Return to EQ index

All information in these pages is copyright (c) 1989-2003 by Roger Nichols. All rights reserved. Permission for personal reference only, and may not be reproduced by any method without written permission.

Behind Door "A"

By Roger Nichols

For The Sake of Comparison

It is possible to hear something and tell whether or not you like it. We do it all of the time. What you can’t do is tell whether or not what you are listening to is the best representation of the original.

If you are recording in the studio you put down a take and listen to the playback to decide if everything is in order. You can determine whether all of the musicians played their part correctly by listening to the playback. You can determine if all of the instruments are in tune by listening to the playback. You can tell if the tempo is correct. You can tell if the groove is what you were looking for. You can also tell if the sound of the song as a whole is good enough to go on the finished CD. What you cannot tell is if the sound you hear on playback is the same as the sound that was heard in the control room during the recording.

The only way to determine if what was recorded is the same as what was played is to compare the two directly. To do this comparison, you need the band to play along with the recorded material. You can then switch between then live material and what is coming back from the recorder. To do this a/b comparison correctly you have to switch sources at the recording console. You have to switch from monitoring the live console outputs to monitoring the tape playback. If you do the comparison by switching the recorder from playback to input monitoring, you will still be hearing the converters and electronics of the recording device, which ads to the coloration of the sound.

If you are still recording analog, you will hear the biggest difference. I am not talking about whether you like the sound of analog or not, but whether the input sounds like the tape playback. They are not the same, and that is just the law of physics. Digital recording does not necessarily sound exactly the same as the input signal, but it is measurably closer to the input than analog recording.

Every process does something to the sound. Some of the processes are done to make the sound more pleasing to the artist, engineer, or producer. The process can improve the sound or make it sound worse. These processes include mic choice, mic position, instrument choice, instrument placement, instrument tuning, player performance, guitar amp settings, drum head mutes, and literally hundreds of other variables that color the sound on the way to the recording machine. The recording method is one of those variables.

Shootouts

Whenever a bunch of equipment gets accumulated in one place there will inevitably be a shootout over which piece of equipment is better than all of the other pieces of equipment. I was involved in one of these shootouts in Nashville a few years ago involving a swarm of digital converters. All of the converters were calibrated exactly so that there would be no level change when switching between them. As a control, a straight wire connection was one of the choices.

About a dozen engineers showed up to evaluate the converters. The source material was fed into the conversion network. The source material varied between live acoustic guitar, live vocal, multi-track analog and multi-track digital material mixed through an SSL with Ultimation. Remember that one of the choices was the original signal right from the console. The other seven choices were the six converters and the straight wire. Only two of us picked the straight wire as sounding the most like the original. Everyone else picked converter #3. They said that they liked converter #3 because it made their mixes sound better than what was coming directly out of the console. They said… IT MADE THEIR MIXES SOUND BETTER THAN WHAT CAME OUT OF THE CONSOLE.

Well, that was not exactly what I was looking for. I wanted a converter that sounded just like the straight wire. I wanted a converter that did nothing of its own to my mixes. I wanted to mix by listening to the output of the console, print the mix to the mix machine, and when I played the mix back, I wanted to hear exactly the same thing. If I wanted anything different, I would change it in the mix, or do it in mastering.

I want the same thing in my multi-track recording also. I don’t want the recorder to do anything to my recording. If I want an analog sound on the drums, I will record them onto an analog machine and then put them back. If I want the hi-hat brighter, I will record it brighter. If I want some compression on the vocal, I will compress the vocal.

During the shootout, if I had not had the original for the comparison, I probably would have selected the same converter that everyone else selected. It sounded good. If there was no way to tell what it was supposed to sound like, then converter #3 would have been the clear winner as the best converter of the shootout. Only with a comparison of the original source would we know whether the converter was doing a transparent job, or not.

CD Formats

This leads me to the latest confrontation on the technology forefront. I was in a high end audio store the other day and there were a bunch of guys comparing the new Sony DSD (Direct Stream Digital) with a 96k 24bit CD player. They were arguing over which was doing the best job of reproducing what was on the CD. Of course they were not comparing the same recording mixed in both formats. They were comparing two completely different performances by two completely different artists. How could you make a determination about which was the better vehicle for the final audio delivery system?

The only way to ever find out which is better is to record a live to stereo session that was encoded into both systems, DSD and 96k 24bit. The only way to truly compare the two systems would be to record in both formats, and play them back while the live band played along. Then do an A/B/C comparison to see which one sounds the most like the original, not, which of the three sounds the best.

Now that I have beaten this horse to death, how do I know that the horse is really dead and not just sleeping? I need a sleeping horse and a dead horse for comparison. Or do I need the original horse before he was dead. Or what if he died while sleeping? Talking about this as making me hoarse. How many bits in a horse’s mouth? I remember when cowboy movies were two bits on Saturday.