BAASnotes.com – Bay Area Audiophile Society

Is Music Math (and vice versa)?

Physicist Glenn Elert has included a nice exposition on the mathematical basis of music in his online physics text (click here).

From Beck "Sea Change" DVD-A

Glenn keeps the tech description at a “Popular Science” level for much of it, but descends to undergrad-college-level in parts. So there’s something for everyone. (nice illustrations too.)

If you think about the content, I believe you’ll emerge with a thing or two to enrich your listening – at home or in live venues.

Bob

PS  The pic at right shows some anamolies I’ve detected on a popular DVD-A. I am shocked and amazed at what I find in some “hi rez” material. (At least this one goes above 22kHz – some don’t.)

Can Loud Music Add Hard Drive Latency? Does it Matter?

Note that this ad hoc demonstration does not represent proof, let alone applicability to home playback environments. Also, extensive use of caching in player software would seem to negate this effect in sound systems.

However, perhaps this is why some austensibly-sensible audiophiles report better sonics with solid state drives (SSDs)?

Thanks to PeterT for pointing this out to me, from reporting in the San Francisco Chronicle.

Bob

Reproducing the Live Event – How Loud is Loud Enough?

We audiophiles are obsessed with frequency response (FR):

• Is that system “flat” to 20hHz?
• How low will these speakers go? I *must* have 20hz!
• There was a “BBC dip” in that Tannoy. No, I would call it a smile.
• Urgh. Those speakers are waaarrrm.
• And so on….

Even looking at the august Stereophile, keeper of the truth of measurements, we see JA focusing primarily on (for loudspeakers) impedance, FR, and time domain criteria.

What’s usually missing, often ignored, is the amplitude response (AR). How loudly will the system play distortion-free, and how softly? As any Lowther or horn fan would say, this dimension is also vital to producing the illusion of a live event.

Wait, you may say, everybody (including JA) publishes sensitivity data (dB/V for speakers, and dB gain for amps) and uses said data for buying decisions. True enough. But our present system of metrics makes it difficult to answer the original question regarding distortion-free amplitude envelope.

More fundamentally, before we even worry about what our own system can do, we might ask the question: how much is enough? How loudly (and softly) must my system play to reproduce “live levels”? (Assuming one cares. And BAAS members seem to have wildly disparate opinions on this point.)

While researching this issue in the context of my own system, I discovered an excellent survey article on the issue of audio dynamic range (click here). The source was a Greek website (unsurprising given the audiophile mania over there). I will warn you that the piece is long, winding, and – given the blazing red background – hard on the eyes. But it rewards the reader with a trove of cool data, from microphones to the digital chain and (eventually) out your speakers and into the room. Along the way, we investigate the noise floor of Davies Hall and Skywalker and more.

I’ll give you the gist. How much dynamic range do I need? More than you think. Can my present system do it? Almost certainly not. What can I do about it? Ummm, 24 bits, 500+ watts, multichannel….

Read the article. It’s a bit of work, but you’ll likely learn something new and useful.

Enjoy your Thanksgiving.

Bob

How Should Your Room Measure?

Yes, the title of this post is correct.

We audiophiles read a lot about how our components should measure, i.e., how to quantitatively specify system performance. You know the drill: 20-20Khz, 8 ohm, slew rate > 10, noise < 100dB, etc.

But how do we specify correct room behavior? And what are the consequences of poor behavior? Should any of this change with dipole speakers?

Sure, these answers can be found buried in textbooks and AES research papers. But few, if any, easily-accessible references are available.

Now, thanks to BAAS member Nyal Mellor, we have such a document. In fact, he and co-author Jeff Hedback wrote it! And they graciously donated a copy to BAAS for download.

You will need to use FTP to get it. See the new “Downloads” button above for instructions. It’s easy!

Bob

In Audio, Digital is Simply Not Digital

I wrote earlier in this blog about how digital audio signals must ultimately be represented by analog signals – particularly when traversing a cable.

This article in EE Times, while geeky, presents more evidence about the slippery problems of digital audio – this time from the perspective of USB.

Don’t think USB cables matter? Read on – if you dare!

Bob

Thunderbolt – The world’s most sophisticated cable?

The geometry of Nordost Odin. The circuit-in-a-cable approach of MIT. The static charge on Synergistic shielding. All examples of how audio cables have progressed beyond the twisted pair.

But these are childlike in sophistication compared to the Thunderbolt cable, as featured in new Apple Macintosh products.

T-bolt, which some suspect to be the ‘next big thing’ in computer-audio connectivity, houses a dozen I/C’s in each cable!

Check out the details in this iFixit blog piece.

Fascinating stuff (for a geek).

Bob

Be a Better Audiophile (Or Maybe Reviewer?)

This just in from member Bob S.:

Click here for Ear Training software from Harmon Intl.  Really cool, and it's free!

Some bugs are discussed here.

Finally, Figure 3 (here) shows the poor performance of a group of audio reviewers (how chosen?) compared to listeners trained by Harmon's method.

Bob S.

Thanks, Bob!

Light Peak – Overkill for Audio Interfaces?

This week, Intel and Apple announced the first incarnation of Light Peak – on the latest iteration of Mac laptops. Apple calls it Thunderbolt.

Cheesy name aside, the 10 gbps performance is staggering. It’s more than an order of magnitude faster than Firewire 800. Want to move a Blueray movie? Less than a minute. USB 2.0 and 3.0 just got leap-frogged, big-time.

For those of us with terabyte+ music libraries, moving these and backing them up just got easier. This interface has 3x the transfer rate of a high-performance SATA drive!

Is Apple positioning this as the successor to firewire? Probably. Will it lead to better fidelity in music server systems? Who knows.

But I want one…

Bob

DSP & Digital Crossovers Using a Mac

Crossover - eFicion F300

I just wrote a lengthy post about digital crossovers and EQ over on the Computer Audiophile forums (click here to read it).

Bottom line, while not for everybody, DSP technology can revolutionize the way you interact with your system.

I also make some specific product recommendations that I thought would be inappropriate for BAASnotes (being as how we’re Switzerland and all….

Bob

The Cool Sound of Valves

Check out this article in IEEE Spectrum regarding why tubes sound better.

It’s a bit long, but its arguments are lucid and well-presented. Be sure to read the sidebars too!

Amazingly (for an IEEE pub), bothe the objective and subjective sides of the story are covered.

Also, note how many Bay Area people are on center-stage.

The money quote:

In spite of the bad press, high-end audio was estimated in a Wall Street Journal article in December 1991 to sell as much as $1 billion per year worldwide, enjoying a 20 percent annual growth rate. Japan, alone, consumes $200 million per year in high-end components.

Bob

Audio Expert to Conduct “Room Measurement” Class

Bay Area audiophiles have an exciting new event coming up on March 12^th.  This will be a room and loudspeaker measurement techniques class taught by Manny LaCarrubba.

The class will be limited to 6 participants, will be held at Manny’s facility, Sausalito Audio in Novato and will cost $150.00.  Class will start at 10am and will end around 3pm with a break for lunch.

Manny, a longtime BAAS member, is a respected recording and mastering engineer, speaker designer and acoustic expert. Among his many accomplishments is the design of the award-winning Beogram loudspeakers.

If you wish to sign up for the class or if you have any questions, please email:

Audio Objectivist Makes Case at AES 2009 (Video)

Most audiophiles who have frequented the forums know Ethan Winer – both as sound trap manufacturer and audiophile skeptic.

While the video below is more about the latter, I think that there are a few good learning points as well….

(Note: If the geeky video below isn’t your thing, try this one – with Mr. Winer on cello!)

Loudspeaker Burn-in: Fact or Fiction?

Our audiophile colleagues in Spain have taken an interesting approach to the question – they asked the manufacturers.

Here’s what top builders said.

Punch line: about 100 hours…give or take 100 hours!

(Check out Audioholics for another viewpoint.)

Tips for Better Mac CD Rips

I’m using a used 15″ Intel-based Macbook from late 2006. I know other models would give me the best sound from a Mac, but I can’t afford them.

You should get excellent sonics from the MacBook. It should not be the weakest link in the digital chain.

Do you use an external drive and software other than iTunes to rip CDs to your Mac computer(s)? Some folks use a Plextor PX-Premium CD-RW drive and AccurateRip XLD software (or something like that). They say the sound is much better.

I use the standard Mac internal drive (which is quite good). I also own a Plextor, but cannot hear a difference when using it (one exception below).Using my library (~1600 disks) for the samples, my own (informal) testing yielded the following results for Mac ripping software:

• iTunes
  • By far the easiest and fastest, but 2-3% of disks have an audible error (which I find unacceptable)
• Max
  • Rip takes 1.5x longer than iTunes, but error rate far less than 1% (acceptable to me)
  • Fairly easy to set up
• XLD
  • Rip takes ~3x longer than iTunes, but no errors encountered
  • Somewhat difficult to set up, and some manual reconfiguration necessary for every disk
  • I find both the rip times and difficulties unacceptable for normal use – I only use this for “problem disks”

A note about the ‘audible errors’: In my experience, these only occur in the first several seconds of “Track 1.” They are most commonly momentary dropouts. This behavior suggests to me that they are caused predominantly by “disk offset” issues. XLD explicitly allows the (power) user to compensate for the offset. It has other cool error-avoidance features as well. But it’s hard to live with….

Do you use a special USB cable? From whom?

I run Firewire, which I generally recommend for Mac users. I have never heard a USB or Firewire cable make a difference. But some in-the-know folks claim that they do. (If so, this would be another ‘audiophile mystery’ to scientists.)

Is it FLAC files that won’t play with iTunes on a Mac.

Correct. (iTunes can be made to play FLAC, but it’s not worth the effort!)

What converter do you use? I tried hunting for one, but couldn’t get it to work.

Max does a great job (as does XLD). I recommend using MP4/Apple Lossless format. I have never heard a difference between lossless compression and no compression.But a Mac using Firewire is a far more robust machine than PC/USB. To give you an indication, I often receive email, browse the web, download a file, and backup a hard drive simultaneously – while listening to flawless 24/192 (losslessly compressed). The only way that I’ve been able to adversely affect playback is to play (at 1x) the same CD that I am ripping (at 4x)!

Does Bandwidth Beyond 20khz Improve Sound Quality?

Some folks over at the SlimDevices Forum (here) are currently debating the impact of this 2000 psycho-acoustic scholarly research from Japan.

The results of the study support some remarkable and far-reaching conclusions:

• Frequency content beyond 20khz enhances human perception of music – test subjects favor it over lower-bandwidth material.
• Brain activity can be measured and used to correlate subjective assessments of the subjects.
• The form of ABX testing most commonly used in listening tests – several seconds of sound separated by less than a second of silence while switching samples – is inappropriate to measure a human’s assessment of musical content. The brain takes longer to form aesthetic judgments.

These tests were conducted in a controlled setting using double-blind protocols, direct measurement of brain activity, and  a statistically relevant sample size. Pretty solid.

These conclusions could have far-ranging implications to audiophiles:

•  A high-bandwidth system is necessary to full enjoyment of music.
  • Ever hear of a super-tweeter?
• Properly done, “high resolution” formats (e.g., SACD, HrX) enhance listening enjoyment.
  • More material please! And affordable high-resolution DAC’s….
• The rapid-fire “test tone” method of audio ABX testing is inappropriate to measuring human satisfaction of reproduced music.
  • Ban their use in audiophile settings and discussions.

Wow. Weighty stuff to consider this weekend.”Hypersonic” indeed.

In a future piece, I’ll discuss some of the factors that both the paper and the often-didactic forum discussion miss….

On Bandwidth (and Sampling Rate)

For our technically-minded members, I’m sure that you’re aware of Fourier Analysis and the Fast Fourier Transform (FFT) that revolutionized signal processing. For others, the core idea is that any signal can be efficiently decomposed into a number of sine waves. The more sine waves, the better the approximation.

The graphic at left shows a number of such approximations to a square wave. Engineers (even audio engineers) love square waves because of their infinitely rich harmonic structure. Listen to this sound clip (sin-sq-440-128k.mp3) to hear the A above middle C (440 hz) – first as a pure tone for 10 seconds, then 10 seconds as a square wave.

Sounds a bit like clipping, doesn’t it? It should – as the illustration below clearly shows.

Getting back to the first graphic (above left), we clearly need a lot of sine waves to approximate harmonically rich sound signatures like a square wave (or a bowed string). And those sine waves must be of increasing frequency. Thus, we need a lot of (analog) bandwidth. If we’re in the digital domain, this translates to a fast sampling frequency.

Hopefully this analysis helps to illustrate why many audiophiles desire 96+ khz  sampling (e.g., SACD) and/or ultrasonic analog system bandwidths (e.g., ribbon tweeters or supertweeters).

But the jury’s still out on whether these characteristics are needed to render a truly accurate musical event.