T+A MP 3100 HV SACD player/streaming DAC Measurements

Sidebar 3: Measurements

I measured the T+A MP 3100 HV with my Audio Precision SYS2722 system (see the January 2008 "As We See It"). Looking first at the MP 3100 HV as a disc player, its error correction was the best I have encountered—there were no glitches in its output with any of the gaps in the data spiral on the Pierre Verany Digital Test CD. The optical and coaxial S/PDIF inputs and the AES/EBU input locked to datastreams with sample rates up to 192kHz. Apple's USB Prober utility identified the processor as "MP 3100 HV HD-Audio" from "T+A elektroakustik" and indicated that the T+A's USB port operated in the optimal isochronous asynchronous mode. Apple's AudioMIDI utility revealed that, via USB, the MP 3100 HV accepted 32-bit integer data sampled at all rates from 32 to 384kHz.

The maximum output level varied according to which input was selected. Playing a 1kHz tone at 0dBFS on a test SACD resulted in an output level of 4.964V, balanced, and 2.483V, unbalanced, both into 100k ohms and confirming the specified levels. With the USB, S/PDIF, and AES/EBU inputs, the maximum output levels at 1kHz were 0.36dB lower, at 4.752V, balanced, and 2.375V, unbalanced. The maximum output levels playing a CD were lower still, at 4.038V and 2.018V, though the latter does conform to the Red Book standard. Both sets of outputs preserved absolute polarity with all inputs. The unbalanced output impedance is specified as 50 ohms. I measured 45.7 ohms at 20kHz and 1kHz, increasing very slightly to 54.5 ohms at 20Hz. The balanced output impedances were exactly twice the unbalanced values, as expected.

The four oversampling digital reconstruction filters are labeled FIR Long, FIR Short, Bezier IIR, and Bezier. The FIR Long filter had a conventional time-symmetrical linear-phase impulse response with 44.1kHz data (fig.1); the FIR Short filter, as its name suggests, is shorter (fig.2). With the Bezier IIR filter selected, which I understand was used by Jim Austin for most of his auditioning, the T+A's impulse response had a single cycle of preringing and two of postringing (fig.3). The Bezier filter offered an almost time-perfect symmetrical pulse (fig.4).

520TA3100fig01

Fig.1 T+A MP 3100 HV, FIR Long filter, impulse response (one sample at 0dBFS, 44.1kHz sampling, 4ms time window).

520TA3100fig02

Fig.2 T+A MP 3100 HV, FIR Short filter, impulse response (one sample at 0dBFS, 44.1kHz sampling, 4ms time window).

520TA3100fig03

Fig.3 T+A MP 3100 HV, Bezier IIR filter, impulse response (one sample at 0dBFS, 44.1kHz sampling, 4ms time window).

520TA3100fig04

Fig.4 T+A MP 3100 HV, Bezier filter, impulse response (one sample at 0dBFS, 44.1kHz sampling, 4ms time window).

With 44.1kHz-sampled white noise, the T+A's response with the FIR Long filter featured a fast rolloff above the audioband (fig.5, red and magenta traces), with the aliased image at 25kHz of a full-scale 19.1kHz tone (blue and cyan traces) suppressed by almost 100dB. The FIR Short filter didn't reach full stop-band attenuation until 28kHz and the Bezier IIR filter until 39kHz (fig.6, red and magenta traces), with the aliased image of the 19.1kHz tone (blue and cyan traces) down by just 12dB. An unusual sculpting of the ultrasonic noise floor, centered on 70kHz, can be seen with the white noise signal. The Bezier filter offered an even slower ultrasonic rolloff (fig.7).

520TA3100fig05

Fig.5 T+A MP 3100 HV, FIR Long filter, wideband spectrum of white noise at –4dBFS (left channel red, right magenta) and 19.1kHz tone at 0dBFS (left blue, right cyan) into 100k ohms with data sampled at 44.1kHz (20dB/vertical div.).

520TA3100fig06

Fig.6 T+A MP 3100 HV, Bezier IIR filter, wideband spectrum of white noise at –4dBFS (left channel red, right magenta) and 19.1kHz tone at 0dBFS (left blue, right cyan) into 100k ohms with data sampled at 44.1kHz (20dB/vertical div.).

520TA3100fig07

Fig.7 T+A MP 3100 HV, Bezier filter, wideband spectrum of white noise at –4dBFS (left channel red, right magenta) and 19.1kHz tone at 0dBFS (left blue, right cyan) into 100k ohms with data sampled at 44.1kHz (20dB/vertical div.).

With the two FIR filters and data sampled at 44.1kHz, the MP 3100 HV's frequency response was flat almost up to 20kHz, with then a sharp rolloff just below half the sample rate (not shown). At higher sample rates, these two filters were flat to 20kHz but started rolling off before half of each sample rate. (I had the analog low-pass filter set to Normal for the testing.)

The Bezier IIR filter was peculiar, in that a small, 0.75dB-high peak was present with 44.1kHz data (fig.8, green and gray traces). The peak was lower in amplitude and higher in frequency with 96kHz data (cyan and magenta traces), and was absent with 192kHz data (blue and red traces). The Bezier filter started rolling off earlier than the other filters and was down by 3dB just below half of each sample rate (fig.9), suggesting that it resembles the slow-rolloff filters first seen in the 1990s with products from Wadia and Pioneer.

520TA3100fig08

Fig.8 T+A MP 3100 HV, Bezier IIR filter, frequency response at –12dBFS into 100k ohms with data sampled at: 44.1kHz (left channel green, right gray), 96kHz (left channel cyan, right magenta), 192kHz (left blue, right red) (0.5dB/vertical div.).

520TA3100fig09

Fig.9 T+A MP 3100 HV, Bezier filter, frequency response at –12dBFS into 100k ohms with data sampled at: 44.1kHz (left channel green, right gray), 96kHz (left channel cyan, right magenta), 192kHz (left blue, right red) (0.5dB/vertical div.).

The T+A's channel separation was superb, at >125dB in both directions below 1kHz and still 113dB at the top of the audioband. When I increased the bit depth from 16 to 24 with a dithered 1kHz tone at –90dBFS (fig.10), the random noise floor dropped by around 22dB, meaning that the MP 3100 HV offers almost 20 bits' worth of resolution, which is enough to give a clean spectrum with a dithered tone at a roots-of-the-universe –120dBFS (fig.11). With undithered 16-bit data representing a tone at exactly –90.31dBFS, the three DC voltage levels described by the data were perfectly resolved (fig.12). With undithered 24-bit data, the result was an excellent sinewave (fig.13)

520TA3100fig10

Fig.10 T+A MP 3100 HV, FIR Long filter, spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with: 16-bit data (left channel cyan, right magenta), 24-bit data (left blue, right red) (20dB/vertical div.).

520TA3100fig11

Fig.11 T+A MP 3100 HV, spectrum with noise and spuriae of dithered 1kHz tone at –120dBFS with 24-bit data (left channel blue, right red) (20dB/vertical div.).

520TA3100fig12

Fig.12 T+A MP 3100 HV, FIR Long filter, waveform of undithered 16-bit, 1kHz sinewave at –90.31dBFS (left channel blue, right red).

520TA3100fig13

Fig.13 T+A MP 3100 HV, FIR Long filter, waveform of undithered 24-bit, 1kHz sinewave at –90.31dBFS (left channel blue, right red).

The T+A MP 3100 HV offered very low levels of harmonic distortion into high impedances. With a full-scale 50Hz tone, the second harmonic was the highest at level, but at –104dB in the left channel (fig.14, blue trace) and –96dB in the right channel (red trace), these are inconsequential. Higher-order harmonics all lay below –110dB (0.0003%). However, the output clipped with the balanced outputs driving a full-scale tone into 600 ohms. I had to reduce the signal level to –6dBFS to reduce the levels of the distortion harmonics to below –90dB (0.003%) into this very demanding load.

520TA3100fig14

Fig.14 T+A MP 3100 HV, spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

While the T+A's intermodulation distortion was very low in level, the second-order difference product produced by equal-level tones at 19 and 20kHz with the combined waveform peaking at 0dBFS lying around –104dB (0.0006%), the suppression of the ultrasonic aliased images of these tones depended on the digital reconstruction filter in use. While the aliased tones were attenuated by 90dB or more with the FIR Long filter (fig.15), the Bezier IIR filter suppressed them by around 20dB (fig.16) and aliased images appeared in the audioband, though these are still at a very low level.

520TA3100fig15

Fig.15 T+A MP 3100 HV, FIR Long filter, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 0dBFS into 100k ohms, 44.1kHz data (left channel blue, right red; linear frequency scale).

520TA3100fig16

Fig.16 T+A MP 3100 HV, Bezier IIR filter, HF intermodulation spectrum, DC–30kHz, 19+20kHz at –6dBFS into 100k ohms, 44.1kHz data (left channel blue, right red; linear frequency scale).

When I tested the T+A processor for its rejection of word-clock jitter with 16-bit TosLink J-Test data, all the odd-order harmonics of the LSB-level, low-frequency squarewave were at the correct levels (fig.17, sloping green line). No sidebands were visible when I repeated the jitter test with 24-bit optical J-Test data, but a very low-level pair of sidebands appeared at ±80Hz with 24-bit USB data (fig.18), these of unknown origin but most likely inconsequential.

520TA3100fig17

Fig.17 T+A MP 3100 HV, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 16-bit optical data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

520TA3100fig18

Fig.18 T+A MP 3100 HV, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 24-bit USB data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

The T+A MP 3100 HV did well on the test bench, though those rare preamplifiers with an input impedance of less than 1k ohms should be avoided.—John Atkinson

COMPANY INFO
T+A elektroakustik GmbH & Co.
T+A North America
(207) 251-8129
ARTICLE CONTENTS

COMMENTS
barrows's picture

With DACs like this T+A unit, which have a totally separate, discrete, conversion scheme for DSD playback, it would be desirable to see a set of measurements for DSD playback. Many audiophiles are using these types of DACs with playback software which allows all file formats to be oversampled to DSD 256 or 512 for playback, because many find that this approach to D/A conversion offers improved sonics over PCM conversion, especially with sophisticated computer based oversampling to DSD allowed by playback software like Roon, and HQPlayer.
I would suggest, that at this point in the development of discrete DSD DACs, not doing and publishing measurements for DSD playback is only telling half of the story, indeed it is not telling much of the story at all for those users whose intent is to only send the DAC DSD data. Please, please consider doing measurements for DSD playback in the future.

MhtLion's picture

I agree 100%.

Bogolu Haranath's picture

In the measurements section, Bezier IIR filter impulse response, Fig.3, looks like short minimum phase filter ...... However JA1's description of this filter sounds like a hybrid type of filter ....... JI preferred this same type of filter when reviewing the T+A DAC8 ...... JA2 also preferred this same short minimum phase filter :-) .......

Ortofan's picture

... aside from being able to play SA-CD discs, what can this device do that a Marantz ND8006, for example, cannot?
https://www.us.marantz.com/en-US/shop/hificomponents/nd8006

Bogolu Haranath's picture

It has a choice of 4 different digital reconstruction filters ...... A better comparison would be the new $5,500, Mark Levinson 5101 ....... That ML SACD/CD player/DAC probably also has a choice of 7 different digital reconstruction filters, like the ML 5805 integrated amp/DAC, reviewed by Stereophile :-) ......

Guileshill's picture

Earlier comment withdrawn. Confusion over the product enumeration.

doxycc's picture

There is considerable value in all-in-one units - shelf space, fewer cables (and lower cost), sonic point of view of manufacturer. There are similar downsides - lack of upgradeability depending on architecture (modular upgrades might be available but for how long after introduction of unit). Residual value of brand especially on the digital side (audiogon can give you a sense of the 50-80% discount on 3 year old digital even among the most well known brand). Some non-US brands have stable distribution and dealer networks others change frequently. At this price, good to have a checklist and determine risk/reward as s consideration separate from the sonic qualities.

Long-time listener's picture

With all this incredible engineering, at $21,000 it still can't provide the 21-bit resolution available from the $2000 NAD M51 or the most recent Benchmark DAC. Why not?

dcolak's picture

It costs 800 USD and offers true 21bit resolution.

https://www.audiosciencereview.com/forum/index.php?threads/topping-d90-b...

Bogolu Haranath's picture

Topping D90 is a DAC ....... Yes, its measurements are excellent ......But, it has no SACD/CD player ..... Also, it has no wi-fi access :-) .........

Bogolu Haranath's picture

BTW ...... It would be nice to see a Stereophile review of Topping D90 DAC :-) .........

Bogolu Haranath's picture

Stereophile has a $400 Class-A listed DAC ....... So, Stereophile can also review another under $1k, probably another Class-A list-able DAC :-) ........

MhtLion's picture

Would the SACD/CD player worths 26.25 times the price? One thing for sure, the sound quality gap between High End DAC vs Budget DAC had shrunken a quite lot. My personal guess is that whoever replaces their $20k DAC with this $800 DAC won't miss too much.

Bogolu Haranath's picture

I'm not justifying the price of this T+A player ....... I was mentioning that, it is not exactly apples to apples comparison between the D90 DAC and this T+A player ....... One could buy a $1,200 Denon CD/SACD player for example ....... The $400 Pro-Ject, Stereophile Class-A rated, DAC/Pre-amp/Headphone amp which I mentioned, is even less expensive than the Topping D90 :-) ......

MhtLion's picture

Agreed 100%. I once tried to purchase T+A DSD 8 DAC. At that time, the only U.S. distributor I could find was an unknown name to me, and they tried to charge a full retail price + shipping + sales tax even though I did not live in the same state. So, I found a European retailer who had a small sale plus VAT saving to sell out of the Eurozone. The final price including shipping was at least 40% less than the U.S. price. But, at the last minute, they realized that they have a contractual agreement not to sell to the U.S. The salesperson said other none Euro countries were fine, but U.S. He also said he sold other brands to U.S. hence sorry for the confusion.

Math is funny. $60/$100 is 40% discount. But, $100/$60 is 66% premium. A 66% premium to buy the exact same product over other markets was a bit too much for me.

Bogolu Haranath's picture

T+A DAC8 was reviewed by Stereophile ....... Stereophile Class-A+, listed ....... That DAC8 has similar DAC and similar 4 digital reconstruction filters as this T+A player ....... DAC8 costs less than $5k :-) ........

barrows's picture

I had the D-90 here for a few weeks, and would suggest, respectfully, that your comments are wrong. While the D-90 is a very, very good DAC, For The Money, it cannot compete with a player such this T+A with its much more developed output stage (I have heard the DAC version of the T+A).
I still would recommend trying the D-90, in your system, to anyone looking for a DAC under around $4K. Especially with DSD 256 playback in "DAC mode".

JRT's picture

JA1's measurements show that the device under review has varied output depending on input selected, and that output well exceeds the 4.0_Vrms output of the Topping D-90. If the devices under comparison are not very closely matched in output level, the device with higher output level is usually at an perceptual advantage in the comparison.

barrows's picture

JRT, if your comment is aimed at me, I am well aware of the need for precise level matched comparisons, especially when compared components close in performance (which many DACs are). I typically level match by measuring output voltage with pink noise signal. I work in audio product development, and have been "professional" in this industry for some 20 years now.
The D-90 is excellent for the money, but DACs with more robust output stages and power supplies of my experience do outperform it.

dcolak's picture

D90 is far above Benchmark DAC3 so I would REALLY love to see your measurements.

Bogolu Haranath's picture

Which measurements of D90 are 'far above' Benchmark DAC3? ....... Just curious :-) .......

MhtLion's picture

The fact we are mentioning $700 DAC and $21,000 DAC in the same sentence indicates the market is going through a big change.

There were two groups of engineers. Some were born rich and some had great careers elsewhere. Some had wealthy wives. They liked the music and equipment. So, they worked hard: 2-4 hours a day to bring a product or two every couple years or so. It was hard work, but never that hard. After all, only a few had an engineering background. Many hired the engineers or better yet outsourced 97.56% of everything they ever produced. The very idea, concept, and principle of their innovation were ALL OUTSOURCED. They sold a few, but the extravaganza prices kept their lives comfortable. Perhaps they did not care for money. It was LOVE that drove them hard at audio venues and showrooms, where they spent the majority of the time.

The others were mostly with an engineering background. Some liked the music, some rarely listened to anything at all. They were after the hard profit. For their lack of love, they made up by working their a*s out, 10-15 hours a day to produce $800 DAC with great specs and OK sound. After all, most customers judge the audio equipment with their eyes anyway - by the aesthetic and specs. Their work ethics or the lack of it were mostly copying others without ever paying. Some discerned audiophiles scoffed, calling those cheap knockoffs.

But, an unexpected, rather a comical trend starts to emerge. After years and years of hard work, while the other was hard drinking. First, $3,000 DAC starts to sound as good as $10,000 DAC. Then, $300 DAC starts to sound as good as $3,000 DAC. Finally, $700 DAC sounds not too bad right next to $20,000 DAC.

Ortofan's picture

... TEAC to be "cheap knockoffs"?

The UD-301 and UD-501 cost no more than $600.

https://store.acousticsounds.com/d/100294/Teac-UD301_PCM__DSD_USB_DAC-DA_Converter_or_Processor

https://store.acousticsounds.com/d/88975/Teac-UD-501_PCM__DSD_USB_DAC-DA_Converter_or_Processor

http://archimago.blogspot.com/2013/05/initial-impressions-teac-ud-501-usb-dac.html

https://archimago.blogspot.com/2013/05/measurements-teac-ud-501-pcm-performance.html

Allen Fant's picture

An excellent review- JA2
The musical selections, albeit short, are on point.

Anton's picture

But where did the MQA go?

Bogolu Haranath's picture

MQA sems to be DOA :-) .......

Bogolu Haranath's picture

Or ..... MQA went into the 'Dead man's Chest' :-) .........

X