Naxos 'dumbs-down production standards'
'To those who see nothing to worry about in the cheapness of recordings, I'd like to point out that without a sensible stream of revenue the whole process of making recordings will be compromised - if indeed it is not already.
Naxos, for one, pays very, very little to its freelance production staff. I can say from personal experience, that not only does the current economic climate make it almost impossible to earn a living wage as an independent classical music engineer or producer, it also has the inevitable knock-on effect of dumbing-down production standards: it's now almost impossible to invest in new equipment. Quite obviously this impedes qualitative improvements for which classical recording engineers are traditionally renowned.
But who cares about quality anymore? Naxos among others blithely and inaccurately claim that their audio streams offer 'Near CD quality', which at 128k/sec is little more than 10% of CDs 1141Kb/sec. Anyhow, you won't need those B&W 803's for much longer - a couple of Tandy ear buds will do just as well.'
This comment was added to my article Is classical music too cheap? but I thought it worth running as a new post. It was written by Guthry Trojan who is a freelance classical and jazz recording engineer based in Paris. (Check out his new blog The Crunch). These are Guthry's personal views, and he raises some interesting questions.
We have an awful lot to thank Naxos for and this blog regularly recommends their recordings, but will low budget recordings lead to a long term erosion of technical standards as Guthry suggests? Just one example from my CD collection is the intrusive mains hum during the first dance in Set 2 (Op 33) of the otherwise excellent Naxos recording of Malcolm Arnold's English Dances, (catalogue number 8553526, track 5). The audible hum which lasts for the duration of the 3 minute 14 second track must surely have been noticed when the disc was being edited. Would a fist full of split notes from the brass have been allowed onto the final release?
Are the bit rates in streamed audio really adequate for serious listening? If the soloist is playing a 1697 Stradivarius what is the point in listening at 'low-fi' quality? Why develop new hi-res formats like SACD (which Naxos are backing) when audio streams via the internet are low-res? And what about music over digital radio (DAB)? How many stations actually broadcast at the optimum 192kb/s, and how many users follow the BBC and reduce bit-rates as low as 80kb/s to compress more channels into the same bandwidth?
Isn't it rather short-sighted to agonise over subtle differences in performers' interpretations when, as the table below shows, there are orders of magnitude quality variations in the replay chain? If the CD is dead, as so many pundits are telling us, did high quality sound die with it? In the brave new world of the internet has quality been traded for accessibility?
More questions than answers I'm afraid - 'For the times they are a-changin'
Sample rates for the competing audio formats:
MP3 (x-High) = 192Kbps
MP3 (High) = 128Kbps
MP3 (Mid) = 64Kbps
DAB = 192Kbps
CD = 1413Kbps
SACD = 2.8 Mbps
DVD Audio, varies from 192 KHz for two channels to 96/48 KHz for five channel surround sound
I know that sample rate does not have a linear relationship to sound quality, but the relationship cannot be ignored. And interestingly by far the best performer using this measurement is a vinyl LP - as the signal remains in the analogue domain the effective sampling rate is infinite.
Comments welcome on this fascinating, and controversial, topic. But they must be attributed - no anonymous comments please, and no 'Jerry Springers' either thanks.
Image credit - Microphone from Coutant.org, my trusty Thorens TD125 from Decibelhifi.com
Image owners - if you do not want your picture used in this article please contact me and it will be removed. Report broken links, missing images and other errors to - overgrownpath at hotmail dot co dot uk
If you enjoyed this post take An Overgrown Path to Music-like-water and Naxos blog and Paying the piper.
Naxos, for one, pays very, very little to its freelance production staff. I can say from personal experience, that not only does the current economic climate make it almost impossible to earn a living wage as an independent classical music engineer or producer, it also has the inevitable knock-on effect of dumbing-down production standards: it's now almost impossible to invest in new equipment. Quite obviously this impedes qualitative improvements for which classical recording engineers are traditionally renowned.But who cares about quality anymore? Naxos among others blithely and inaccurately claim that their audio streams offer 'Near CD quality', which at 128k/sec is little more than 10% of CDs 1141Kb/sec. Anyhow, you won't need those B&W 803's for much longer - a couple of Tandy ear buds will do just as well.'
This comment was added to my article Is classical music too cheap? but I thought it worth running as a new post. It was written by Guthry Trojan who is a freelance classical and jazz recording engineer based in Paris. (Check out his new blog The Crunch). These are Guthry's personal views, and he raises some interesting questions.
We have an awful lot to thank Naxos for and this blog regularly recommends their recordings, but will low budget recordings lead to a long term erosion of technical standards as Guthry suggests? Just one example from my CD collection is the intrusive mains hum during the first dance in Set 2 (Op 33) of the otherwise excellent Naxos recording of Malcolm Arnold's English Dances, (catalogue number 8553526, track 5). The audible hum which lasts for the duration of the 3 minute 14 second track must surely have been noticed when the disc was being edited. Would a fist full of split notes from the brass have been allowed onto the final release?
Are the bit rates in streamed audio really adequate for serious listening? If the soloist is playing a 1697 Stradivarius what is the point in listening at 'low-fi' quality? Why develop new hi-res formats like SACD (which Naxos are backing) when audio streams via the internet are low-res? And what about music over digital radio (DAB)? How many stations actually broadcast at the optimum 192kb/s, and how many users follow the BBC and reduce bit-rates as low as 80kb/s to compress more channels into the same bandwidth?
Isn't it rather short-sighted to agonise over subtle differences in performers' interpretations when, as the table below shows, there are orders of magnitude quality variations in the replay chain? If the CD is dead, as so many pundits are telling us, did high quality sound die with it? In the brave new world of the internet has quality been traded for accessibility?
More questions than answers I'm afraid - 'For the times they are a-changin'
Sample rates for the competing audio formats:
MP3 (x-High) = 192KbpsMP3 (High) = 128Kbps
MP3 (Mid) = 64Kbps
DAB = 192Kbps
CD = 1413Kbps
SACD = 2.8 Mbps
DVD Audio, varies from 192 KHz for two channels to 96/48 KHz for five channel surround sound
I know that sample rate does not have a linear relationship to sound quality, but the relationship cannot be ignored. And interestingly by far the best performer using this measurement is a vinyl LP - as the signal remains in the analogue domain the effective sampling rate is infinite.
Comments welcome on this fascinating, and controversial, topic. But they must be attributed - no anonymous comments please, and no 'Jerry Springers' either thanks.
Image credit - Microphone from Coutant.org, my trusty Thorens TD125 from Decibelhifi.com
Image owners - if you do not want your picture used in this article please contact me and it will be removed. Report broken links, missing images and other errors to - overgrownpath at hotmail dot co dot uk
If you enjoyed this post take An Overgrown Path to Music-like-water and Naxos blog and Paying the piper.
Comments
They are running it on their MP3 and related news feeds with the photo of the Thorens TD125 record deck displayed against the headline. Check this link to see what it looks like.
And I can't imaging people with high fidelity speakers listening to a 128KB quartet at all. Let alone organ music.
As we've now been warned for CD production problems, I think they should be spelled out loud and clear in the review. The market will do the rest, i.e. I certainly will never buy a CD with a production issue, how good it otherwise might be.
I posted some thoughts on this at my site. While the first-hand accounts of Naxos' engineering budgets are interesting, I don't know if there is a causal link between following the current state of the industry w.r.t. internet streaming and making poor recordings.
best,
wb
It's an insult to the downloaders.
Making white noise available for free downloads, what a coup.
That said, I have to point out an inaccuracy. You say, Pliable:
"... in the analogue domain the effective sampling rate is infinite."
That's not true. The closest analogy we can make to anything like a sampling rate in the analog domain would be the grain of the vinyl record. This varies widely from one release to another, with audiophile/virgin vinyl/180 gram LPs at the extreme quality end of the spectrum and recycled vinyl grot often heard on punk 7"s and flaky acetate on dub plates on the grimey end.
Vinyl's "sampling rate" is infinite because it exists in our world, not the computers, and there is no real sampling rate. But the quality of a vinyl recording is just as variable, if not moreso, than the quality of digital ones. It's just less quantifiable. Even with the highest quality vinyl releases, we can measure surface noise that is quantifiably more obtrusive than the -96 dB noise floor of a 44.1 compact disc. This surface noise comes from the molecular structure of the vinyl material -- microvibrations triggered by the grain within the groove. A similar thing happens with silver halide photographs. When you blow them up big enough, you can see little particles of toner that form the smallest constituents of the image. The primary difference is that in analog, real-world media the grain exists in a random pattern of randomly-sized quanta, whereas in digital media the quanta are regularly-sized and aligned in a grid. (This equally true for visual or aural media.)
I'm a fan of vinyl because I'm a fan of reified objects and the philosophical baggage they carry. I adore the way the sound coming off of the stylus in the turntable interacts with the electrical audio signal in the room, and how immediate and simple and elegant the process of listening is when compared to the ridiculous contorted FFTs we have to go through to hear music in the digital domain. I'm even a fan of the above-mentioned surface noise and the personality and patina it adds to a recording as a unique object deserving of its own special attention.
But I'm also a Naxos subscriber and an eMusic subscriber too, because I think they offer something that I can't have with my treasured vinyl -- wide, cheap, easy variety.
The world will always have a desire for analog media. As its civilizations grow increasingly infoheirarchical, though, the world will increasingly have a NEED for universally-available access to the artistic treasures of humanity.
So -- I'm with you in the spirit of your post, which I would summarize in two words: "Quality first!" But I would modify it like so: "Access for all, but never forget about the importance of quality!" Naxos and other labels do need to start paying more attention to sound quality on both ends of the process (recording and delivery) or the digital copies of performances we have/listen to won't be worth the hard drive space they take up or the fees we pay the distributors.