I finally received my replacement Explorer 2 from Meridian of America. It was pre-loaded with the latest firmware so no more contending with the updating rigmarole. I’ve been listening to – what else – the test tracks from the 2L Hi-res Test Bench. It’s the only set of test tracks that have both the hi-res PCM version along with the MQA encoded version. Plus, they’re free.
The more I read about MQA, the more I started to become interested in the format. It seemed like quite the technical achievement – to somehow encode a higher bitrate signal into the noise spectra of a lower bitrate signal while preserving fidelity in both instances. What’s not to like?
For this mini-review I listened to all tracks through three separate systems to get a good sense of how MQA sounded across different types of systems and equipment. I first listened to the tracks through the NAD VISO HP50’s through the Explorer 2’s headphone amp. Then I hooked up the Explorer 2’s DAC output to my main headphone-based system which consists of the Audeze LCD-XC and DNA Sonett 2. Finally, I connected the Explorer 2 to my speaker system – an Ayre AX-5 Twenty paired with Vandersteen 3A Signatures.
For each comparison, I listened to the MQA track immediately followed by the hi-res PCM file. Software used was Windows 7 / foobar2000 with output set to ASIO and Linux / aplay with ALSA set to direct hardware control of the Explorer 2. The blue LED lit up on the Explorer 2 whenever MQA material was fed to it.
The first track I listened to was “Et misericordia” from the album MAGNIFICAT. Taken from a DXD master, this is a beautiful, reflective choral piece with piano and string accompaniment. It took a few repeats of both tracks to get a good initial understanding of how the MQA encoded file differed from the downsampled 192 kHz / 24-bit PCM file. Chief among these differences is that the MQA track sounded more vibrant and more detailed. This is the kind of presentation I feel that most audiophiles associate with good sound. Yet switching in the PCM track, it became apparent that this came at the price of midrange tonal saturation. Vocals and strings sounded a touch more forward and brittle. Imaging was a tie, though it seemed MQA offered a larger degree of instrument separation. However, it came across as a little artificial due to the MQA track’s tonal balance.
Next up was “Boisterous Bouree” from Britten’s Simple Symphony No. 4. The main difference between the PCM version and the MQA track here was in terms of bass performance. Though the MQA track was a fine rendition, it lacked the PCM version’s rhythmic drive from the contrabass. Strings sounded a tad sweeter over PCM with more air from the acoustic. To my ears, it seemed that Meridian’s encoding scheme was shelving down the bottom octaves and some of the midrange. Was this the ‘de-blurring’ effect that others have noted? I don’t know, though I can see how lower output in these registers can lend one to believe that the sound is clearer. However, this has the unfortunate effect of sucking some of the soul out of the music.
The final two files were taken from a jazz album, Quiet Winter Night. The instruments in “Blagutten” were placed in a spacious acoustic. On the MQA version, the trumpet came across with an icy clarity that bordered on being too aggressive in certain passages. The PCM version softened this somewhat by presenting a fuller tone to the brass. It also served up piano notes with more body. Drums had kick, and the bass sounded more powerful through the PCM version. Both versions imaged well, so no major differences there. Overall, the track was simply less musically satisfying through the MQA encoding process. The MQA version lacked warmth and it never really engaged my emotions.
Given these limited impressions on the Explorer 2, all I can say is that MQA on Meridian’s portable DAC / headphone amp leaves a lot to be desired. Now I need to caveat this with the fact that the Explorer 2 can’t “unfold” the content all the way to the native DXD rate of the master. Perhaps if it could play back the content at 352.8 kHz / 24-bit then all these differences would disappear. At this point, I feel MQA certainly isn’t a silver bullet to the problem of storing hi-res content, nor can I be sure that it isn’t somehow modifying the actual content of the original file during the encoding process since the toolchain is proprietary. In the end, I hope that MQA clicks with enough people that it can be iterated on and improved as it shows some promise.