of super-speakers that have earned rave reviews around the world. The effort started in 2011 with the $27,000/pair SS-AR1, and I've been hoping the work done on that speaker would trickle down into more affordable products. Until now, though, the least expensive of the company's new and improved tower speakers was the $10,000/pair SS-NA2ES. That's an unremarkable price in a world where $50,000/pair high-end speakers are common, but still it's more than most audiophiles have to spend.

Additional Resources
• Sony HAP-S1 Hi-Res Music Player Reviewed at HomeTheaterReview.com.
• Sony Adds New Portable DAC and Integrated Amplifier at HomeTheaterReview.com.
• Visit Sony's brand page at HomeTheaterReview.com.

This spring, Sony shocked me when it announced the SS-CS3, a tower speaker clearly influenced by the SS-NA2ES but priced much more affordably. "So...$5,000/pair?" you might be thinking. Nope. Try less than one tenth as much.

The SS-CS3 lists for $239 each, or $478/pair. For a tower speaker with dual 5.25-inch woofers, that's not a bad price at all, especially when you consider it's not easy to find a really good-sounding bookshelf speaker at that price...and when you consider that the SS-CS3 looks nice. A little plain, perhaps, but not in any way cheap.

The main element the SS-CS3 shares with the SS-NA2ES, besides the look, is its 0.75-inch super-tweeter, which is intended to reproduce frequencies up to 50 kilohertz. What does that extended frequency response do besides annoy your Chihuahua? It reproduces the wider bandwidth of high-resolution digital audio files that use DSD or 24-bit/96-kilohertz PCM.

You can't hear that high, but it's said that expanding the bandwidth of a recording shifts the phase effects of the anti-aliasing filter in the analog-to-digital converter to frequencies where you can't hear them. Of course, a conventional tweeter also introduces its own phase anomalies as its frequency response starts to roll off above 20 or 25 kHz or so. Theoretically, then, the super-tweeter's relatively flat response way up to 50 kHz preserves the pristine phase information that high-resolution audio captures.

And if all that just sounds like a bunch of technical gobbledygook...well, you can still annoy your Chihuahua.

Below the super-tweeter sits a one-inch conventional tweeter, with an ordinary polyester fabric dome. The twin woofers use cones made from dual-layer foamed mica. What's that? I have no idea, but I can tell you it feels about as light as a doped-paper cone, but it's stiffer and seems better-damped, which means it should distort less.

The enclosure is ported below the bottom woofer, and the back panel has speaker-cable binding posts that are nicer than those on many high-end speakers. A removable grille snaps on to cover the drivers, but the speaker looks and sounds much better without the grille. The matte-black-finished cabinet is well built and fairly solid for something in this price range.

The SS-CS3 is the top model in the line. There's also the $169 SS-CS8 center speaker, the $239 SA-CA9 10-inch subwoofer, and the $219/pair SS-CS5 bookshelf speaker.

The Hookup
Because the SS-CS3 is small (at just over 36 inches high) and light (at just over 30 pounds), it's much easier to unpack and set up than most of the speakers I review. I worried that the towers might need to go closer to the wall behind them to reinforce the bass, but no - the balance sounded just right with the speakers in the same places I usually put my Revel F206 towers. That means the backs of the speakers were 24 inches from the wall behind them; the speakers were nine feet apart and 10 feet from my head.

As usual, I used my Krell S-300i integrated amp for the review. Of course, it's unlikely anyone will use a $2,500 amp with a $478/pair speaker, but it's the amp I'm used to and I'm sticking with it.

This speaker is inexpensive enough that it might find itself mated with one of those $20 Lepai or Pyle Class D amps from Parts Express. That would be a shame because my measurements have shown those amps can deliver sub-optimal results when connected to even a medium-impedance load like the six-ohm SS-CS3. No, it's not like you have to connect the SS-CS3 to a Krell or even an NAD, but at least use a decent stereo receiver.

My audio sources were a Sony PHA-2 USB DAC/headphone amp connected to the Toshiba laptop that holds my music collection - a good choice because the PHA-2 handles high-res DSD and PCM audio, which is what the SS-CS3 was designed in part to handle. I also use a Music Hall Ikura turntable connected through an NAD PP-3 phono preamp, plus a Samsung BD-C6500 Blu-ray player.

Performance
Often when I'm listening to high-end audio gear, I think to myself, "It's so great to be able to listen to the music I love through such nice gear." I think it every single time I listen to my Revels. What surprised me was how often I thought it when I was listening to the SS-CS3.

Here's a great example. I was messing around with a preproduction sample of the Constellation Audio Cygnus, a very high-end music streamer with an unusual Web-based control interface. I was clicking buttons on my Web browser trying to get it to play, when suddenly Holly Cole's unaccompanied voice blared out of the speakers at extremely high volume as she tore into her rendition of "If I Were a Bell." (I'd failed to anticipate that the Cygnus would have a much higher output level than the phono preamp I'd just been using.) I quickly hit pause and turned down the volume on the Krell, while marveling that the voice sounded clear, smooth, and entirely free of distortion even at such a loud level. "Good thing I had the Revels hooked up," I thought. But when I went to change some connections in the system, I realized that I'd been hearing the Sonys, which were right next to the Revels. Not bad for under $500, huh?

But anyone who wants a crank-speaker isn't going to buy this one. If you're putting the SS-CS3 into a two-channel rig, I figure you're buying it to listen to music of above-average refinement at moderate levels. Maybe you'd listen to something like jazz saxophonist Charles Lloyd's classic Forest Flower album - which I spun through the Sonys because I happened to have the original vinyl release on loan from a friend.

On material like Forest Flower, the Sony speakers really shine, approaching the sound quality of much more expensive speakers like the Revels. The dual 5.25-inch woofers had plenty of oomph to carry the weight and groove of Cecil McBee's upright bass, but more important was the stereo imaging and dynamics. I loved hearing Keith Jarrett's jarring solo toward the end of side one, with Jarrett pounding full-bore in the piano's top octave, then hitting and plucking at the strings. This is tough for a speaker to play well because dynamics are so hard for a tweeter to handle, but the SS-CS3s sounded completely clear and natural, never seeming troubled by Jarrett's theatrics. Even though the mix puts the piano hard in the left channel, I still got a convincing portrayal of the body and size of the instrument.

Go to Page Two for more on Performance, The Downside, Comparison & Competition, and Conclusion...

Performance (Continued)
Even better was Muddy Waters' rendition of "Good Morning Little Schoolgirl" from Folk Singer, downloaded in 24/96 PCM from HDTracks. This, to me, is the kind of thing you buy a high-end system for: an uncannily lifelike, captivating portrayal of a performance. This recording is famous in part for the natural, uncolored tonality of Waters' voice, and the SS-CS3 seemed to do almost nothing to add any color of its own. This kind of vocal rendition is too rare, even in some high-end speakers; to hear it coming from a $478/pair speaker is simply amazing.

When I watched a few movie clips from Blu-ray discs through the SS-CS3, I was impressed by the speaker's high output level and its astonishingly low bass distortion, even when I had the Krell cranked almost all the way up. The scene from the first Thor movie, in which Thor battles a metallic, fire-breathing creature called the Destroyer, is one of my favorite bass test scenes because the bass isn't all that low in pitch, but it's extremely dynamic. The pair of SS-CS3s working together delivered about the same impact and bass power I'd expect from a typical $300 or $400 10-inch subwoofer. I actually felt a mild sense of impact in my chest, and most important, I heard not one bit of strain or distortion in the low notes, even at very high volume. Pretty good woofers for a $478/pair speaker!

Measurements
Here are the measurements for the Sony SS-CS3 speaker. Click on each image to view in a larger window.

Frequency response
On-axis: ±3.6 dB from 58 Hz to 20 kHz
Average: ±3.8 dB from 58 Hz to 20 kHz

Impedance
Minimum 4.5 ohms/237 Hz/+-4°, nominal 6 ohms

Sensitivity (2.83 volts/1 meter, anechoic)
87.7 dB

The first chart above shows the frequency response of the SS-CS3, the second shows the impedance. For frequency response, two measurements are shown: at 0° on-axis (blue trace) and an average of responses at 0°, ±10°, ±20° and ±30° (green trace), all measured on the horizontal axis.

The SS-CS3 has a pretty smooth response with no major peaks or dips except for a -2.5dB dip at 3 kHz. The main characteristic is a slight downward tilt to the tonal balance, meaning the treble may sound a little soft. I'm going to take a guess and say that the de-emphasis at 3 kHz is what gave dialogue in movie soundtracks a slightly bright character, by emphasizing the frequencies above the dip. The averaged "listening window" response from 0° to ±30° is very close to the on-axis response, which is good - that'll help make the SS-CS3 less sensitive to positioning, toe-in, and room acoustics.

These measurements were done without grilles. The grilles make the response just slightly less even, introducing a -2dB dip at 2,850 Hz, a -3dB dip at 6,150 Hz, and a boost of +1 to +1.5 dB between 7 and 13 kHz.

Sensitivity of the SS-CS3 is average at 87.7 dB, measured quasi-anechoically from 300 Hz to 3 kHz. Expect about +3 dB more output in-room. Nominal impedance is 6 ohms, no problem for even most cheap A/V receivers, but don't try to run the SS-CS3 off one of those sub-$50 amps from Parts Express. It'll probably work, but it might shut off if you push it.

Overall, these are good measurements for the price.

Here's how I did the measurements. I measured frequency responses using an Audiomatica Clio FW 10 audio analyzer with the MIC-01 measurement microphone and the speakers driven with an Outlaw Model 2200 amplifier. I used quasi-anechoic technique to remove the acoustical effects of surrounding objects. The SS-CS3 was placed atop a 28-inch (67-cm) stand. The mic was placed at a distance of two meters. Bass response was measured using ground plane technique, with the microphone on the ground two meters in front of the speaker. Bass response results were spliced to the quasi-anechoic curves at 210 Hz. Results were smoothed to 1/12th octave. Post-processing was done using LinearX LMS analyzer software.

The Downside
As impressive as the SS-CS3's output is for its size, you can only do so much with two inexpensive 5.25-inch woofers. When I cranked the volume up during more rocking tunes like The Cult's "Wild Flower," the midrange, treble, and upper bass stayed clear, but the lower regions of the bass (below 80 Hz or so) compressed. Thus, the sound thinned out a lot. It wasn't bad. It just wasn't kick-ass.

When I compared the SS-CS3 with my Revel F206s, I could hear that, while the Sony is a great speaker, it's lacking things that a good high-end speaker delivers. While it was pretty shocking to hear how closely the SS-CS3 matched the tonality of the F206, the F206 just sounded a little bit better in almost every way. It was both smoother and more detailed, especially in the midrange; Waters' voice sounded the same in many ways, but through the F206 I got a better sense of his breath and the natural resonance of his voice. It was similar in ways to the kind of improvement you get by going from a 128-kbps MP3 to a 256-kbps MP3: basically, everything sounds a little better.

I noticed also that, when I was watching movies through the SS-CS3 - again, with no subwoofer in the system - voices sounded a little too trebly and a little bit edgy. Not to the point that my enjoyment was lessened, but it was a noticeable flaw. I expect that adding a subwoofer to bring in a little more bottom end and tilt the tonal balance a little toward the bass would help.

Comparison and Competition
There are quite a few decent tower speakers in the SS-CS3's price range, including most notably the Polk TSi300 and TSi400, the Cambridge Audio S70, the Infinity Primus S363, and the Klipsch Reference RF-52 II. All of these cost around $400 to $450 per pair, and most are similar in size and driver layout to the SS-CS3 (although none of them has a super-tweeter, and the Infinity has dual 6.5-inch woofs plus a four-inch midrange).

Unfortunately, I haven't reviewed any of those specific models, although I have reviewed other models in all of those lines. Based on those impressions gained in past reviews, and on how well the SS-CS3 performed in a comparison with my $3,500/pair Revel F206s, I expect the Sony could at the very least hold its own against any of those speakers.

Conclusion
Sony did an astonishingly good job on the SS-CS3. In fact, it's so good that I can't imagine anyone not liking this speaker. What's not to like? It has a mostly neutral sound, strong imaging, great dynamics, and quite respectable bass for a small tower speaker. Of course, serious audiophiles will probably want to step up to something better, but I expect they'll have to spend at least twice the SS-CS3's price to get something that significantly outperforms it.