Thanks to shadowlight, I was able to activate a pre-launch demo account on Qobuz, a new streaming service that is hoping to compete with Tidal. While there are minor difference between the two the most significant difference it that while Tidal’s highest fidelity option involves the streaming of MQA encoded files, Qobuz will be streaming native high res files, up to 24/192. While US pricing has not been announced yet, if you use the European pricing as a rough guideline, Qobuz will cost a little more than what I not pay for Tidal, but not by that much.
While Qobuz will stream files at up to 24/192, the optical receiver on my 1st generation EE DAC will only sync up to 24/96, so I set Qobuz to limit me to that speed. But that in fact turns out to be a good thing for right now, because my Bluesound Node only does MQA software unpacking, which the MQA literature says delivers a sound that is similar to what one would expect from a 24/96 file.
I set up a playlist on the Node that included copies of the same songs both in MQA from Tidal and 24/96 from Qobuz. Since both sources will be played through the same equipment using the same Bluesound software I have pretty much isolated the comparison down to a single variable, the MQA encoded 24/96 vs the native 24/96 from Qobuz. (The Bluesound outputs the unpacked MQA file to my DAC as a 24/96 bitstream so even that is the same)
I listened to comparisons of a number of cuts and some clear patterns emerged. Keep in mind that the baseline here was pretty darn good sound to begin with, so nothing sounded anywhere near bad, but as audiophile we like to split hares and sweat the details , so that is what I am going to talk about now.
First up, the title track from Lee Morgan’s Candy album.
Three things jumped out at me as I did this first comparison. When compared to the native file, the MQA had an overall darkness in the soundstage. The naive file sounded much more open and alive and gave a more realistic impression of listening in a real acoustic space. Morgan’s trumpet also had a sharper attack on the native file, much closer to what you hear when you are in the room with an actual trumpet player. And lastly, the native file also gave you a clearer picture of the brass of Morgan’s trumpet resonating as a separate and distinct sound from the fundamental note produced by the air being blown through it. On the MQA version the note and the resonance was more of a homogenous single sound. This was not a matter of tonality, which was pretty much consistent across the two formats, but rather of detail.
Next was ZZ Top founder Billy F Gibbons’ new album The Big Bad Blues.
Even though it is not what I would consider an audiophile recording, I thought it would be a good thing to include because it was only released a few days before my listening test. I am going to make a leap of faith here, but seeing as how new it is I felt it is a pretty safe to assume that both versions were sourced for the same mastering and which would take possible differences in the source used to make the two files off the table (on would hope). Overall the MQA was just little less defined that the native files. Instruments had a definite space in the soundstage, but it was more of an area than a spot. With the native file the sound stage was a little bit wider and had a better sense of depth but more importantly the instruments occupied and more cleanly defined pace than in the MQA version. Instead of something being over towards the right is was clearly specific distance inside the edge of the right speaker. Lastly, the guitar on the MA was a little mushy when compared to the sound of the Qobuz native file. Interestingly the Qobuzz Hi Res file of this album is only 24/44.
On the Wish You Were Here Symphonic album I listened to the first cut with Alice Cooper doing the vocals on the title track.
Here my notes on the MQA version said clear and spacious. Switching to the native file there were subtle differences, but they were much less in magnitude that in the first two comparisons. He native file had a little better sense of space and a little more wood body sound than the MQA , but again , not much more. The biggest difference was again in speed. In the instrumental introduction to the song there is a section series of piano notes are interspersed among sets of chords. On the Qobuz file (which is only 24/48) the piano notes have not only a faster rise time, but also a much faster fall time which results sonically in your being able to hear the note fade all the way to silence before the next one starts up, as opposed to going half way down and running into the next note. I my mind I liken this to using the sustain pedal to hold a note verses the note being cut off clean when the pedal is not engaged.
From Elton John’s Goodbye Yellow Brick Road I compared the two versions Funeral for a Friend/Love Lies Bleeding
(For this, and all of the remaining comparisons, Qobuz was back up to 24/96.) Again, the MQA was no slouch, but when I switched to the native file the soundstage got a bit wider and deeper. But more importantly it became significantly more defined. Instruments took ownership of the space they were in. In addition to hearing the exact spot they were in, you could also get a much greater sense of the silence in areas where the instruments were not located. As with earlier comparisons, the rise and fall times were faster resulting in sharper transient response. One other thing that I noticed was that the vibrato on one of the electronic keyboards had a much cleared and distinct sound as opposed to being simply a fuzz riding along with the fundamental note. This is not unlike what I heard on Lee Morgan’s horn but it is the result of an electronic overlay instead of two different physical surfaces vibrating. But I guess in terms of the musical signal details are details regardless of how they are generated. Interestingly the instruments in the native file sounded smaller in size and in the MQA version. I am not sure which is right, but since this is a complete studio creation I’m not sure if there is really anything that is right, but is was an interesting observation none the less.
Time to move on to a full symphonic performance. I had a tough time tracking down something that was in both MQA and Qobuz Hi Res, but with a bit of searching I came across the Mercury Living Presence Stereo performance of The Nutcracker, led by Antal Dorati.
The MQA version of this recording had a soundstage that was strictly limited in width by the centerlines of each speaker. When I switched to the native version the stage immediately widened to two feet beyond the outside edges of the speakers, quite a difference. The stage also opened up and gave the listener a much greater sense of the hall ambiance. As with other recordings the native file provided a greater level of detail on each of the instruments.
The next thing I listened to was another new release, this time a duet album by Tony Bennett and Diana Krall.
I listened to the title track, Love is Here to Stay. As with previous comparisons on the native file I heard a more open room sound. There was also better decay and clearer vibrato on the vocals, better edge definition on the piano and more shimmer on the cymbals and snare drum, all indicative of faster speed and better detail.
Lastly I pulled up the I’m an Old Cow Hand track from Sonny Rollins’ Way Out West album.
Here the differences were minute, if any. But it is worth noting that on this album each of the instruments is either full right or full left, so there is basically no soundstage to speak of in the original recording. I confirmed this by putting on my Analogue Productions LP of the same album. No stage at all, just 100% ping pong stereo.. Not that this makes it a bad album, I still love it. It is just not a good tool d=for doing sonic evaluations. (Wow a recording that is only good for enjoying the music it contains…. What a concept.)
Once this was all done and I looked back at the results, some very distinct patterns started to emerge. This led me to wonder if this was the result of some kind of “house sound” differences between the two companies. So I pulled up CD quality versions of the Tony Bennet/Diana Krall cut from both systems. One would think that they would simply load the CDs onto a server and they should sound identical when played through the same equipment, right? Well think again. The same type of differences that I heard between the two hi res formats existed between the redbook versions, although the gaps were not nearly as wide. So much for bits are bits, huh? I don’t know if it a function of the digital equipment and/or software that they use or if one or both of them are doing something to “improve” the sound before saving the files on the server, but at least one if not both of them are displaying a definite house sound flavoring. For the record, both companies’ hi res versions sounded better than their own CD quality versions.
When I get some time I think I am going to try to cross compare the CD version from one service with the hi res from the other and see if this sheds any further light on anything. If I come to any interesting conclusions on that front I will share them here.
As far as the rest of Qobuz goes I still have not had enough time with it to form any solid conclusions. First impressions, Qobuz seems to have a bigger library of the type of music I looks for (vintage Jazz, classical and classic rock) and opposed to Tidal which is pretty much rap/hip hop centric (which should be no big surprise to anyone considering who owns it.) The software is not the friendliest, but it has still not been officially released in the US, so I am sure all of the interfaces are still a work in progress. More specifically, the Bluesound interface for Qobuz definitely needs some cleaning up, for example there is currently not way to tag an album as one of your favorites within the Bluesound software. The clickbox is there, but it does not do anything. Currently I have to go out the the Qobuz app to tag favorites and the return to Bluesound to play them. Kind of a PITA, but considering the service is still not officially launched in the US I am sure the software is a work in progress and these things will be worked out.
Let’s keep in mind that this is just a single comparison and only reflects how things sound in my house on my equipment. Of course YMMV. And it would also be interesting to compare the Qubuz 192/24 to full hardware decoded MQA to get a sense of what each can do at its best. Unfortunately, I do not currently have the hardware to do that kind of comparison. But it is really the first time I have been able to do anything that even came close to isolating MQA as a single variable. Even so we only have enough information to consider these results anecdotal, at best. While it is not enough to come to any conclusive results on how good MQA is in comparison to other alternatives, it is a good first step.
More to follow………