Monthly Archives: July 2011

How HAECO-CSG Kills The Music

“Wrecked,” as one audio engineer calls it.  That is what HAECO-CSG does to music.  What is it, and why was it so bad for recorded music? To understand the process, you have to turn the clock back to the late 60s, when both monaural and sterophonic playback equipment were commonplace in many homes back in the late 1960s.  In some ways, the two were compatible, but in others, they weren’t.  HAECO-CSG attempted to cure that problem.  Little did anyone realize what a sonic mess it made out of recordings.  How could something with good intentions lead to such bad sound?

You have to consider what happens when you play a stereo recording, summed to mono.  The left and right signals stay the same volume, but since music more in the center of the soundstage were present in both channels, they could be as much as three decibels higher than the far left/right signals, which would throw off the balance (the “mix”) of the recording.

Rather than issue two separate versions of a recording as had been done in the past, Howard Holzer, A&M Records’ chief engineer in Los Angeles, created a system that would electrically alter the recording so that when the stereo recording was “folded down” to mono, the balance would be mostly preserved.   HAECO was the Holzer Audio Engineering Company, and CSG was the Compatible Stereo Generator. Mission accomplished?

Not quite.  While a casual and non-critical listener may never hear a difference, the end result is an effect where the stereo soundstage is smeared.  To give an example, let’s use an example of a human, male voice.  Say, Sergio Mendes, on the track “When Summer Turns To Snow” from the Fool On The Hill album, one on which CSG was used to master the album.  A human voice consists of the fundamental frequency (the pitch of the voice…Sergio sings in a baritone), and sibilants (or “formants”), which are like the rasp of the vocal cords, the whistle of air between the teeth, or other high frequency components that are not the main pitch.

Normally in a stereo recording, you can pinpoint the voice by both the formants and the fundamental frequency coming from the exact same spot in the soundstage.  Not so with CSG.  What happens is that the image is smeared.  The formants can be pinpointed, but the fundamental frequency is smeared across the soundstage in a “phasey” sort of way.  The sound also has more of an overly-full presentation to it.  The net effect of CSG with a voice like Sergio’s, and the rest of the music, is almost the same kind of phasey effect you get with the “fake” stereo that was also popular at the time.

The real problem, today, is that many recordings were mixed to two-channel stereo with the CSG processor in the chain, so no two-channel tape exists without the CSG processing.  The only way to properly undo the CSG effect is to remix from the original multitrack master tapes…if they even still existed.  Many CDs have been reissued over the years that contain the CSG processing.  They sound about as good as

There is another fix, and I will outline this in my next installment.  Stay tuned.