Friday, December 10, 2004

Hi-fi article # 3

-------------------

Removing Sibilance From Vocals
by Dan Shanefield

One recording of a female singer, when played back on my home system, is among the most realistic-sounding that I've ever heard. This is the Sheffield Labs album "I've Got The Music In Me," sung by Thelma Houston. When she belts out "Don't Misunderstand," in my living room, and I go into an adjoining room and look away, I'd swear she was actually right there behind me, in the middle of the carpet. (By the way, amazon.com identifies this CD as ASIN B000008GPY, and they usually can link you to the seller of a used copy. Failing that, you could probably get one at that multi-linked www.gemm.com or at www.amoebamusic.com.)

A problem with this recording, however, is that some of the "s" sounds at the beginnings of words have too much sibilance. Although the overall impression is quite real, it's somewhat unpleasant. Playing this recording on other peoples' excellent home systems, or via my own with different loudspeakers, etc., the sibilance is always a bit too much. You can buy a component called a "de-esser," which combines filters plus companders and tends to remove sibilance, but it's usually in the category of pro equipment and very expensive.

After experimenting with various filters and companders inserted into my home system, I was surprised to find that a ten-band graphic equalizer can remove the sibilance without detracting from the overall sense of realism. I just have to decrease (by 6 dB) the band centered around 4 kHz, and the sibilance is gone. (I haven't been able to make a tone control work as well as this.) The EQ also works with some other vocal recordings that I have, particularly when female singers get so "essy" that I can imagine a fog of saliva spray settling onto my carpet. Maybe your imagination is not as hyperactive as mine, but you might try this simple EQ cure, in case there is too much sibilance in some of your favorite vocals.

--------------------
(Printed in July 2004 LC)

Double-Blind Testing of Prozac
by Dan Shanefield

Although Prozac and placebo pills are both somewhat effective in treating psychological depression, there is amazingly little difference, as shown by extensive double-blind testing. Details and literature references are reported in the website
PROZAC .
The philosophical discussion by the two medicine-researcher authors of that site provides an example of why some of us audio-researcher authors have trouble convincing "golden ears" that our tests are really valid.

-----------------

Mass Producing Electronic Components: Five Horror Stories
by Dan Shanefield

If any LC readers are curious to read about some of the things that can go wrong (sometimes very wrong!) when components are made on a large scale, they could look at five very short stories that I posted on the web. These tell how I tried to be an innovator in what was then the biggest company in the world, the old AT&T. Of course, the bigger and more complex the operation, the more chances you have to really mess things up (as some of you might already know from your work with the big automobile companies).

These little stories are arranged in a continuous circle, where the reader can go to the bottom of any page and then click on an internet link, which is the colored title of the next story. Perhaps the best (or at least the funniest) place to start is at
MONSTER ,
which is described in another engineer's textbook as "Shanefield's Green Monster." (That quote is on page 231 of the book "Tape Casting," by Richard Mistler.) I used to tell these tales to legions of Rutgers students and consulting clients, as examples of what not to do. Maybe that's why the amazon.com Customer Review of one of my books mentions "the infamous DJ Shanefield." Anyhow, the stories also tell some ways in which an engineer can manage to survive, even in the tricky electronics industry. (One way is, if you can't beat 'em, join 'em, and another way is to publish some articles and patents, and then sell lots of licenses. But there were some pretty humorous moments, like when the Vice President of the company insists on calling me "Dave Sheffield" as he totally trashes my project, etc.)

------------------------------


Let's Hear Three Cheers for Wretched Excess!
by Dan Shanefield

The July 2004 issue of Stereophile magazine has what might hold the world's record for the ultimate golden-ear review, at least until something wilder comes along. It's Michael Fremer's article on the Wavac SH-933 vacuum tube amplifier, which costs $350,000 and weighs about half a ton. Using this amp for playback of good recordings, the reviewer says he heard all kinds of details that he never was able to perceive before. Like when a guitar player was tapping his foot on the stage during a recording, Fremer could "make out how far the vibration spread across the floorboard before dissipating" (page 79). Unfortunately, he neglected to tell us exactly how far that turned out to be. Nevertheless, the whole business is amazing and amusing. To see the revu, click on this:
REVU

-------------------

Archiving Favorite Music Tracks Via MP-3 by Dan Shanefield

Most of my LP discs (and also my old 78s) contain one or two tracks that are my favorites, plus many that I don't like very much. I have recorded a bunch of cassette tapes that only contain the favorites, so I can continuously play just the stuff that I like the most. However, now that many of those cassettes are 20 or 30 years old, the tapes are beginning to develop mechanical problems such as squeaking, jamming, or even breaking. I am getting tired of laboriously using Radio Shack cassette repair kits to get them working again.

Now that CD-R and CD-RW recorders are readily available (one is built into my Macintosh G4 computer), I decided to copy my old tapes onto CDs, where there is nothing mechanical in the recording itself, and various electronics magazines have said that the reliability is of "archival" quality.

To be efficient about it, I tried putting about 6 hours of music on each CD, via MP3 compression. Although I wouldn't exactly say the process of doing this is "intuitive," even with the Mac, once you get the hang of it, the whole thing is not very hard to do. (After various screw-ups, I finally managed to do it right, and then I wrote out very detailed instructions to myself, which made all the later "burns" quite easy.)

The question arises, "Isn't MP3 only good enough for teenagers who have already degraded their hearing by listening to too much loud bass at rock concerts?" Well, I tried using the maximum compression, at 128 kilobits per second, and that did audibly degrade the sound, when playing back classical music. (That's pretty much what the popular magazines said, also.) But at 160 kbps, I could only barely hear any difference when ABing the original versus the MP3 CD, and any loss of quality has been quite tolerable to me. (I did not do any blind testing, and anyhow, this is all pretty subjective, regarding what is "tolerable" to a particular listener.)

At this point I've got a lot of my all-time favorite music efficiently stored on the hard drive of my computer, in addition to being on the 4 CDs that I have made so far. My computer is linked to my main stereo system, so it's easier than ever to play any of this favorite stuff. Although most old CD players can't handle MP3, most new DVD players can, and at least I can play back the music via more than one component. Anyhow, MP3 players are probably going to become more and more common.

Of course, this sort of thing involves very personal choices (convenience versus quality, etc.), so MP3 storage isn't for everybody. But for me, it has turned out to be a useful thing.

--------------------
To download a jazz file that's super-high fidelity, even though it's MP3, click on the colored capitalized word below:
JAZZ
.
For technorati tags: