general comments


Today of all days, I am exploring techniques for restoration of old tape recordings. In a very old box of open reel tapes, I found one completely unlabeled tape. It contains a recording of an LP. The tape is half-track, mono. One side is one LP and the other is another. In any case, this gives me a chance to experiment with various restoration tools.

If you even think you know who the Alto Soprano singing this is, let me know! find me at: andrew at noisefloor.org. Thanks!

restoration_sample , Mono  7.9mB, MP3, 192kbps,  Length 5:56

We have a winner! She is likely to be Bidu Sayao (1900-1994), who was a friend of the composer: Heitor Villa-Lobos (1887-1959). The work is
Bachianas brasileiras. It seems likely, due to the age of the source open-reel tape (1955-1965~) that this could be the priemier recording of this work. I am just in amazement. And to think that this tape sat in a box in storage for oh so many years. Thank you Christorpher Purdy! and also, thanks to dear Ray Dawson, a family freind who died so many years ago, and left me many of his tapes. I am most certainly going to listen to them all now.If you have more info – find me at: andrew at noisefloor.org.

It seems I've chosen one of the most difficult types of music to accurately and noiselessly capture. I'm old enough to remember when CDs premiered as a new medium. Many of us had our ears opened when we heard our first high quality CD.

The first thing I noticed was how far down the noise floor was. Prior to CDs, our best (and only) media for music reproduction were vinyl LPs and magnetic tape. At the time, many audiophiles thought that vinyl was superior to nearly all forms of magnetic tape available to consumers.

In the 1980's, a form of noise reduction called CX hit the scene. CX works by encoding a compressed audio signal to the medium. The softest parts of the program are kept at a higher level (above the noise floor) than they would be without CX. That is, the dynamic range of the music is compressed. When the program is played back, the compression is 'undone' by a separate box called an expander. What that does is to effectively lower the volume of the quietest parts of the recorded program compared to the loudest parts. The end result is that the noise of the medium (a vinyl LP, let's say) is rendered much quieter than it would be without CX, and the dynamic range of the source program is essentially retained. The noise is reduced, and the music program is presented unchanged.

Being fascinated with this concept, I bought and built a CX decoder kit. (I still have it, and it still works.) I can confess that it works as advertised. It wasn't flawless, but it worked. During silent passages between cuts on an LP, there was no noise to speak of. This was my first taste of a reproduction method that produced what we today have come to accept as standard. CX didn't do at all well in the market. CDs soon came out, and CX became unheard of outside of the LaserDisk.

With the CD being so new, most audiophiles welcomed it as superior to all that had come before, at least as far as what consumers could lay their hands on. I knew some people in the 80s that thought the high end of CDs was all wrong. "Too harsh" or something like that, they'd say. I thought they were crazy at the time. But now, I am not so sure!

Today, the CD has most certainly come of age. In fact, a CD with a capacity of 800 megabytes ( an hour or so) and an audio encoding of 16 bit depth is considered barely adequate. The industry knows this and has left the CD behind – most people don't know that DVDs have an audio word length of 24 bits.

Which brings me to my thought for today. Over and over again, I have lamented that so many recordings I've made suffer horribly when mastered to 16 bit CD when the program gets extremely quiet. What happens in that case, is that due to the extremely low audio levels, very few of the available 16 bits get used to quantify the signal. What I heard was an Aliased signal. There's nothing musical at all about an aliased signal. I'd rather hear vinyl groove noise with a quiet program than aliased signals. So I always try to make sure that whatever I am recording is modulated as high as possible. Then I have to worry about clipping! There's no way to win on either end; There just isn't enough room for the whole program at 16 bits.

If I was recording pop or rock, This wouldn't be any issue at all. But for me (and for you, too!) it is a serious issue. There is no more dynamic program material commonly recorded than sacred and classical music in churches or music halls. The recording engineer is faced with a program that at times can be as loud as any rock concert (if you don't believe me, then you've never been to a hymn sing with 500 lusty singers and a blisteringly loud pipe organ) Then at times, perhaps at the same event I must record a single person singing at very little more than a whisper from 100 feet away. The temptation to crank up the gain is powerful, but has to be avoided if some dynamic context is to be retained.

As a producer of audio media for my clients, I cannot wait for the 16 bit CD to go away. If you have some attachment to the CD as it is today, may I please suggest that there is something better waiting for you. In the mean time, I am going to suggest to my clients audio DVDs in place of the standard CD of their event.

I have no new noises to post this time. I want to write about what will likely become my new church home. I recently moved to the Columbus, Ohio area. One of the first things I wanted to do when I got here was to find a chuch that had not only an excellent music department, but an important pipe organ, too. My previous post should make it clear that I delight in pipe organs. I want to be as close to them as possible for as long as possible.

The very first church I went to was recommended by a few good friends at my last church home. It is the First Congregational Church, UCC, in downtown Columbus. I visited this wonderful place Sunday. They have two pipe organs, A Beckerath up in the rear loft, and a 1931 Kimball (opus 30324) up front behind a screen. The Kimball was recently completely renovated. Both are large organs, 3 and 4 manuals respectively.

I arrived a little late. I thought they started at 10:30. I must have confused them with another church I wanted to visit. Went I walked in the rear entrance, I heard a soft, distant rumbling noise, a noise that is unmistakable to a serious pipe organ enthusiast. No, it wasn't music. I followed the sound down a hallway. Louder. Down a flight of stairs. Louder. Down another short flight into the sub-basement. Very loud. Around a corner and through an open door into a small pitch dark room. Roaring loud! Flick the light switch and there it is: the massive Spencer blower for the Kimball organ. What a way to meet a pipe organ, first by its blower. It was immaculately clean. This organ has been renovated recently.

The Beckerath has a feature I haven't seen before. Each draw knob is sequentially numbered. Perhaps there is a combination action that is programmed with stop numbers.

There were at least 300 people there. Scott Hayes from Muller organs, also music director at First Baptist Church in Springfield, played organ. (Tim Smith the Minister of Music and Artistic Director was on vacation). Scott's prelude was "Alleluyas" by S. Preston.

We were treated to a few truly fascinating pieces played by a saxophone quartet led by Kelley Gilbert. Handel's "Sarabanda" and for the postlude (where everyone is sitting and listening instead of getting up and leaving) a refreshingly fun piece called "Gaguenardise" by J. Francaix. They were flawless in every respect. One doesn't hear near enough saxophone quartets.

I am delighted to learn that OSU records and re-broadcasts recitals and performances there on the WOSU stations. Their Lessons and Carols service is broadcast live.

If I'm fortunate, I'll have a small, somehow noise-related audio clip to post regarding this amazing church soon.