How much preamp integration before sound quality suffers?
Boomzilla -- Mon, 12/14/2009 - 21:33
This question has to do with the hair-shirt minimalist philosophy vs. the "complication isn't necessarily bad" one. I see preamps (let's take an Audio Research as an example) that do one and only one thing - amplify & switch input signals before passing them on to the power amp. At the far end of the spectrum, I see AV preamps that have equalization, digital processing, room analysis, and the kitchen sink.
The prevailing audiophile dogma seems to be that the complicated preamps sound not so good as the simple ones. Yet I see startling examples where this isn't so - to wit, the recent Absolute Sound kudos to McIntosh for a variety of their (less than simple) solid state and tube preamps. The McIntoshes include tone controls, significant switching, and far more parts than the Audio Research of the same price range, yet both receive rave reviews.
Of course, I understand that the quality of parts and engineering can overcome some of the design choices (compromises?) that must be made to integrate more functions into the preamp. Nevertheless, is there a point at which no amount of parts and engineering quality can overcome the disadvantages of the design's complexity? If so, where does that point reside? Do the addition of:
Tone controls
Tuner
Balanced and unbalanced inputs & outputs
DAC converters
Surround sound decoding
Room analysis & correction
Etc.
damage the sound beyond redemption compared to the Audio Research that lacks these features? Which features do the most damage and which are most amenable to parts quality and engineering remedies? If features are irrelevant and circuitry design is all that matters, why don't the minimalist companies add features wholesale?
Before I invest a significant amount into a preamp that I may have to purchase without a meaningful audition, I'd like to clearly understand what features to look for and what to avoid. I realize that this may be a hopeless quest due to the number of variables, but I'd feel negligent if I didn't at least make an attempt to understand.
All content, design, and layout are Copyright © 1999 - 2011 NextScreen. All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Your question is certain to spark a debate, including a debate between TAS writers. Some feel that tone controls (and a balance control) give the user the ability to tailor the sound of a particular recording and enhance the listening experience. Others think that such signal processing is not only an adulteration of the signal, but also puts additional sonically degrading circuitry in the signal path.
There's no question that the addition of DSP room correction, A/D conversion, D/A conversion, and surround-sound decoding compromises sound quality (in my experience). I've reviewed a number of outstanding AV controllers and even the best don't approach the performance on stereo sources as a dedicated two-channel preamplifier. That's why I have both a controller and 2-channel preamp in my system; 2-channel signals never go through the controller.
As for balanced inputs, how the signal is handled inside the preamp is key. Some preamps with balanced inputs and outputs actually have unbalanced internal architecture. The balanced signal at the input is converted to unbalanced by a differential amplifier. The signal then goes through the preamp as unbalanced, and is converted to balanced by a phase splitter. When you use the preamp's unbalanced inputs and outputs, the differential amplifier and phase splitter are not in the signal path. This preamp thus adds two active and unnecessary stages to the signal path. Other preamps that are fully balanced, have four identical signal paths through the preamp (left +, left -, right +, right -) and a four-element volume control. This approach is obviously more expensive to implement, but is the right way to handle balanced signals.
The presence of digital clocks inside a preamp is bad news for analog signals. Digital circuits radiate noise that pollutes the analog signal.
I'll have to put myself in the minimalist camp.
Thanks, Robert - I appreciate the informed opinion. Off the top of your head, do you know which manufacturers provide a "pure" balanced input-output path and which process in unbalanced mode? Is there any way for the consumer to tell from the manufacturer's literature?
My intention, by the way, was never to spark a debate, but only to better inform myself. Divergent opinions are perfectly acceptable provided that the justification is clearly stated. The questions are esoteric enough that there doesn't seem to be a plenitude of (credible) information available...
Thanks again & Happy Holidays! - Boomzilla
A good sense of humor makes it ALL sound better!
The way to tell if a preamplifier is truly balanced is to ask the manufacturer or pop the top and see if it has four signal paths or two. If the preamp uses a potentiometer for a volume control, you can look at the number of elements; a two-element volume control indicates a single-ended circuit (left and right) and a four-element pot a balanced circuit (left +, left -, right +, right -).
THAT, even I can understand - Thanks!
A good sense of humor makes it ALL sound better!
Hi guys,
I have been considering this qustion as well. I have been saveing up for some time now to dip my foot in to the hi-fi pool. My system will be placed in my living room and will have to do double duty as far as music and movies. I listen to music every day and watch movies a couple times a month.
My plans / hopes have been to find a set of main speakers with two sets of binding posts so that I can hook them up to both a stereo amp and an avr. It is my thought that this will allow me to get the best sound for music, and still enjoy a movie now and then. Am I thinking correctly?
Hi Jers -
Alas, to the best of my knowledge, no speaker made has two "full-range" pairs of binding posts so that you can switch between two different amplifiers. The speakers that DO have two pairs of binding posts typically have one pair going to the woofer of the cabinet and the other going to the midrange/tweeter section. Such speakers are typically fitted with a jumper between the woofer and other pair so that a single speaker cable (2 wires) can serve the speaker.
You need to consider one of two things, IMHO - one would be a speaker switch box that would allow either amplifier to feed the speakers OR just run wires from both amplifiers to the speakers and plug in only the amplifier that you wish to use at the time. You DON'T want both amps connected to the same speakers simultaneously - bad electrical mojo!
Happy listening! - Boomzilla
A good sense of humor makes it ALL sound better!
Thanks for the reply Boomzilla,
I had been thinking of buying a pair of Focal 837V at one time. I was considering running two sets of wires as I did not want to add another box to the chain. Looks like thats what I will be doing (with the wires, not the Focals). The lure of seperates is just to strong!
Thanks again,
Jers
I think this is a great topic that can be taken one step further. How about all the recent dacs (Berkeley, Metric Halo, PWD) that have preamplifier functionality built in? Does adding a separate preamplifier degrade the signal and make the music sound worse? At what level of preamplifier do you have to reach where the benefits of using it, instead of the built in preamp function are evident.
Hi Hiro!
Now THAT'S a good question - I almost bought the highly-rated Benchmark DAC1-HDR with the DAC and preamp in the same box. By Mr. Harley's claim, having the two functions in the same box degrades the analog sound due to digital signal leakage. However, Mr. Robert E. Greene, who writes for the same magazine as Mr. Harley (The Absolute Sound) found the Benchmark product to be excellent.
In fact, by definition a DAC actually HAS an analog output, so EVERY such product has both digital and analog circuitry in close proximity and in the same box! By Mr. Harley's definition, every digital source is contaminated with "digititis" by proximity when it is converted to analog. Therefore, only analog sources (with no digital in the signal chain at all) can remain free of digital contamination.
Taking the train of thought yet farther, if the analog signal is to be inevitably contaminated with digital leakage, no matter whether the conversion is done in a dedicated DAC, a DAC-preamp, or an AV receiver - does it really matter AT ALL where the conversion and contamination occur? In fact, is the signal degradation that Mr. Harley refers to a function of digital signal leakage or of other factors entirely? If Mr. Harley can conclude that a state-of-the-art DAC (let's say, an Audio Research or a Bryston) is just as contaminated by the digital conversion as is the signal in an audio-video receiver (AVR), then the digital-to-analog conversion itself IS the source of the problem, and all digital music is damned forever to sound worse that its equivalent analog source. On the other hand, if the high-quality DACs sound significantly better than the AVR, then the conversion and "signal leakage" are not the core problems at all, but rather artifacts of poorer-quality parts and engineering in the cheaper component.
Mr Harley?
Thanks - Boomzilla
A good sense of humor makes it ALL sound better!
The amount of signal processing (and thus digital noise) in a controller is vastly greater than in a DAC. In addition, controllers have video signals going through them. Moreover, DACs are a lot less crowded than controllers, allowing designers to better shield the analog electronics from noise in DACs.
I find that DACs with variable output driving a power amplifier directly sound better than when going through a preamp, even a world-class preamp. I have the output of a Berkeley Alpha DAC and a dCS Scarlatti feeding my preamp, but when I want a special listening session to high-res, I move the interconnects driving the power amplifier from the preamp to the Berkeley or to the Scarlatti.
Thank you for the clarification, Robert. My follow-up questions are:
1. Provided that there is no video signal actually routed through a controller, does it still sound worse than a plain DAC?
2. When you speak of "DACs with variable output," do you mean those with digital volume attenuation only? I understand that those attenuate the resolution of the signal?
If you mean a DAC with a passive preamp, the passive preamp is still analog, AND in the same box as the digital signal
(although with additional physical separation).
If you mean a DAC with an active analog buffer or gain stage, then isn't that a DAC with a preamp built in?
I'm confused...
A good sense of humor makes it ALL sound better!
In my experience, no controller matches the sound quality of a dedicated two-channel preamplifier. It is likely that the reason isn't just the presence of the other circuitry inside the box, but the fact that the manufacturer must dedicate so much of the parts budget and design time to the complicated decoding functions, not to mention eight channels of circuitry. To address your first question directly, the video clocks are still running even when there's no video signal going through the unit. Some controllers allow you to shut down all non-essential functions when listening in analog-bypass mode (no A/D or D/A conversion, or signal processing).
Question 2: DACs that attenuate in the digital domain lose 1 bit of resolution for every 6dB of attenuation. The Berkeley DAC I use happens to produce a comfortable listening level in the range of 3dB to 12dB of attentuation. Berkeley claims that their attenuator actually sounds better than an analog volume control because of the implementation. If you start with longer word lengths, the reduction in resolution is less significant. In addition, digital filters operate better when the signal they are processing is below full-scale.
Once again, thanks for shedding light, Robert. The fact that the video circuitry is still active in the AV preamp makes sense. Some AV preamps don't have an analog bypass mode, and those that do don't seem to be very forthcoming about what gets shut down and what stays active when the bypass is engaged.
Despite the fact that your DAC does attenuation in the digital domain, does it not have an analog buffer stage downstream of the actual converter chip? I'm not yet understanding why a DAC, which has a digital input and an analog output (usually in close proximity and in the same box) would sound different in any way from a DAC that had either a passive analog preamp or a preamp gain stage in the same box. Either way, the digital signal is in close (often immediate) proximity to the analog outputs. Even with digital attenuation, your Berkeley isn't outputting a digital signal, but rather an analog one. After conversion/attenuation, you are STILL in close proximity to the digital "noise" source inside the cabinet.
I can easily understand why a DAC with its own attenuation would sound better routed directly into the power amp than with an intermediate preamp. Less circuitry is usually better circuitry...
As to digital attenuation, do non-up-converting DACs also have that luxury, or is the additional word length provided by upsampling necessary to high-quality digital attenuation? I see that products such as the Logitech (and indeed, Apple's own iTunes) offer attenuation (presumably in the digital domain), but neither product (to the best of my knowledge) does any upsampling.
A good sense of humor makes it ALL sound better!