Lector Strumenti Audio Digitube S-192 D/A converter

Ten years ago, the average consumer was unaware that he or she needed an e-book reader. Since that time, neither those people nor the authors whose books they consume have changed very much. But the people in between have grown restless and unsatisfied, and it is they who call the tune. Consequently, many of you have gone from owning books to sort of, kind of owning books (and sort of, kind of not).

Just as the publishing industry has devised a new way to empty your wallet, so has the record industry found a new way to entice you into buying Kind of Blue for the umpteenth time (footnote 1). That's depressing. But, on the bright side, the latest way of buying digital music has ushered in a new way of playing digital music at home: through a perfectionist-quality digital-to-analog converter with a USB input. And because that technology brings with it a new and honestly better way to listen to CDs—by playing them as perfectly ripped files, without waking the noisy and cognitively challenged guard dog of error correction—I am inclined to simply, in the words of onetime Texas gubernatorial candidate and Petroleum Hall of Fame inductee Clayton Williams, lie back and enjoy it.

And so another promising variation on a new and fruitful formula has come my way, this time from the same region of the present Italy (footnote 2) that gave us Nicolï Amati, Andrea Guarneri, and Antonio Stradivari. Lector Strumenti Audio, a 32-year-old company that made a splash in the US not long ago with tubed CD players that received praise for both their musicality and their reasonable prices, has introduced their Digitube S-192 ($3595), a multiple-input D/A converter that is decidedly USB-friendly, and whose model name gives a clue to its reportedly high-resolution performance.

The Lector Digitube S-192 is built into a chassis that's well styled without silly excess. A two-piece steel clamshell comprises the major portion of the enclosure, with steel front and rear panels and a sedately pretty faceplate of acrylic, with a tinted window for the digital display. The faceplate also includes a small pushbutton for toggling through input choices—the Digitube's only user control, apart from its side-mounted rocker power switch—and a row of five blue LEDs to indicate which input is currently in use. These correspond with five sets of rear-mounted input jacks: two electrical S/PDIF (RCA, BNC), one optical S/PDIF (TosLink), one AES/EBU (XLR), and, of course, one USB (Type B).


The well-finished chassis, supported by three nicely made alloy-and-rubber isolation feet, is filled with a total of seven circuit boards, on the largest of which are the power-supply and audio-output components—the latter including the pair of ECC81 dual-triode tubes that account for another portion of the Lector's name. Two boards adjacent to the rear panel handle the digital-input chores, the smaller built around a Tenor TE8802L USB streaming controller chip. Yet another board plays host to a 32-bit AK4397 DAC chip from Japan's Asahi Kasei Microdevices Corporation (AKM), supported by an AKM AK4113VF digital audio receiver chip. The parts quality is very good throughout, and I was impressed that current-to-voltage conversion appears to be handled by discrete resistors. In contrast with the makers of other recent source components, Lector has eschewed the use of a switching power supply, opting instead for a more traditional supply built around a toroidal mains transformer of reasonable size and apparently good quality. An unusually hefty, hand-terminated, detachable AC cord is supplied as standard.

Installation and setup
I used the Lector Digitube S-192 as a line-level source in my usual system, with Shindo's Masseto preamplifier and Corton-Charlemagne mono amplifiers, and with Altec Valencia and DeVore Fidelity Orangutan O/96 loudspeakers (having recently purchased my review samples of the latter). USB cables were a 1m length of AudioQuest's high-value Carbon and a 2m-long Wireworld Revision—the latter a gesture toward at least minimal parity with my current digital reference, the Halide DAC HD, which is hardwired with 2m of WireWorld's Starlight USB cable. The tubed Lector ran slightly but not excessively warm to the touch.

In addition to using the Digitube as a USB-input converter, I also tried two of its remaining four inputs: TosLink, driven by my Apple iMac's PCM audio output, and RCA coaxial, driven by the non-DSD digital output of my Sony SCD-777ES SACD/CD player. But I primarily relied on the Lector Digitube as a USB source with my iMac, using Stephen Booth's Decibel (v1.2.11) music-playback software for all music files and Apple iTunes for streaming FM broadcasts. Regarding the latter, and while noting the unsuitability of MP3 files for most reviewing chores, let me also note that WCKR, my favorite Internet radio station, sounded fine through the Lector, with sufficient sonic presence that the cowbell in Bix Beiderbecke's 1927 recording of "I'm More than Satisfied" made my own cognitively challenged dog bark from the other room.

The Digitube S-192 is supplied with a CD-R containing a user's manual and the various device drivers required for Windows installations. A driver is not required for Apple OS X systems, but in my first few installation attempts I noted that my iMac had difficulty finding the Digitube. Lector anticipates this—the manual advises users so confounded to simply break and remake the USB connection. I did, and that worked just fine, the Lector now appearing in OS X's Sound/Output window as a selection named "lector-a." After those minor early difficulties, my computer seldom failed to recognize the Lector, even after multiple un- and re-installations for review purposes.

The Lector Digitube S-192 otherwise performed without apparent flaw during its time in my system, its only idiosyncrasy being a rather too audible relay, the clicking of which could be heard from across the room when I manually changed tracks. The sound didn't disrupt the music, of course, but it got a bit old after a while.

Right off the bat, and in comparison with the far less expensive Halide Design DAC HD ($450), the Lector Digitube S-192 had a more powerful, more "physical" bottom end and, to an even greater extent, a more extended treble range. The latter quality brought with it a greater-than-average capacity for conveying texture, which served well the very sweetest and highest-quality recordings—and made a small handful of noisy ones a bit less pleasant. An example of the latter was "I'm Not the One," from the Black Keys' Brothers (ripped from CD, Nonesuch 523994), the grungier textures in which were laid a little too bare for my tastes, compelling me to switch from the mildly relentless Altec horns to the more civilized DeVore O/96s for the remainder of my listening. That said, the Lector did an exceptional job of communicating the impact of the kick drum, and the subtler nuances of force in the electric bass lines.

Footnote 1: I intend no condescension. Ten years ago, I was unaware that I needed handmade braided leaders for my fly-fishing lines, a conical-burr grinder for my coffee, and another guitar. Who could have seen those coming?

Footnote 2: It wasn't until 1870, well after the time of the historically great luthiers of Cremona, that Italy went from being a loose collection of city-states to the present unified nation.

Lector Strumenti Audio
US distributor: Hudson Audio Imports
143 Bergenline Avenue
Closter, NJ 07624
(201) 768-6986

remlab's picture

What does this mean? That the "absolute sound" of a component is all that matters? Horrid measurements have no effect on what some of the best ears in the business can perceive? Yikes!

1audio's picture

I think something is misconfigured. I have never seen or measured an AKM device that was so far from correct operation. You can see measured results for the chip here: http://www.akm.com/akm/en/file/ev-board-manual/AK4397EQ.pdf Everything indicates a problem with the board or with the specific device.

Is it possible that the analog electronics could be the cause of the non-linearity? It seems unlikely but getting the digital side so misconfigured and still working is not easy.

Obviously non-linearity at -90 dB doesn't scream at reviewers saying "I sound bad!" but MP3's demonstrated how much audio can be removed and still have "decent" audio.

Kal Rubinson's picture

FWIW, in the "Manufacturers' Comments" section, Lector Strumenti's US representative states that the designer "deliberately sacrifices measured performance to achieve his design objectives, the most obvious one being superior musicality" and seems not to think that the measurements are unexpected.

Archimago's picture

I read the comment. Looks like a well written manufacturer damage control comment that just regurgitates a number of adjectives from the subjective review (which more likely demonstrates hearing limitations).

Good for John Atkinson to say it as it is. A device that costs a significant chunk of change, that's marketed with big numbers (ooohhh "32 bit DAC from AKM" as seen on the web site, capable of 192kHz and 384kHz) but in reality you'd be wasting dollars feeding this baby >16-bit audio data.

It'd be a shame to think that "musicality" is achieved by sacrificing dynamic range.

remlab's picture

..is that in order to make a digital product sound good, you need someone incompetent to design it. Sign me up!

corrective_unconscious's picture

Damage control is not to double down on a bad result. Damage control is normally to attack the measurer or to "discover" on its return that the unit was broken somewhere along the line. Or to announce revisions to the product "since" the review.

This manufacturer has a far more interesting response than damage control.

qwerty's picture

graph 12 is for sony?
Fig.12 Sony HAP-Z1ES, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 16-bit internal data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

and comment is wrong?
maybe this measurement are all fake!
manufacture web site say: setup the dac in service mode for measurement !

John Atkinson's picture
qwerty wrote:
graph 12 is for sony?

That's correct and fig.8 is for the MSB Analog DAC. Both are examples of what the measured performance on these tests should look like.

qwerty wrote:
maybe this measurement are all fake!

No, they are not fake. The manufacturer stated that the measurements correctly characterized the product in his published comment.

qwerty wrote:
manufacture web site say: setup the dac in service mode for measurement !

As the reviewer didn't audition the Lector DAC in "service mode," I didn't feel that to be relevant.

John Atkinson
Editor, Stereophile

corrective_unconscious's picture

I think as a matter of testing protocol it would have been better to additionally do the measurements and do some listening in service mode.

For the measurements then we could at least have an inkling whether the manufacturer is even capable of designing gear which will measure conventionally, a pretty important piece of data. Then listening to service mode would possibly indicate whether the manufacturer's choice of poorly measuring configurations actually yielded some audible benefit...in one person's system, at least.

I understand then in terms of workload it's like doing two reviews, though, but I'd be really curious about all this.

(Also, as an aside, in my quite old Camino trying to preview and edit my post twice before hitting "save" results in my getting trapped in an endless loop always returning me to preview mode...even when I do eventually hit "save." Either that or I will have double posted, as happened sometimes with the old messaging system here.)

qwerty's picture

service mode is for measure test non for listening.
feedback email with manufacture, without service mode setup internnally switch the dac stay in soft mute condition, so the measure are wrong

corrective_unconscious's picture

A claim to be tested as is any other claim about specs from...whom? A manufacturer?

Speaker companies say their speakers have a certain efficiency; then the efficiency of the speakers gets tested.

One mode allegedly measures well. We don't know that. It wasn't tested.

Not too difficult.

SNI's picture

You write that the Lector uses 32 bit AK4397 DAC chips.
AK4397 is a 24 bit DAC chip which is obsolete.
If it is in fact a 32 bit chip, then it must be AK4399, which I think is also obsolede by now.
You also describe that the I/V conversion is done discretely with use of only resistors.
This cannot be true, as both AK4397 and AK4399 are both U (voltage) out converters, which in general do not require I/V conversion at all.
Anyways I cannot believe that sound would not be bettered, if the measurable components were optimized.