By: Bill Dudleston, Legacy Audio Chief Designer
Which begs the question: Are you really 24-bit capable in your system?
When Sony and Philips introduced the Compact Disc in 1982 consumers were greeted with a 16-bit audio format that achieved a 90 dB S/N ratio (96.33 dB theoretical). Today’s recordings are mastered at a bit depth of 24 bits while quality DACs achieve a 125 dB S/N ratio (144.49 dB theoretical).
Nearly four decades have passed since the CD’s debut and consumers can now stream high-resolution recordings with 35 dB greater dynamic capability.
So, you have this extra 35 dB…where does that go? At the same playback level, that’s 17.5 dB more upward dynamics with a noise floor 17.5 dB lower. This demands a more powerful, yet quieter amplifier to take full advantage.
You will enjoy a lower noise floor in your recordings and stronger crests. While that might not make your speakers play louder, the real difference is not in averaged RMS level – it is heard in an increased resolution.
My former colleague Dr. Bela Julesz, while teaching at Rutgers, explained that our perception/cerebral processing of audio and video share many similarities. Let’s recall analog VHS tape rendering a mere 260 lines on a 480 scan-line analog monitor. Now recall the increased definition afforded by 720p video and the excitement that followed with the first look at a crisper 1080p image at 2.1 megapixels. Today we can stream 4K at 8.3 megapixels into our homes and the images have taken on a new dimension in depth. This is largely due to our ability to resolve subtle shadowing in all this information. This is not a matter of just brightness or gain, but dynamic contrast.
Today’s higher resolution recordings benefit from amplifiers offering increased voltage capability and significantly lower noise floor. Speakers with large motors can generate high back-EMF on these recordings and will benefit from amplifiers with high damping characteristics. Additionally, many full-range speakers now employ two or more woofers and can demand higher current than single woofer configurations of the past.
In the spectrum shown below, you can see the difference between the music waveform at 24 bits vs. 16 bits. Notice the lower height (lower voltage) at points 1 and 2 in the 16-bit waveform. This is compression and lower dynamics. At point 3, the height is the same, and at point 4, the voltage is actually a small amount higher. This still represents dynamic compression, as the difference between points 1 and 2, and point 4 is even less.
You can click on the spectrum to see a larger version which will make it easier to discern the height differences.
Power is the product of current (i) and voltage (V). Higher amplifier power assures the proper snap in the transients and prevents clipping that can damage high-frequency drivers. Higher power means dense musical arrangements will avoid the smearing, strain, and the inevitable distortion as complex passages reach higher levels.
Let’s first recall that the 35 dB realizable increase in total dynamic range difference benefits the noise floor greatly in 24-bit recordings. For calculation purposes, let’s assume that fully half of the 24-bit overall level advantage is allocated to a reduction of the noise floor. Therefore only 17.5 dB of this increase of 35 dB in dynamic range is allocated to potential upward dynamics.
17.5 dB = 20 log V24/V16
Solving this equation V24 = 7.5 V16
If a 16-bit recording is maximized on a four-ohm pair of speakers driven by 80 watts, you would need 750 watts to realize the potential increase in dynamic articulation afforded by 24-bit while enjoying a music lower noise floor.
If you thought loud car crashes, rocket launches, and dinosaur footsteps are all high power is good for, you might consider the benefits of lower noise floor and increased dynamic resolution.
Legacy’s i·V Series amplifiers are the most powerful home theater amplifiers to date. Featuring state of the art ICEedge® technology, the powerhouse delivers up to 1kW into 4 ohms at less than 0.005% distortion to each of its channels. The elegant amplifier series operates at nearly 80% efficiency at the rated output and delivers more than 600 watts into 8 ohms all channels driven while remaining stable into low impedances.
By: Carlo Lo Raso
- You started Legacy Audio in 1983 (that’s 37 years ago now), in your opinion what is the most significant thing that has happened in loudspeaker and amplifier development over that time?
Rather than a single thing, I’d say it’s the powerful combination of DSP and Class D amplification. This marriage allows a new level of precision in the time domain with greater dynamics and control. The benefits are huge. While driver material technology of course has advanced with the use of Kapton, Neodymium, graphite composites, ceramics and now graphene, the moving coil dynamic driver is not so unlike their century-old predecessors in operating principle. Even AMT drivers have been around for fifty years.
- In the United States in the ’60s and ’70s, there was the concept of “The East Coast Sound” and “The West Coast Sound” regarding speakers. There was also an established “European sound” from the likes of the BBC, KEF, and Bowers & Wilkins. Then in the ‘80s, several Canadian companies used research pioneered by Floyd Toole and the NRC to inform the designs of their speakers. As a designer and engineer, what do you look to accomplish when designing and voicing Legacy speakers? What motivates you here?
You are spot on with your historical overview, Carlo. Bottom line, the Legacy difference is greater dynamic range, lower distortion, higher acceleration, and wider bandwidth. Our designs exhibit room ready carefully optimized radiation patterns which weigh off-axis response in the overall voicing.
The end goal in a Legacy speaker is a natural and articulated voice that remains effortless even at high volumes. As a long time recordist, I realize that I am a hypercritical listener. From my four-thousand album library and daily use of high-resolution streaming services such as Qobuz, I have chosen several hundred tracks as a reference for voicing. Some are immaculate, some a touch warm, some known to be bright. I also listen to pink noise (not just measure it) when I voice a speaker. Having tuned tens of thousands of speakers in over 30 years, it is extremely useful and familiar as a sonic gauge. We refine the voice of a loudspeaker progressively as we gain more and more experience with it in dozens of rooms.
I work persistently on driver array development. The progress from this development becomes the inspiration for new designs before DSP or internal amplification is specified. Since Legacy is virtually completely performance-driven, the motivation remains pure.
- There is always the back and forth out there between Objectivists and Subjectivists when discussing audio. When you are working on a speaker design where do you balance the weight of measurements with the weight of listening? Have you ever had a design that measured superbly but did not sound that way?
Of course. It’s usually a result of initially emphasizing a single measurement parameter while overlooking others. For example, flat response without regard to distortion, time domain, or lobing pattern.
There’s no valid reason to trade off objectivity as this is where design goals are well chosen. Subjective interpretation is essential to evaluate whether the sound is real, etc. Subjectivity allows us to evaluate and set new objectives. I would never abandon the objective perspective.
The Fletcher Munson curve is an example where both sides are equally important. We prefer different frequency response shapes based on a range of listening levels. Objective analysis documents the subjective conclusion. One has to weigh a very broad sample of recordings to feel that at a typical listening level the balance is there.
And since you’ll always have recordings that are outliers -I include features like the contour controls in our Wavelet processor interface to help with those.
- Out of all the speakers and electronics you have designed, do you have a personal favorite? What is it and why?
The speaker that taught me the most is Whisper. It is the first truly full range directivity-controlled design. It offers increased clarity, greater placement flexibility, and a significantly broader sweet spot. While guys like Harry Olson, Earl Geddes, and Richard Modaferri understood these benefits long ago, the hi-fi world was slow to understand. It took 10 years to overcome 20 years of Bose touting diffusive designs. Now, room correction and directivity control are well-accepted techniques with pros and audiophiles. Look at the plethora of woofer-based dipoles employed out there now to improve directivity. But even now they are still not employing differential alignments using DSP to delay the front array which is the key to the design.
- In the above Op/Ed, you go into detail about what is required to have a “24-bit ready system” and you mention how the Legacy’s new iV amplifier series are well suited for this. What other traits are required for an ideal modern amplifier and how do the new iV series power amplifiers meet those goals?
The designs eliminate DC offset which is essential for speaker linearity. It also offers wide bandwidth, stability in a wide range of impedances, and a high damping factor (low output impedance). This is important because, under high power conditions, woofers can generate up to 50V of back EMF.
- Legacy has been using DSP processing with loudspeakers (ie: via the Wavelet processor) for some time now. Do you see the room as being the primary obstacle to getting good sound and what does the future hold for DSP and Legacy speakers?
The end goal is for the listener to perceive the complete sonic portrait of the recording event. To do so, there must be compensations in a number of ways. Compensations for the limitations of the recording process, compensations for the loudspeakers’ interactions within the environment, and compensations even for how the sound decays within the environment. Certainly, the room and its interactions are the biggest remaining challenges beyond the original recording itself.
My life’s work has been dedicated to the reconstruction of a sonic event. This wave launch reconstruction involves a sequence that ultimately compensates for deficiencies in each of the steps from recording to the loudspeaker’s internal deficiencies to the room’s interactions and ultimately even the ambient field and how it decays. Obviously, with headphones, the room’s interface is replaced by the direct coupling phenomenon, which is a little bit simpler, but still quite complex.
- Headphones seem to be a consistently growing segment of the audio market, including high-end audio. Has Legacy ever toyed with the idea of building headphones or a headphone amplifier?
We have research going on in that area, along with some binaural research. In particular, I’m interested in developing algorithms that will provide externalization of the sound source in a more natural way. Presently, it is virtually impossible to get a sense of “forward” in an image via headphones. Previous circuits like the Bauer shuffle circuit were designed to reduce the unnatural separation due to a lack of crosstalk to the opposite ear. There’s lots of work to be done here. While there are products out there designed for gaming and creating Atmos mix-in of many channels, this is really about trying to get realistic stereo reproduction correct.
Frankly, whether playback is via loudspeaker or headphones, it comes down to the recording process itself and how it’s properly interpreted. The following method is what we’re using to refine the stereo playback process.
Record an event such as a jazz ensemble performance using traditional ORTF or Blumlein methods, which provide traditional stereo encoding. Record the same event with conventional spot mic’ing with competent mixing. This usually pans the vocalist equally into the left and right channels to create a center image.
Simultaneously on adjacent tracks record the performance with a binaural head in an optimal listener position (in the case of a symphony, many conductors will tell you it’s from the conductor’s podium). With those factors in mind, we capture the recordings in a binaural playback as the experience reference.
Now play back all of these recordings that were captured using the above techniques through loudspeakers at an optimal listener position and/or headphones, again being recorded by the binaural head. Finally, you have an apples to apples comparison of what’s being heard from an optimal position at the live performance vs. how it’s being presented to the listener through playback methods. You can even sum the two with one in reverse polarity and hear only the differences if you pre-record a click track to align. These differences specifically indicate the errors introduced in the transcription. Crosstalk is analyzed the same way.
That’s the kind of research we’re doing right now. The algorithms for DSP, the compensation for loudspeakers, headphones, and the amplification are all considered part of the same project.
- Surround sound and home theater are now well established and accepted when it comes to watching movies at home. Multi-channel surround sound for music stalled and never quite got the same acceptance. How do you feel about music in surround and do you see it as possibly having a resurgence at all?
Yes, but I see it more through the two-channel playback medium. One of the reasons why two-channel has remained so popular is the magical effect of a phantom center channel appearing between two speakers. With our algorithms at this point, instead of a soundstage between stretched like a clothesline between the left and right speakers’ tweeters, we’re now able to externalize outside the left and right speakers and define planes of depth. This is based on room correction being applied first, and an Omnio/SUT (Stereo Unfold Technology) circuit being applied second.
The research done for our VALOR system was precisely what this is about– multi-channel sound from two loudspeakers in the front. Virtually any musical event is something that is being presented in front of you and then accompanied by the hall or room effects that can be enhancing. It’s not the same as trying to fly a fighter plane with someone closing in on you from behind. 99+% of music content is in front of us at a live performance, so stereo has a very solid future.
- When not focused on designing and testing, what music does Bill Dudleston enjoy when he just wants to sit down and listen?
I listen to such a wide range of music from jazz to Zydeco to symphonic to folk and some alternative. I like recordings that are naturally mic’d with limited use of sampled tracks. Sounds and mixes that are rich with color but not congested. If it has a gospel-like soulful feel and a backbeat I’m usually hooked. I like singers that can really sing – not just stylize. The qualities that Roy Orbison, Tom Jones, Brook Benton, and Etta James possessed. Today the talent is there in singers like Beth Hart and Adele. I witnessed Christina Aguilera belt out “At Last” live acapella when she was 17-18 years old. She weighed about 95 lbs. and demonstrated enormous talent. I wish her studio releases weren’t so overproduced.
- How has the COVID-19 situation affected Legacy Audio?
We’ve been blessed. Orders kept rolling in and our staff is back in full gear.
- Beyond how the industry has been affected by the COVID-19 situation, is there anything that concerns you about the state or the future of the Audio Industry as a whole?
There will always be a demand for quality sound – of that I am certain. But speaker efficiency needs to increase industry wide. Less than 2% of electrical energy is converted to acoustical energy. Remembering the consumer is usually the winner as technology advances, we need to accept that Millennials will expect things wireless and will not tolerate bundles of large cables. Wireless streaming to even esoteric high-end designs is inevitable.
- What are your 3 “Dessert Island” Albums?
Ry Cooder – Buena Vista Social Club
Van Morrison- Magic Time
Fritz Reiner- The Complete Chicago Symphony Recordings on RCA, 408 songs over 63 Discs
Bill Dudleston is President, founder and Chief Engineer of Legacy Audio, a high-performance audio and home theater equipment manufacturer located in Springfield, Illinois. Bill is a member of the Audio Engineering Society and the American Institute of Chemical Engineers and holds numerous patents for various circuit topologies and acoustic alignments.