NAD Masters Series M10 streaming integrated amplifier Trying Dirac Live Room Correction

Sidebar 1: Trying Dirac Live Room Correction

To use Dirac Live LE with the M10, I plugged the microphone supplied with the amplifier into the 3.5mm jack on the USB adaptor, also supplied, and plugged that into the USB Type A port on the M10’s rear panel. I installed the Dirac Live app on my iPad mini, after first making sure that it and the M10 were connected to the same network. When I ran the app with the M10 driving my KEF LS50s, it found and identified the amplifier, reduced the playback volume, and performed a level check. I then followed the on-screen instructions, placing the microphone in each of the nine positions specified by the app and performing a “chirp” test at each. After the last test, Dirac calculated a correction filter and asked for it to be named and saved to the M10’s internal DSP. Imaginative as always, I called the correction filter “LS50.”

When I then ran the BluOS app on the iPad, the Audio Settings menu now included an on/off switch for “LS50.” The target response for the filter is shown as the dark blue trace in fig.1 and can be compared with the before and after responses for each speaker (respectively thin and bold green and red traces).

1219naddirac.Diracfig1

Fig.1 Target response, 50Hz–500Hz, calculated by Dirac Live for the KEF LS50s (dark blue), with responses in JA’s listening room of left loudspeaker (corrected, bold green, and uncorrected, thin green) and right speaker (corrected, bold red, and uncorrected, thin red).

I was listening to J.S. Bach’s “Wachet Auf,” from a favorite album, Yo-Yo Ma and the Amsterdam Baroque Orchestra’s Simply Baroque II (16/44.1 ALAC files ripped from CD, Sony Classical 60681) when, using the BluOS app, I turned on the low-frequency correction Dirac had calculated for the KEF LS50s. The low frequencies became richer and the orchestral sound acquired greater bloom. This was definitely a step in the right direction.

I continued listening to recordings I knew well, both classical and rock. Well-recorded classical piano fared particularly well with the small KEFs corrected by the M10’s Dirac Live. The weight added to the left-hand register of Evelina Vorontsova’s instrument on her hauntingly beautiful reading of Rachmaninoff’s second Piano Sonata (16/44.1 ALAC file ripped from CD, STH Quality Classics CD 1416092) was addictive. And that weight was not accompanied by any blurring or boom. However, when I listened to the Simon Rattle Beethoven symphony I had used for my comparisons with the Luxman amplifier, I couldn’t escape the impression that there was too much upper bass with the Dirac EQ.

Accordingly, I measured the LS50s’ spatially averaged in-room response with the speakers driven by the NAD M10 with and without the Dirac-calculated correction. (I used the same methodology I have been using for 30 years—averaging 20 spectra, taken for the left and right speakers individually using a 96kHz sample rate, in a vertical rectangular grid 36″ wide by 18″ high and centered on the positions of my ears. The only change from past practice was to show 1/12 octave–smoothed spectra rather than 1/6 octave, in order to reveal more detail.)

The results are shown in fig.2. (Ignore for now the green trace.) The blue trace is the in-room response of the uncorrected speakers. It slopes down below 200Hz and above 2kHz. The former is due to the fact that the speakers are positioned well away from the room’s sidewalls and front wall and don’t, therefore, benefit from any low-frequency boundary reinforcement. The latter is due to the room’s increasing absorption at high frequencies—the KEF’s highs actually sound neutrally balanced at the listening position.

1219naddirac.Diracfig2

Fig.2 KEF LS50, spatially averaged, 1/12-octave response in JA’s listening room with the NAD M10’s original Dirac correction (red), with the revised correction (green), and without correction (blue).

The red trace in fig.2 shows the LS50s’ spatially averaged in-room response with Dirac Live correction applied between 50Hz and 500Hz. You can see that the corrective filter calculated by Dirac has flattened the balance in the lower midrange but has overcompensated for the KEFs’ shelved-down low frequencies. There is now an apparent excess of energy of 2–5dB from 40Hz to 150Hz, correlating nicely with my perception. Note that the measured responses roll off rapidly above 20kHz. This is due to my feeding the diagnostic test signal to the M10’s analog inputs, which digitize it with a 44.1kHz sample rate. Repeating a response measurement with digital data sampled at 96kHz and 192kHz fed to the M10’s coaxial S/PDIF input indicated that the Dirac filter did function identically at the higher sample rates.

Though both are centered on the position of my head in my chair, the grid spacing and layout I use for my traditional measurements and those used by Dirac Live are slightly different. (This shouldn’t make a big difference because once the number of positions where the responses are measured reaches nine, the spatially averaged response converges on a consistent result.) I also use a high-end Earthworks omnidirectional microphone for my measurements, though I doubt that would account for the differences between the red and blue traces in fig.2. Nevertheless, I plugged the NAD mike and USB adaptor into my laptop, where it was identi fied as “NAD USB Audio” and could be set to sample rates of 44.1kHz or 48kHz with a bit depth of 16. I did some tests comparing the NAD mike with my Earthworks mike, and the two had identical responses below 1kHz.

More experimentation was necessary. I turned off the original LS50 filter and ran the Dirac Live app again. However, this time, when I had finished the measurements and the app had calculated the correction filter, instead of accepting what I was shown, I used the Control buttons—the solid circles superimposed on the target trace in fig.1—to reduce the intended levels below 200Hz by 2dB or so (fig.3).

1219naddirac.Diracfig3

Fig.3 Modified target response, 50Hz–500Hz, calculated by Dirac Live for the KEF LS50s (dark blue), with responses in JA’s listening room of left loudspeaker (corrected, bold green, and uncorrected, thin green) and right speaker (corrected, bold red, and uncorrected, thin red).

The new spatially averaged room response is shown as the green curve in fig.2. Now when I listened to the Simon Rattle Beethoven recording, the mid- and upper-bass regions were in better balance with the midrange. And there were sufficient lows for the Rachmaninoff piano sonata recording still to sound magnificent. The subtle organ bass pedal line in the “Kyrie Eleison” from the sublime Robert Shaw/Atlanta Symphony Orchestra recording of the Duruflé Requiem (16/44.1 ALAC file ripped from CD, Telarc CD-80135) was more evenly balanced with the Dirac EQ than it was without. The extra bass energy didn’t seem to stress the KEFs, and though I haven’t shown them, the responses for the left and right speakers at the positions of my ears now matched very closely below 500Hz.

To sum up, NAD’s incorporation of Dirac Live in the M10 is a huge added value, especially if you have a small-to-medium–sized room and minimonitors that you like or need to place a ways away from the room boundaries. As much as I love the unequalized LS50s, their presentation corrected with Dirac Live was addictive.—John Atkinson

NEXT: Specifications »