CD, analog, HRA ( High Res Audio)

Analogic, digital, sisteme audio, casti, revelatii si review-uri
Post Reply
User avatar
LEKTOR
Posts: 1774
Joined: 05 Oct 2008, 16:08
Location: Audiofilsburg

CD, analog, HRA ( High Res Audio)

Post by LEKTOR »

Un articol interesant

" Oct 21, 2014 - Two panel discussions of high-res audio brought some major talent to bear on the issue.

Last week, I attended the AES (Audio Engineering Society) convention at the Los Angeles Convention Center. Since it's a trade show for the pro-audio industry, you might wonder why I went. For one thing, I used to cover that industry, so it was like old-home week for me, and I got to see many folks I know from those days. But more importantly, pro audio is where consumer audio begins—AES is a show for the engineers who record and master the music we all listen to, so it's important for consumers to understand what they're thinking about and what tools are available for them.

I was particularly interested in anything related to high-resolution audio (HRA)—and there was plenty of it, thanks in part to several panel discussions on the subject. The High-Res Audio Production Workshop panel included Leslie Ann Jones, moderator and director of music recording and scoring at Skywalker Sound; Chuck Ainley, a well-known producer/mixer/engineer who works mostly in Nashville; John Burk, Chief Creative Officer of the Concord Music Group record label; Bob Clearmountain, a legendary producer/mixer/engineer; and Ryan Ulyate, another highly regarded producer/mixer/engineer.

The first point of discussion was the definition of HRA. For the most part, the consensus was "anything higher than CD specs"—i.e., PCM at 44.1 kHz sampling rate and 16-bit word length, aka 44.1/16 or 16/44.1. (None of the panelists currently use Direct Stream Digital or DSD, the format used by SACD.) Ulyate uses 24/48, while all the other panelists use 24/88.2 or above. Ainley commented that "HRA is whatever pushes the capabilities of current technology." And of course, lossily compressed files like MP3 are nowhere near HRA no matter how they were originally recorded.

Clearmountain wouldn't go quite as far as Ainley. He said there's not much difference in sound quality between 44.1 and 96 kHz, though the higher sampling rate and bit depth makes a huge difference when digitally mixing and equalizing, because they provide much more mathematical "headroom" to perform the necessary computations without introducing overload distortion. He also said that 192 kHz is actually not as good because there's more jitter, and analog electronics—which he uses extensively—can go into oscillation with all that ultrasonic information, so we might as well record at 96.

Speaking of analog, Ulyate made an interesting comment about mastering for CD—he converts the 24/48 recording to analog and then back to digital at 16/44.1 with no sample-rate conversion so he can hear the final result as he's working on it. As he said, "You must follow the files all the way to the label, including FLAC compression," which he claimed can alter the sound depending on how it's done. On the other hand, Clearmountain said that modern sample-rate converters can downsample from 24/96 to 16/44.1 with no problem whatsoever.

Everyone on the panel agreed that standardization is very important, and the lack of any such standard is tough on them. For example, some plug-ins—software modules for digital-audio workstations (DAWs)—are not yet high-res capable. Ainley told a story about how renowned engineer Elliott Scheiner used a plug-in on the main mix that was limited to 44.1 kHz, which compromised the entire project. Then there's the all-too-common situation in which different audio elements to be used in a project originated at different resolutions—a nightmare scenario.

Ulyate commented that HRA "lets us hit the reset button on the loudness wars," referring to the fact that most recorded popular music on CDs is heavily compressed to encompass a very narrow dynamic range, which makes it sound loud all the time. (I'm not talking about data compression like MP3, but dynamic compression.) By contrast, HRA recordings are often untouched by dynamic compression.

The panel for the High-Res Audio Super Session included Maureen Droney, moderator and managing director of the Producers and Engineers Wing of the National Academy of Recording Arts and Sciences (NARAS); Bruce Botnick, producer and engineer and VP of content acquisition for Pono Music; George Massenburg, engineer, inventor, and educator at McGill University in Montreal; Andrew Scheps, freelance producer/mixer/engineer; and Bill Schnee, another freelance producer and engineer and a guest on the Home Theater Geeks podcast.

This panel began with a discussion of the difference between studio and consumer "hi-fi" sound. Droney noted that engineers often talk about high-end consumer gear as being more concerned with visual appearance than with sound quality—echoing some of Mark Henninger's comments in his report from the recent New York Audio Show—while consumers complain that the music coming out of studios is compromised, especially with heavy dynamic and data compression.

Botnick added that studios have had high fidelity for many years, and that consumer gear is finally getting good enough for us to hear what engineers have heard all along. He also said that the goal of Pono is to provide consumers with music "the same way we hear it in the studio."

Next, the discussion turned to lossy data compression (e.g., MP3) and how much that impacts the sound quality. This is the subject of basic psychoacoustic research into how people respond to music being done at McGill University and elsewhere, but Massenburg did not have any definitive results to share yet. Scheps offered some anecdotal evidence he calls the "head-bob test." If you play an MP3 version of a tune, those in the audience who know the tune will bob their head along with it; if you then play an uncompressed version, even a CD, many more heads will bob.

Scheps explained his theory about why this happens, which he emphasized is not proven. He said the brain makes predictions about what comes next, and the sensory input either meets the prediction or not; when it doesn't, strange things can happen. Lossy codecs like MP3 remove information the brain is expecting, even though the brain ultimately ignores it, so the experience is less satisfying.

One aspect of this is the fatigue factor. Scheps said that he gets fatigued after listening to Spotify for a while, but not uncompressed audio. He believes the major factors are lossy compression and the quality of the playback system, not the resolution. Botnick added that a really good 16/44.1 recording blows away any data-compressed file, not because of the frequency response, but because of the openness of the uncompressed audio. On the other hand, Scheps argued that limiting the dynamic range to 16 bits, even without dynamic compression, can be more fatiguing than the greater dynamic range offered by 24 bits.

To demonstrate the difference between data-compressed and uncompressed music—as if anyone in the room needed to be convinced there's a difference—Botnick played Neil Young's "Heart of Gold," first at 24/192 and then in AAC at 256 kbps. Even on the funky PA speakers in the room, the difference was obvious—the compressed version was much more closed in, and the highs were quite muted compared with the HRA version.

Like the previous panel, this one agreed that HRA is critical for mixing and EQ, but many of the panelists thought that HRA as a final consumer format is more difficult to justify. Still, Botnick said that artists want their music to be released with as high quality as possible, so they need to promote HRA to the public, just as Neil Young is doing with Pono.

In the end, it all comes down to the music. As Schnee said, "Mediocre-sounding great music is better than great-sounding mediocre music." Still, everyone on the panel agreed that great-sounding great music is the best of all.
... living a vi(ntage)king analog lifestyle in a digital world ...

User avatar
brighty
Posts: 898
Joined: 06 Nov 2008, 19:05
Location: Bucuresti

Re: CD, analog, HRA ( High Res Audio)

Post by brighty »

Un foarte recent interviu "ingineresc" cu muzicianul, producatorul si sunetistul britanic Gary Numan, pe care multi dintre noi il cunoastem datorita fenomenalului hit "Cars" de la inceputul anilor '80:

http://www.digitaltrends.com/music/inte ... ary-numan/" onclick="window.open(this.href);return false;

Teaser:
Digital Trends: What’s your view of 96/24 high-resolution audio and recording at the best quality you can?

Gary Numan: The highest I’ve gone to is 48k. Personally, I think once you put music on a CD or MP3 and play it on some arguably fantastic system in your house or some shitty system in your car, it almost doesn’t matter. You can spend a huge amount of money and time worrying that your record is at the highest possible level it can be, and then somebody will be listening to it on some shit headphones and an iPad.

48k is a very high standard. If I sing something in the studio and then I play it back that way, it sounds exactly the same. The thing that’s most important to me is the sounds of the songs and the songs themselves. If you’ve got a great song, whether it’s in 48 or 96 or higher, it’s still a great song. People will listen to that more than they will the clinical side of it. When we moved to CDs, I thought that was a lovely step forward. And then when we moved onto digital music after that, I thought it was fantastic.[...]

User avatar
remoteviewer1
Posts: 925
Joined: 14 May 2011, 11:13
Location: Bucuresti

Re: CD, analog, HRA ( High Res Audio)

Post by remoteviewer1 »

Foarte interesant interviul, pe aceasi pagina exista si un interviu cu un membru al formatiei Kiss ce face o declaratie destul de transanta cu privire la moartea rock-ului. Interesante sunt comentariile abundente din subsolul articolului. Pe scurt omul spune ca industria nu mai poate produce si sustine noi talente dupa aparitia download-ului de muzica.
“Without music, life would be a mistake.”― Friedrich Nietzsche

Post Reply