Inside Audio: Dave Moulton

THE GOOD, THE BAD AND THE UGLY

The Brave New World

A couple of years back, I got an irate e-mail from a guy in northern California who calls himself Bro Duke. He took me to task for worrying about Surround Sound, saying, "As far as I'm concerned, the TV medium hasn't even begun to manage mono properly. How is it going to contend with multichannel?" So, I did a column on this very topic (TV Sound: Mean Streets? TV Technology, November 30, 1998).

Well, Bro Duke is still on the warpath. He's really unhappy with the quality of much broadcast audio, and he thinks we (TV Technology and the FCC, ostensibly) should do something about it. Although I'm a little uncomfortable with the notion of deputizing the Bad Audio Police – particularly as a division of either TV Technology or, even more so, the FCC – I also am not all that keen on the quality of sound I hear from either TV or radio sources.

So, when Bro Duke threatened to send me a CD documenting said lousy sound, I took him at his word, and promised to have a listen, do a column, even rant if necessary.

DUKE’S CD

Well, Bro Duke sent me a CD all right, and it's pretty interesting. He has examples of "bad audio" and "good audio," plus some commentary on each. His sources are mostly broadcast, but he also included several pop record re-issues, comparing them with either the original or an alternative reissue.

Now, the first thing to note is that some of the "bad" examples are indeed pretty dreadful. As in "what could they have possibly been thinking, drinking, smoking" bad.

The second thing to note is that many of the examples sent by Bro Duke occur either in reissue or rebroadcast. The original work may have sounded pretty decent, but then something really bad happened when the work was rebroadcast or reissued.

Bro Duke takes the general position that we did better in the good ol' days. We didn't have all these processors we didn't know how to use, and as a result, we did a better job – necessarily working with less. In other words, the then obligatory "Keep It Simple, Stupid" syndrome kept us out of trouble. And Duke's CD, not surprisingly, tends to support that viewpoint. Let's take a look – er, listen.

A QUICK LISTEN TO BAD AUDIO

The first two examples – both record reissues – displayed some really ham-fisted engineering. One was a major label reissue of a pop guitar instrumental in which the engineer, along with doing some massive re-equalization, tried to REMOVE THE REVERB from the guitar solo!

It's awful. The second reissue had a combination of gating and compression that managed to just about wreck the recording. Duke then shared with us a Time-Life infomercial for a compilation CD of soul recordings. The CD sounded like it was using a ducker backwards, but it was probably actually a soft squelch circuit. The resulting effect is that when the voice-over stops, the level of the music bed gets turned down, erratically.

Following that were some inexplicable examples of reruns of Jay Leno. These sounded to me like a poorly adjusted Burwen dynamic noise filter (remember those?) was involved.

When the level of Leno's voice dropped, a low-pass filter seemed to sweep down, roll off the top-end and further reduce intelligibility. Why? Why? Why in the world would somebody do this?

Another infomercial, with terrible "phone-quality" voice-overs, followed. This stuff is just plain hard to listen to!

Harsh, compressed, midrangy, no top, no bottom, just squashed hard little voices barking at us – forever, it seemed. Egad!

OTHER SIDE OF THE COIN

Then Duke treated us to some "good" commercials, some "good" Johnny Carson reruns, some "good" Leno and some "good" Conan. (I think Duke likes to watch TV late at night.)

Now, with all due respect to Duke, who really IS the guy wading through all the crud here, the "good" examples aren't all that hot either. Duke liked the Carson examples, but I found the upper midrange boost really irritating and the compression on the audience made me feel like I was being bludgeoned by a moderately firm 2 x 4.

Duke liked Carson's voice and the quality of room tone (which is much more "live" than more contemporary broadcast techniques). He attributed this to the use of a desk mic (Carson used an AKG 414 during the latter part of his career, if I recall correctly), plus booms rather than lavalieres on the guests and talent.

SOME OBSERVATIONS AND COMMENTS

None of the problems presented here had anything to do with stereo versus mono (only one example exhibited any stereo-ness, and it was pretty benign).

There were some serious problems with spectrum, along with some massive distortion, some mix balance problems, etc. In addition, there was some real ambivalence in the examples about ambience (a point we will discuss next month). But the big, big problem that occurred over and over in these examples was in level management, the bad use of AGC, gates, squelch and compression, often in combination.

Now, compressors and related gain-regulating circuits are among the most difficult devices to "hear out" and learn to use. Moreover, their behavior is often counterintuitive and sometimes yields artifacts that are quite brutal, as well as entirely unexpected and unintended.

This is particularly so in systems and chains of systems where we end up with multiple gain-regulating stages in series. It is a common occurrence in broadcasting, as the audio signal goes tripping down the production staircase from studio to control room to post-suite to distribution to satellite to cable/regional station to transmitter to couch potato.

Partly, of course, the poor quality is due to simple incompetence. There are people dealing with audio who simply don't know (and may not care) what they are doing. It is an article of faith that audio is largely forgotten in the TV production and transmission world, and it shows.

Another reason for the poor quality is simple inattention. Gain regulation that worked fine for some program material may not work for other material. The price of quality is eternal vigilance, like the Duke said.

A third reason is poor or incomplete training. The poor schnook that tried to remove the reverb from a guitar solo whose quality was based on that reverb, apparently lacked the musical training necessary to recognize that inherent production value. The result was inappropriate production goals.

There's more to this, however. Next month, I'll discuss some of the other issues surrounding these misadventures, plus my own look at the current state of audio coming out of the tube.

Thanks for listening.

Dave Moulton usually falls asleep before Jay Leno comes on, but he'll try to stay awake and check it out! You can complain to him about this or anything else at www.moultonlabs.com.