The QB-9^DSD replaced the DAC that I use on a regular basis (Teac Esoteric D70/dCS Puccini U-Clock) and was used to drive the rest of my usual system (Melos Plus Series Line, Parasound HCA3500, ATC SCM-50PSL). The media player used was the last version of foobar2000 installed on a Win7 netbook. Driver installation was no problem and anyone capable enough to follow the simple instructions given by the Ayre (though uploaded only to their website and not included in the manual, which offers only some basic info), will not encounter any problems even with the DSD setup.
Given that this device is an upgraded version of the QB-9 tested some time ago, the first question that would be good to be answered is whether it offers any real noticeable difference. Since an “old” version of the QB-9 was not available for a direct comparison, I turned to the recorded listening sessions for both devices and performed a blind test. The result was surprising: The differences were rather clear (given the price range and the fact that the first version of the QB-9 was already a top-class DAC) and the result of the blind test showed that I could differentiate between the two machines (at the level of the recorded files, at least). For those who like the details, the blind test was performed with the ABCHR software set for mid-level differences and the result was 7/8 successful identifications, corresponding to a Type I error probability equal to 0.035. Based on this result, Ayre's claims of significant differences between the original and the upgraded QB-9 versions appear, in fact, reasonable (note, however, that this is not a formal ABX blind test in the strict, academic, sense since there was just one listener, a limited set of tracks and it was not double blind. Therefore from a statistical point of view one is not entitled to project the result to the general population, in the mathematical sense. Nevertheless it was a quite enlightening experiment). The main differences identified in this phase were a better sound stage and a much better feeling of balance characterizing the QB-9^DSD.
The next step was to evaluate the effect of the two different filters that are available. The measurements showed that the differences are small and a comparative listening session demonstrated that this is, indeed, true. With some effort and through the use of appropriate music program, QB-9^DSD seemed to offer a more clear stereo image (losing slightly some of its width and depth) and sounded a little faster and dry in the high frequency part of the spectrum by using the Measure mode. On the other hand, using the Listen mode, it gained a bit more towards the ambiance feeling and, also, sounded somewhat more mild and pleasant. During the review of the previous version I used mainly the Listen mode (the mode proposed by Ayre, after all) but this time I decided to listen to the QB-9^DSDin Measure mode for a much longer period of time since the result was more close to what I'm used to from the reference converter. Anyway, the differences are very small (please refer to the listening sessions recording at the end of this page) and the choice is more a matter of personal taste and less of objective correctness. It's a pity the switch for the filter selection is difficult to access, at the back panel …

The first impression one gets from the QB-9^DSD is that of a highly transparent and balanced converter, one that gives a powerful sense of wide bandwidth and offers very good attributes in terms of dynamic range (probably because of its low noise levels). Sound stage was very clear with good depth and width and the device seamlessly managed the details of the recording and appearedmore than enough capable to focus individual sound sources in space. With the QB-9^DSD in place, the system retained its full potential to describe complex scenes with large orchestras offering very good sense of the movement and of the air surrounding the instrument groups while smaller ensembles were described with great realism and presence.
Very low frequencies appeared well controlled and nimble with good description of the attack part and clearly contributed to an impressive, yet balanced, sound character, without creating any sense of excess. Those attributes are maintained towards the low and the higher low part of the spectrum offering a good sense of scale (especially in large acoustic instruments such as the double bass) the result being extremely comfortable to listen to. Rhythm section was described with very good control, quick and massive as it should and the result was truly impressive.
Rendering of the mid part of the spectrum was characterized by great detail and a sense of openness and flow. Listening proved to be very relaxed without any feeling of pressure or fatigue, even at high levels. To this end, it surely helped that there was a slight displacement of the image behind the horizontal axis defined by the loudspeakers, exactly as needed to create both a feeling of space and offer the listener a good sense of the mix details.
QB-9^DSD proved quite a performer at the top part of the spectrum, offering a feeling of very good extension and being fast and transparent. In “Measure” mode, it seemed to be a bit hasty during the release parts of the time envelope, rigorous and slightly dry, but with more energy, a character that easily could be changed towards a more mild, laid-back and somewhat rounded one, when the “Listen” mode was chosen. Harmonics were well balanced between richness and precision and the end result offered a satisfactory feeling of brightness.
Listening to high resolution files lead to even better results. There is no doubt that the QB-9^DSD can easily fulfill the requirements set by a genuine oversampled recording and supply the rest of the system the details and the feeling of comfort, especially at high frequencies, that high resolution material offers. The soundstage became even more realistic with a strong “been there” feeling and the overall result was more relaxed and pleasant, both for PCM and DSD files.

Conclusion

Skipping various commonplace phrases like "the good got better" that rightly come to mind, it's worth staying to the perceived facts: And these are that, given its price tag, the QB-9^DSD is a dramatically improved version of the QB-9 and one that is worth the cost of the upgrade. The new device has better behavior at critical points, offers DSD compatibility and retains all the positive aspects of the previous version (such as the two digital filters and the output stage architecture). Those who already have a QB-9 are advised to plan on upgrading the earliest, while those looking for a high performance USB DAC, can rejoice: The QB-9^DSD will rise above their expectations!

Listening Sessions Recording

The following recordings were made with a DV-RA1000 Tascam master recorder (using 24bit/192kHz sample rate) and you can download them to have an impression of what the device under review sounded like. It is obvious that any recording of this kind could not be absolutely transparent but, according to our experience, the majority of sound attributes we listen to, during the actual listening sessions, are preserved. No need to say that you should use these samples cautiously and for informative purposes only. Do not rely on them exclusively to make any buying decisions. The file format is .wav, so expect that the zipped files will be quite large (even if the clips are about a minute or less long. You could use an ABX listening tool like the ABX plug-in for foobar2000 if you want to make some more elaborate experiments. Contact us if you have any questions.

Listening Sessions Recording TOC
File #01	Ayre QB-9DSD,16bit/44.1kHz, Measure Mode
File #02	Ayre QB-9DSD,16bit/44.1kHz, Listen Mode
File #03	Ayre QB-9DSD,24bit/96kHz, Measure Mode
File #04	Ayre QB-9DSD,DSD 64x, Measure Mode
File #05	Reference DAC (dCS Puccini u-Clock/Teac Esoteric D-70

Previous | Next