By John Siau
April 10, 2014
It's on your iPhone, your Android and your computer. It's even on those CDs you put on a shelf somewhere. Audio that goes to 11.
If 10 is the clip point of digital audio, you actually have digital recordings that go to 11. Nigel Tufnel of Spinal Tap was on to something in 1984 when he explained that his Marshal amps "go to 11". If you have never seen "This is Spinal Tap" I suggest watching this short clip before reading on. Nigel's brilliant discussion sets the stage for this application note.
But, it's not just Spinal Tap recordings that "go to 11"; every recording you own may also "go to 11"! How is this possible? If 10 is the clip point of digital audio, how can there possibly be an 11? And, if we use Nigel's logic; if 10 is good, why isn't 11 better?
As strange as it sounds, audio that "goes to 11" is hidden in between digital samples. This is especially true when the recorded samples just reach "10". Digital systems take a snapshot of the audio signal thousands of times per second. These snapshots or "samples" represent the audio signal at an instant in time. In between successive samples, the audio is always changing. Digital sampling systems often miss short audio peaks which occur between these samples. These peaks often "go to 11", but are entirely missed by the sampling system.
Nevertheless, the short peaks between samples are not lost! These peaks that "go to 11" can be reconstructed from the surrounding digital samples. The DAC (digital to analog converter) in an audio system is equipped with digital reconstruction filters that can recover these inter-sample peaks. These filters work wonderfully until the digital processing overflows. Peaks that hit "9" or "10" will not cause an overflow, but peaks that "go to 11" may cause an overflow.
If you attempt to divide 1 by 0 on your calculator, the digital processing will overload and an error message will be displayed. Likewise, if the digital processing in your audio system overloads, bad things happen. Overflows that occur in digital reconstruction filters can produce a burst of distortion that persists for many samples. This distortion is non-musical and foreign to the natural sounds around us. These overloads often add an unnatural harshness to the digital playback system. But this does not mean that digital audio is fundamentally flawed. Some DACs can reproduce signals that "go to 11" without clipping. Benchmark's DAC2 is one such device.
Benchmark scanned over 5000 CD tracks to determine the severity of the inter-sample peaks in commercially available music. We discovered that most tracks contained peaks that were 1 or 2 dB above a full-scale "10". A peak that is 1 dB above full scale is 1.1 times as high as a full scale sample. A +1 dB inter-sample over is audio that goes to exactly 11! Nigel was right!
But back to our survey of CD tracks: we discovered some tracks had peaks that were 3.1 dB higher than full scale. This is 1.4 times as high as a full scale "10", and is audio that goes to 14 on Nigel's scale. You may own some recordings that go to 14, and you most certainly own many recordings that "go to 11".
Another twist to this situation is that MP3 compression seems to increase the occurrence of peaks that exceed full scale. This can make MP3 files sound worse than they should.
Once these problems were identified, Benchmark was able to implement a solution. The DAC2 reduces the signal level of the digital signal by 3.5 dB before it enters the digital interpolation and reconstruction filters in the DAC. This gain reduction is made up by increasing the analog gain after the D/A converter chip. The result is a DAC that not only "goes to 11", it is a DAC that "goes to 15". A peak of +3.5 dB is 1.5 times full scale (or "15" on Nigel's scale).
The Benchmark DAC2 goes to 15! Nigel should be impressed.
Secrets contributor Sumit Chawla recently caught up with Benchmark’s VP and Chief Designer, John Siau to get a little more in-depth on several subjects.
Q: "Benchmark is one of the few companies that publishes an extensive set of measurements, but you also balance that with subjective testing. Can you talk about the equipment, the listening room, and the process for subjective testing?"
Q: "Was there ever a time where you learned something from a subjective test that was not captured by measurements?"
Q: "You conducted some listening tests to determine whether distortion in the “First Watt” was audible. What test material did you use for this, and what did you find?"
Q: "The AHB2 amplifier incorporates THX Audio Achromatic Amplifier technology. When and how did the partnership with THX come about?"
Q: "Linear power supplies have been and remain quite popular in high-end devices. You favor switch-mode power supplies. When and why did you make this switch?"
... and more!
At Benchmark, listening is the final exam that determines if a design passes from engineering to production. When all of the measurements show that a product is working flawlessly, we spend time listening for issues that may not have shown up on the test station. If we hear something, we go back and figure out how to measure what we heard. We then add this test to our arsenal of measurements.
Benchmark's listening room is equipped with a variety of signal sources, amplifiers and loudspeakers, including the selection of nearfield monitors shown in the photo. It is also equipped with ABX switch boxes that can be used to switch sources while the music is playing.
Benchmark's lab is equipped with Audio Precision test stations that include the top-of-the-line APx555 and the older AP2722 and AP2522. We don't just use these test stations for R&D - every product must pass a full set of tests on one of our Audio Precision test stations before it ships from our factory in Syracuse, NY.
Paul Seydor of The Absolute Sound interviews John Siau, VP and chief designer at Benchmark Media Systems. The interview accompanies Paul's review of the LA4 in the December, 2020 issue of TAS.
"At Benchmark, listening is the final exam that determines if a design passes from engineering to production. But since listening tests are never perfect, it’s essential we develop measurements for each artifact we identify in a listening test. An APx555 test set has far more resolution than human hearing, but it has no intelligence. We have to tell it exactly what to measure and how to measure it. When we hear something we cannot measure, we are not doing the right measurements. If we just listen, redesign, then repeat, we may arrive at a solution that just masks the artifact with another less-objectionable artifact. But if we focus on eliminating every artifact that we can measure, we can quickly converge on a solution that approaches sonic transparency. If we can measure an artifact, we don't try to determine if it’s low enough to be inaudible, we simply try to eliminate it."
- John Siau