The emerging awareness of High Resolution Audio and its commercialization are headed for rough waters if the providers of “so-called” high-res audio downloads don’t start coming clean about what they’re offering. I’m seriously worried that the continued confusion regarding what is and what isn’t a high-resolution audio track will result in the death of what could be a revolutionary move away from the same old thing. Why am I bringing this up again? Because I joined and downloaded the sample high-resolution audio files from the French site QoBuz.com…and found more of the same hype and inaccuracies that have dominated most domestic sites.
QoBuz.com seems to be a big player in the E.U. Their site is extensive, offers content from the major labels, has “high-resolution streaming”, a high-end audio magazine and lots of user choices for each track. They will sell you a low resolution MP3 file up to “Studio Quality Masters” at 192 kHz/24-bits! You identify your OS and they recommend which format you should download. The choices are quite extensive as you can see from the list below:
Figure 1 – The pull down menus on QoBuz.com showing the various OS, music player and format options available from their site.
I signed up and established an account yesterday. My Downloads folder was populated with a few promotional tracks, so I downloaded the tracks that are described as “2 pistes pour tester les possibilités de votre appareil Hi-FI (24 bits / 192 kHz et 24 bits / 176 kHz) Interprètes Divers”. In English, they’re saying that I can download 2 tracks that will show the possibilities of my equipment…that their files are 24-bits/192 kHz or 24-bits at 176 (really 176.4 kHz). I downloaded their Downloader (which is very nice and seems solid) and then proceeded to download the two “high-resolution audio” tracks to evaluate “different sampling frequencies”.
If you were running a high-resolution audio download site, wouldn’t you want the first experience…the one that you provide free…to be of the highest quality and unimpeachable with regard to fidelity and specifications? I can’t imagine what the people at QoBuz.com are thinking when they give everyone who signs up for an account two files that are so obviously not high-resolution!
Here’s the spectragram of the first file. The title of the piece is “Mere de Dieu Vierge” and it is a performance by an a cappella vocal ensemble. Human voices don’t put out a large amount of high frequency sound so I wasn’t expecting too much with this track. What I discovered was a track that had been rolled off at 23 kHz! This was advertised as a 192 kHz/24-bit track and is intended to used to evaluate very high sampling rate material. Take a look:
Figure 2 – A spectragram of a “192 kHz/24-bit” demonstration file from QoBuz.com [Click to enlarge].
The fact that this plot goes to absolute zero means that they used a low pass filter to exclude any frequencies that might have been in the recording space beyond standard CD resolution. I just can’t figure it.
The second track was recorded at 176.4 kHz. It’s a jazz trio featuring an acoustic bass. We would expect it to contain a reasonable amount of high frequency material due to the drum set. This time there’s no roll off…although the track would probably benefit from a LPF because there’s a few oscillations in the upper spectra and a large amount of noise in the highest frequencies. My guess is that this track was recorded using analog tape and that there was some ultrasonic noise present in the studio. Here’s the spectragram.
Figure 3 – A spectragram of a “176.4 kHz/24-bit” demonstration file from QoBuz.com [Click to enlarge].
It’s pretty clear that the French site has some misunderstandings regarding high-resolution audio. Unfortunately, the amount of misinformation being put forth about high-resolution audio is pushing what could be a promising new development closer to the cliff. I’m discouraged.
The new iTrax is in progress. I’m going to include spectragrams on all albums and try to be clear about the provenance…we’ll see if it makes a difference.