Sunday 6 October 2013

MUSINGS: Updates & The Value of Objective Measurements...


Life has been busy getting things done with the new home. Also, I went ahead and bought a new Paradigm SUB 1 to add to my home theater room along with a Signature C3 center channel - piano black of course :-)...  Should be an exciting fall/winter as I get things up and running!

In the midst of this, I'm going to try getting a few comparison measurements of the pre-amp characteristics of the new Emotiva XSP-1 vs. my old Denon 3802 AV receiver to see objectively what differences can be found. There does appear to be a significant audible difference plugging a DAC into the XSP-1 then into the Denon external input and using the Denon as an amp compared to just plugging the DAC straight into the Denon as a pre-amp.


A couple of quick updates:

1. It looks like JRiver has new beta versions of JRMC 19 (19.0.51) which supports the TEAC UD-501's native ASIO DSD128 mode. No problem with using "2xDSD in native format" upsampling of PCM to DSD now. Thanks to InflatableMouse for getting TEAC and Matt over at JRiver talking. With this "fix", there's no need for ASIOProxy any more but I suppose the technique could be useful for other DAC's.

2. I continue to update the list of suspected upsampled 44 & 48kHz PCM-to-SACD titles. Thanks for the entries from various friends and E-mails over the months. Again, I think it's useful to have a look at this list if you're a collector of SACD's. Useful to ponder just what is the benefit of DSD based on these examples as well...

The Value of Objective Measurements: A Case Study

Today, I want to discuss an interesting device which I heard back in early 2012 when a friend bought one (I think he might have even been on the wait list to be one of the first to get it)... The Wadia 121 Decoding Computer.

I remember listening to it and thinking "this isn't bad". Details seemed reasonable.  The remote feels good. I like the idea of good "lossless" volume control built into the DAC. I wasn't blown away by it though. Unfortunately, this was before I started writing this blog and spending time with measurements.

Over the months since then I remember keeping an eye on what others were saying about the device. I assume Wadia did a good job sending out "loaner" units to the various audio reviewers...  A Google search shows quite a number of subjective reviews of this device. At a price point of MSRP $1299, it's not a "top end" DAC based on price but that's still quite a chunk of change. Understandably, reviewers of "high end" gear did not suggest this was the best sounding DAC they've heard, but on the whole, it received very decent, positive remarks...  Let's have a look; I'll throw up a few summary quotes in no particular order based on the reviewers' conclusions trying not to take things too far out of context: [As usual, I present these quotes as "fair use" for the purpose of discussion, criticism, and research.]

The Computer Audiophile (August 16, 2012):
"The Wadia 121 Decoding Computer is more than competent and competes with products double, triple, and quadruple its size...  New computer audiophiles seeking their first entry into this wonderful next phase of high end audio can't go wrong by starting with the 121. They may never need another digital to analog converter."

Enjoy The Music.com (August 2012):
"The Wadia 121 Decoding Computer is the best affordable digital-to-analog converter that I've ever heard. No, I have not heard every affordable DAC on the planet – and there are new DACs in all shapes, sizes, and prices being released even as we speak. But given their track record, it is a safe bet that Wadia has not only put a lot of thought into the design of the Wadia 121, but this DAC won't be bested by any DAC for quite a few years to come..."

AudioStream (June 22, 2012):
"To my way of listening, the Wadia 121Decoding Computer jumps right onto my short list of recommended components. It strikes me as being at once refined yet not overly resolute, with a voice that sounds like music. Sweet music. I enjoyed every listening minute spent regardless of the recording..."
Wins the "Greatest Bits" award.

Sound And Vision (March 14, 2013):
"...Though the music sounded like high-res digital—not vinyl—my brain still involuntarily registered surprise at the lack of clicks and pops. I suppose it associates them with a relaxed listening experience.

"Home theater buffs tend not to think much about source components for music: We figure that as long as we own an Oppo, and maybe a turntable, we’re covered. That worked as long as music streaming was a low-res medium, merely a convenient plaything for background listening. But the advent of high-res downloads demands an upgrade if you want to get the best out of your investment in components and headphones. You just may need something exactly like the Wadia 121."
Awarded 5/5 stars in the "Performance" category.

Ultra High-End Review (June 20, 2012):
"... Reading reviews is helpful (I hope), but I think a proven track record of producing high quality components is perhaps even more important. Here Wadia, well known for producing some of the finest digital playback equipment available since the earliest days of the medium, has brought its considerable talents to bear in producing a DAC which is operationally bullet-proof at an unexpectedly modest price. This is not simply another DAC-in-the-box with off-the-shelf parts and a marketing slogan, but a component with highly sophisticated software realized in DSP which has been decades in the making, coupled with an analog section which, to my ears, is completely transparent. And with no separate preamplifier needed, your budget for speakers has just doubled. I can’t recommend it highly enough."

The Absolute Sound (Feb 28, 2013):
"I can state confidently that few, if any, potential purchasers will be disappointed by the 121’s sonics or ergonomics. I know that I could happily live with the Wadia 121—it’s that good."

So, these are words of subjective reviewers. As I noted, for the price this DAC cannot really be considered "reference" level at least from the perspective of folks who likely have heard DACs in the $5000+ range and have some expectation of what these expensive DACs sound like. There are of course comments about how this DAC doesn't quite reach the level of those รผber-DACs. Here's a nice quote from the Computer Audiophile: "What separates the 121 Decoding Computer from the rarefied air of great but greatly expensive DACs is reduced depth, air, and low level detail when reproducing the best recordings from labels such as Linn Records, Naim, and Reference Recordings." Fair enough.

So, eventually, in the July 2013 issue of Stereophile, we get their full review. Jon Iverson's subjective comments were clearly not as positive:
"After more than a month of use and listening, when I used the 121 strictly as a DAC, I found that, in most cases, its sound had a marginally burnished or rounded quality that could help tame a recording with an unruly top end, or slightly veil a great recording."
What was somewhat stunning however was what John Atkinson found on the test bench:
"Fig.4 shows the spectrum of the 121's output while it decoded dithered 16-bit data (cyan and magenta traces) and 24-bit data (blue and red traces) representing a 1kHz tone at –90dBFS. The increase in bit depth drops the noise floor by around 9dB, implying ultimate resolution between 17 and 18 bits. To generate this graph, I fed the data to the Wadia from the Audio Precision using an AES/EBU link. To my astonishment, when I repeated the analysis using a coaxial S/PDIF link to transmit the 24-bit data, I got 16-bit resolution. The blue and red traces in fig.5 repeat the spectrum with 24-bit data and an AES/EBU link; the cyan and magenta traces in this graph were taken with the 24-bit data transmitted with the coaxial S/PDIF link. I repeated the analysis using a TosLink connection from the Audio Precision, but with no difference in the result. To check that the Audio Precision was working properly, I then used a TosLink connection from my MacBook Pro. However, I got the same result: 24-bit data but 16-bit resolution. Finally, I used a USB connection from the laptop, and although I made sure that the connection was correctly set to transmit 24-bit integer data, the noise floor was around the 15-bit level (not shown)." (Emphasis mine.)

You can also see the noise level demonstrated in Figure 6 with the undithered -90.31dB graph. Not good. [I posted on this test back in August to show what it looks like with some of my DACs.]

Basically, what the objective results show is that we have here a fancy looking DAC with some really cool "talking points" - well respected manufacturer Wadia, "ClockLink" asynchronous USB, "DigiMaster" interpolation, 32-bit 1.4MHz upsampling. But at the end of the day, it's not capable of achieving >16-bit resolution with USB, TosLink, and coaxial inputs. Even with the AES/EBU balanced digital cable, it's "only" capable of 17-18 bits. Unless one were to just use AES/EBU, there appears to be no point feeding 24-bit high resolution audio into this DAC - all those 24-bit HDTracks/Qobuz downloads would be wasted (unless you feel >44kHz sampling is much more important). To make matters worse, it seems like the USB input cannot even achieve a full 16-bit resolution - arguably THE most important interface these days. Knowing this, how can any reviewer hand out awards or grade this device as 5/5 on performance? Even if you like the way this device "sounds", isn't it still a sign of failure that it could not profit from the higher bit depth? Of course, it appears the purely subjective reviews could not comment on this "inconvenient" piece of information.

Now, admittedly, there could have been something wrong with John Atkinson's measurements I suppose, but as of October 2013, I do not see any addendum to the review. I would imagine that a manufacturer would correct this situation ASAP!

I feel that this is a good case study (one of many IMO) into why objectivism has an important if not essential place in audiophile equipment reviews. Bias and placebo are well recognized in domains of research where human qualitative evaluation is involved. I would argue even more so when reviewing "high fidelity" gear where at a certain level of quality, differences are likely very small and effects of biases become even greater - "look and feel" (pretty metal box with lights and metal remote), manufacturer reputation (ooohhh... Wadia), price ($1299 must be a pretty decent DAC right?) all can (and likely do) end up in the final evaluation of sonic quality in the absence of objective information. In some forums / web sites, it almost seems that certain reviewers feel that they are immune to this phenomenon, or even worse, have developed so much faith in their "golden ears" that they feel there is no benefit to empirical evaluation.

Remember, thoughtfully designed audio devices are engineered. They were made based on electrical and (in instances like speakers or turntables) mechanical properties. Without examination of these properties to at least verify claims (eg. that a hi-res DAC is capable of >16-bit dynamic range being fed into it, or an amplifier is capable of the claimed watts with minimum distortion, etc...), I believe the reader cannot place strong value in the reviewer having fully appreciated the limitations/strengths of the device. I'm of course not opining that there be no subjective evaluation - fit and finish, ergonomics, ease of use, reliability, visual esthetics are all important. Likewise, sound quality needs to be checked subjectively. But there's no shame in admitting that in many ways, the human ear/brain is limited and measurement devices can easily enhance the quality of a review synergistically. There's no need to see this as black or white, subjectivist vs. objectivist.

I've said many times on the various forums how I still have a subscription to Stereophile. I look forward to reading the opinions, music reviews, and of course the gear reviews. After all that's said in the subjective portion, it's always good to study those numbers and graphs to make sure the device appears to be delivering all that was promised. I wish more magazines could do that... (Don't worry guys, I wasn't paid off by Stereophile, just wanted to give credit where credit is due.)

[BTW: Perhaps it goes without saying, I feel objective measurements are especially important with devices where the putative effects seem to be without clear scientific basis (eg. anyone know what the Synergistic "Tranquility Base" does yet?)]

Musical selection tonight: "Respect the classics, man!" -- Fillmore from Disney's Cars 
Jimi Hendrix - Band Of Gypsys

Until next time... Enjoy the tunes!

7 comments:

  1. Good rundown. Objective measurements are great if we are educated to what they mean and what the standard is.

    ReplyDelete
    Replies
    1. Absolutely. I would appeal to audiophile magazines, websites... to help educate the hobbyist to better understand the technologies involved and to use objective measures to actually demonstrate hypotheses.

      A good one is jitter for example - used to "explain" differences between cables, expensive gear, etc... but commonly without further exploration as to whether this is in fact the case.

      Delete
  2. Aloha Archimago, another excellent post! With respect to the Wadia subjective reviews, I suspect it is a matter of ear education. For me, correlating listening tests with objective measures shows my bit-depth audibility threshold to be around 12 or 13 bits. This result is consistent with audibility threshold research in the field of psychoacoustics.

    Folks and possibly audio reviewers, can try these audibility threshold tests at: Fun With Digital Audio – Bit Perfect Audibility Testing The how to procedure is documented and audibility files can be downloaded. For those that have never attempted this before, it can be a humbling experience and puts directly in context what we can and cannot hear.

    Keep up the great writings!

    Cheers, Mitch

    ReplyDelete
    Replies
    1. Aloha MItch - are you in Hawaii these days? I see from your bio that you worked up here in western Canada...

      Thanks for the link and the write up as usual! Clearly with your experience on the audio production side, you have the opportunities beyond many of us end-users :-)

      Having said that, I certainly appreciate your "self disclosure" - audibility threshold around 12-13 bits. A little while ago there was a post on the Steve Hoffman forums about financial disclosure and reviewers. That's of course important... IMO just as important are "audibility disclosures" I think! We often have no idea of the reviewer's age, musical background, audiology thresholds... Surely, these parameters are important if we are to believe the subjective impressions since the reviewer's ear/brain becomes the measurement device.

      Keep up the great work... I hope that the serious audiophile (and especially reviewers!) read up on your detailed work and seriously consider the implications for themselves.

      Delete
  3. I have been enjoying reading your blog! I think its great that you are sharing your measurement results using the E-MU 0404USB. I think too many reviews have too much opinion and not enough quantitative analysis. I think it is the marriage of a trained ear + quantitative analysis that make the best reviews of hi-fi equipment. Keep on the good work! As an electrical engineer I am tempted to analyze the performance of my PC SPDIF -> Onkyo NR901 connection which many demonize as full of "jitter" and data corrupting "noise" ... but my ears tell me quite the contrary.

    ReplyDelete
    Replies
    1. What a timely response Leonid!

      Tomorrow, I'm planning to pick up a used Onkyo NR1009 for the home theater. As much as the old Denon 3802 has served me over the years and I'm focused on stereo reproduction, it's time to get more serious with multichannel! I just gotta be able to listen to DTS-HD MA and Dolby TrueHD in my new sound/media room!

      I find it fascinating that HDMI has been demonized in the audiophile world/press... Sure, it ain't perfect but I bet much can be done to improve jitter for example than what has been published based on HDMI receivers of a few years back.

      From what I can see, IMO, HDMI *is* the future for "audiophile" digital audio. Sure, USB will stay for a long time, but once you get beyond 2 channels, that integration between home theater and computer audio can best be realized today and for the forseeable future by HDMI. I suspect much of the animosity is being perpetuated by small manufacturers unable to afford the HDMI licensing.

      I think it'll be fascinating to see what the pre-out looks like using computer --> HDMI --> Onkyo as DAC compared to jitter spectra I have seen to date with 2-channel USB DAC's. Will post in time :-)

      Delete
  4. How did the Wadia 121 sound to you?

    ReplyDelete