Universal and DGG Records have donated six of their new Pure Audio Blu-rays for this month’s drawing. This is the first time these classic symphonic recordings by Karajan, Kleiber and Fricsay have been available in hi-res (96/24), playable on any Blu-ray deck. The four Blu-ray selections you see here and on the Home Page will be random and divided among six AUDIOPHILE AUDITION readers who register this month on our simple form, each receiving two. Here are our reviews of two of them: Carlos Kleiber conducting Beethoven’s Fifth and Seventh, and Herbert von Karajan conducting Mahler’s Fifth. When you register for the drawing we do not share any of your information except your address for shipping. The six winners’ names will appear here next month.
Here are the five winners of the 3-CD Bartok & Kodaly Quartets set from Foghorn Records, our March drawing/giveaway. Congrats to all!: Barbara Stenby, Ronkonkoma NY; Theresa Beres, Portales NM; Michael Krukowski, Milwaukee WI; Wally Babcock, Prattsburgh NY; Eric Reiff, Hampton VA.
[Guest Editorial by John Siau of Benchmark]
What is High-Resolution Audio?
There is a raging debate over the definition of “High-Resolution Audio”. The focus of this debate has been misdirected by a lot of marketing hype. High-Resolution Audio will only truly arrive when the sum total of all defects in the audio chain become inaudible. We are not there yet, but we are getting close. The final chapter will not be higher sample rates and more bits. The final chapter will be achieving the necessary improvements to critical components in the signal chain.
Ignore the sample rate and format wars. There are several competing digital systems that are fully capable of delivering true High-Resolution Audio. These include all 24-bit (and higher) PCM systems with sample rates higher than the 44.1 kHz rate used in standard CD’s. Double speed DSD is a 1-bit format that is also capable of delivering High-Resolution Audio. Avoid falling for the marketing spin that tells you: “if 96 kHz is great, 192 kHz is better”, or “if 24-bits is great, 32-bits must be better”. Don’t fall for the spin!
As a delivery format, 24-bit digital systems are far better than any other part of the signal chain. Converters, preamplifiers, power amplifiers, and speakers cannot deliver the 144 dB signal to noise ratio (SNR) that can be transmitted through a 24-bit system. Our ears have a limitation of about 130 dB (the difference between the threshold of hearing and the threshold of pain). Our ears are essentially limited to the equivalent of 21 or 22 bits (6 dB per bit if you care to do the math). This means that 24-bit systems have a nice margin beyond our auditory system. It also means that the 96 dB available from a 16-bit CD system is somewhat deficient.
The CD was designed to closely approach the limits of human hearing. The format has almost enough bandwidth, and almost enough dynamic range to be transparent. In contrast, High-Resolution digital formats have more than enough resolution to convey every detail that can be heard by the best human ears. Nevertheless, few people are experiencing the promises of the “High-Resolution” or “High-Definition” marketeers.
Why is High-Resolution Audio failing to deliver? Do we need more bits and more samples? The answer is NO! Don’t fall for the spin!
What we need are audio components that are capable of delivering High-Resolution Audio. For example, the D/A converter (DAC) built into the typical DVD player usually has a “24-bit” DAC that only delivers a 90 to 95 dB SNR. This is the equivalent of 15 to 16 bits (CD performance at best). In contrast, the Benchmark DAC2 outboard DAC has an SNR of 126 dB (equivalent to 21 bits). The DAC2 is actually capable of delivering peaks that are 3.5 dB higher, giving the DAC2 a total SNR of 129.5 dB (21.5 bits), matching the limits of human hearing. The DAC2 connects directly to the power amplifier (eliminating a separate “preamplifer” from the signal chain).
But, an outboard DAC is only a partial solution to the High-Resolution Audio dilemma. A second key part of the problem is the performance of the audio power amplifier. A 24-bit audio system is useless if it passes through the typical power amplifier. It is nearly impossible to find power amplifiers that can deliver an SNR higher than about 102 dB. This is the equivalent of 17 bits (adequate for CD applications, but definitely not adequate for High-Resolution Audio). Anyone who thinks they can hear the difference between 16-bit and 24-bit digital audio through a “17-bit” power amplifier is fooling themselves.
Clearly, a system playing High-Resolution Audio through a great outboard DAC, may fail to deliver high-resolution unless the amplifier has adequate performance. For this reason, Benchmark has undertaken the difficult task of building a power amplifier that is worthy of the title “High-Resolution”. The new Benchmark AHB2 power amplifier has a 130 dB SNR. Like the DAC2, the AHB2 is well matched to the limits of human hearing. Together, these two devices form a signal chain that can truly deliver High-Resolution Audio to the terminals on the back of a pair of speakers.
In summary, we need to focus on the performance of our playback equipment instead of participating in the High-Resolution format wars. If your playback system can’t resolve anything better than CD quality, then “High-Resolution Audio” will remain an illusion.
Part II: High-Resolution Audio systems offer the promise of an extended high-frequency range. High-Resolution digital systems now operate at 2 to 4 times the sample rate of the standard CD. This means that these systems have the capability of extending the playback frequency range well above the 22 kHz limit of the standard CD. Does this added high-frequency range improve our listening experience? How high is high enough? Do we really need anything over 20 kHz?
In the previous post I stated that “High-Resolution Audio will only truly arrive when the sum total of all defects in the audio chain become inaudible”. In that post I discussed the need for low noise components. In this post I will look at frequency response. How much bandwidth do we need to satisfy the 20 Hz to 20 kHz limitation of the best human ears? Is a 20 kHz bandwidth good enough for High-Resolution Audio? If not, why not?
Let’s assume we have a playback system consisting of a CD player, a preamplifier, a power amplifier and speakers. If each of these has a 20 Hz to 20 kHz frequency response, is this sufficient to reproduce all of the frequencies we can hear? The short answer is no. Here is why:
The frequency response of a system can be determined by summing the frequency response of each component in the audio chain. If two components are – 3 dB at 20 kHz, then together they are – 6 dB at 20 kHz. If we look at our example system, we have four components in the chain: CD player, preamplifier, power amplifier, and speakers. If all are -3 dB at 20 kHz, then we have a system response that is -12 dB at 20 kHz. Worse yet, this system will typically measure -4 dB at 10 kHz. This system will not come close to meeting the performance of our ears! Each individual component was well matched to the limitation of our ears, but as a system, these components cannot achieve anything resembling High-Resolution Audio.
If we want to accurately reproduce 20 kHz audio, the frequency response of each component must extend well beyond 20 kHz. For this reason, Benchmark products are engineered to achieve a 200 kHz to 500 kHz bandwidth. Is this excessive and unnecessary? To answer this, let’s replace each of the four components in our sample system with components that have a 200 kHz bandwidth. The combined system now measures – 4 dB at 100 kHz, – 0.8 dB at 50 kHz, and about – 0.2 dB at 20 kHz. This simple 4-component signal chain achieves a 100 kHz bandwidth and is well-matched to the 96 kHz bandwidth of a 192 kHz digital sample rate. We could legitimately argue that the region between 20 kHz and 100 kHz may offer little musical content, and even if it does, we may not be able to detect its presence. The real benefit is that we have preserved the entire 20 Hz to 20 kHz bandwidth after passing through four audio components in a typical playback system.
Professional audio systems often have very long signal chains. These applications place difficult demands on the frequency response of each component in the chain. A chain of 16 components, each with a bandwidth of 20 kHz, will produce an overall response of about – 3 dB at 5 kHz. This is telephone quality at best! If the same system is built with 200 kHz components, the overall response will be – 3 dB at 50 kHz, and about -1 dB at 20 kHz. Benchmark’s founder, Allen H. Burdick, identified this problem when investigating the poor performance of television networks in the mid 1980’s. More details on this subject can be found in chapter 3 of “A Clean Audio Installation Guide” by Allen H. Burdick.
In summary, very high bandwidth is required at each link in the audio chain if we want to assemble a High-Resolution system. Audio components and digital formats that just meet the requirements of the human ear may be entirely inadequate when connected together in a chain. The audio chain is nowhere near as strong as the weakest link!
AUDIOPHILE AUDITION began as a local program in San Francisco and then in 1985 as a weekly national radio series hosted by John Sunier, and aired for 13½ years on up to 200 public radio and commercial stations. In September 1998 its web site for program listings was expanded to this free Internet publication.
April 2014 is our 181st issue. All disc reviews are added thru the month as they are written and received, usually daily. The most recent reviews appear at the top of each Section Index. The Home Page lists the five latest published reviews, the Section Index lists the past two months of reviews, and the Archive goes back to January 2001. The Disc Index lists all past reviews. Vinyls and Pure Audio Blu-rays are included in the SACD/Hi-Res Section, as well as exceptional xrcds.
You probably have some friends who would like to know about AUDIOPHILE AUDITION. Please pass on our URL and help expand our elite group of readers and collectors. It’s easy to do at the bottom of every review, by just clicking on the “Email this page to a friend” link. Thanks in advance for getting the word out! And remember that every time you purchase a disc via our Amazon buttons, you are also making a small contribution towards the continuation of AUDIOPHILE AUDITION.
We welcome your feedback, and we have a Reader Feedback section. Please send us your comments, and we will review and possibly post them. Check with us frequently for more reviews and news, and be sure to register for our monthly giveaways! When you do, please give us all the non-intrusive requested facts, and include your first and last name, Email and street address – otherwise we are unable to ship you your winning item! We don’t ask financial details, and we never share your information with anyone else.
— The AUDIOPHILE AUDITION staff
STAFF WRITERS: Dalia Geffen, Laurence Vittes, Tom Gibbs, Gary Lemco, Brian Bloom, Clay Swartz, John Henry, Peter Bates, Ronald Legum, Paul Pelon IV, Calvin Harding Jr., Birney Brown, Jeff Krow, Daniel Krow, Ethan Krow, Hermon Joyner, Bob Moon, Steven Ritter, Randy Haldeman, Max Dudious, John Nemaric, Doug Simpson, Peter Joelson, Patrick P.L. Lam, Brian Whistler, Lee Passarella,, Robbie Gerson, Daniel Coombs, Tim Taylor, Darren Robinson, Pierre Giroux, Mel Martin, Zan Furtwangler, Darren Robinson, Jake Barlow, John Sunier.
If you appreciate the unique audio reviews and information we bring you free of charge, please patronize our sponsors. Simply click on their banners, find their Contact Us link, and leave a short message expressing your thanks for their continued support of AUDIOPHILE AUDITION.
© John Sunier 2014