Audio Class With Professor Hoffman:
Lesson 1: Equalization

While you certainly like your job and don’t want to train your replacement or give away your secrets…any tips for helping us improve the sound on our on stereos—especially for those recordings DCC has not been able to license and let you fix?

Pretty hard to answer. It's like asking to be taught how to drive, just a little. 

Well, in general, the first thing one should do, is find out how accurate their stereo is to begin with. In my college radio station days, we used to use pink noise, feed it through the speakers and then use a 12 band 1/3 octave graphic EQ to make it sound "flat" to the ears. How did we get the pink noise? Well, we patched in an FM Radio, usually a McIntosh or Marantz, and turned off the AFC. This meant that, when you detuned a station you got that nice static; a poor man's pink noise generator. Then we would EQ, so our (Los Angeles TV Station hand-me down) speakers sounded as best as they could. Try it. You can also use the Stereophile Magazine Test CD. It has pink noise and a bunch of other helpful stuff. The "I don't have an FM and I don't want to buy a Stereophile CD" way to do this, is to put a record on your turntable, and EQ so that the surface noise sounds accurate, and not too "pingy", or "bassy". Stupid, but it usually works!

Don't be too shocked at how your system sounds playing back pink noise. This pink noise can show off major flaws in your playback that you didn't even know about. Also notice how just a "touch" of graphic EQ in or out can radically change the sound....

So, we use graphic EQ's to tailor the sound of our playback systems, and we use Parametric EQ's to tailor the sound of our music. Right? Now, you have a good point in which to start. Download (if you can) some parametric EQ programs. A real live Sontec or Massenberg Parametric EQ unit can set you back 9 grand....

Parametric theories in actual english.

The first one is "everything in moderation". Just a touch will do.

The second one is "never add, just subtract what is there already". Most of the CD's you will be trying to fix (especially the new post 1998 "remasters") will need to have stuff removed for them to sound acceptable to your new Audiophile Ears. Some of the DCC Gold CD's might be the exception to this rule, though. You might wish to add a touch of something if you like it a bit "crisper". It's a free country.

EQ pressure points, as originally taught to be by Kevin Gray: Ignoring Bass fixing for the moment, they are....

Lower mid (1000 cycles)

Mid (3400 cycles)

Upper mid (6000 cycles)

Lower treble (8000 cycles)

top (10,000 cycles)

Air (to kill all flying insects (14,000 cycles)

By just using these "EQ pressure points" and a wide curve to subtract some points, one can pretty much fix almost any bad sounding music.) These are pretty much the points that naughty mastering (and mixing) engineers tweak the sound with to give you a headache. Simply by reducing any combination of these points by 1/2 db or so, can fix up many CD's.

Bass EQ points:

Low (40 cycles)

Mid (80 cycles)

Lower cloud (150 cycles)

Upper cloud (250 cycles)

The lower and upper cloud areas are usually shaved off by most mastering engineers to give everything that modern "synth bass" sound. It also robs the music of warmth, and the singer (male and female) of "balls".

Well class, any questions so far? I can write more if anyone is interested...

There will be a test on this later.

In response to many private Emails, I must mention here that what a Parametric EQ does (as invented by George Massenberg) is to be able to dial in certain frequencies, and deal with just them. But, you have to know what to dial in. If you use the "pressure points" I mentioned in my earlier post as a starting place, and simply tweak in and out a little EQ at that point, you will be able to sense right away if it is helping or hurting your sound. Remember, only a db at most!

One db is measured as the LEAST amount of "change" in sound that the human ear can detect. Of course, many of us can notice much smaller changes, 1/4 and 1/2db and so on, but the "decibel" is the starting place for measuring. Get it? So, if I say that on John Fogerty's song "Centerfield" you might need to dial in 10,000 cycles on your Parametric with a broad slope (marked as "1") and:

"take out 3 db at 10k", you will know what I mean...

Have fun fixing stuff!  

Steve: Thanks for the "Lesson 1 - Equalization" interview very useful and informative. Now I know exactly where to nudge to get a little CD warmth when needed and finally understand the "cloud." A couple of further questions:-

If you can say a few more words about "air" am not sure I really have a good sense of what that is, or, sounds like.

What frequencies on a 1/3 octave graphic eq. would you tweak down a hair to remove harshness that shows up on some CD's?

Same question with regard to CD's that are overly bright - what HZ would you routinely try shaving - or is brightness a lost cause, or simply a matter for the old treble control on the amp as the best that can be done.

Very quickly, "air" is the area on a mastering EQ deck that is in the 12,000 to 16,000 cycle range. "Shimmer" is another word.

I NEVER add air. Really bugs me when engineers do. Brings the top end of the tape hiss up and makes my ears crazy. Almost every JVC XRCD does this. Typical Japanese style mastering trick. They LOVE air. Some of you probably do, too. Fine with me. I can always detect it and it bugs me. Sort of unnatural..

On a 1/3 octave, it is pretty hard to remove harshness. The bands are not broad enough. But, try in the 6-8k range. Just a tiny bit, though. Never over do it!

Overly bright? Well, on your graphic, try in the 8 to 12k range. Usually does the trick.

Get yourself one of those funky BSR cheapo five band graphics from the 1970's. Probably find one on Ebay for next to nothing.

Much easier to use, broader, more flexible. Don't tell any of my audiophile buddies that you have that in the system though....
 

***Something to be considered is that Steve only used analog equalization. He NEVER uses plugins on things like CoolEdit Pro. Only outboard, professional level analog EQ. An explanation of why comes from the Stevehoffman.tv forum.

Grant
Steve: What's the difference between analog and digital EQ? 
You have often stated that you don't like the sound of digital EQ. Why is that, and what are the differences between that and analog EQ?

 I've never heard a good sounding digital EQ. The new Sony Oxford plug in system is supposed to be great sounding, but then again it's trying hard to imitate an analog EQ sound. Since I work in the analog domain, I don't need the thing to imitate anything, I just use the real deal. 

My buddy at Sony sent me this about digital EQ; it's just so hard for digital EQ's to sound pleasing: 


There are many ways in which errors can crop up in digital EQ and there has been much research and publication of the subject. Many different architectures (algorithms) have been proposed to optimize the situation and they all have advantages and disadvantages depending on the type of EQ and processor intended. In all cases these different algorithms are compromises that trade-off one undesirable effect against another. 

Another issue that has differentiated the sound of digital EQs from their analogue counterparts is HF response cramping. This phenomenon occurs when EQ curves approach the HF area closest to the half sampling frequency (Nyquist frequency) and manifests itself as an increase in the steepness of the EQ curve at the upper most part of the response. 

The effect of the cramping is to reduce the HF content of the EQ curve, restricting the openness of the sound and adding to the effect of harshness due to the predominance of mid frequency action within the unbalanced EQ curve. 

Since this effect is related the closeness to the intended response to the Nyquist frequency, the problem is greatly reduced if the system is run at 2FS (88.2KHz or 96KHz) and this may be part of the reason why over sampled systems are often preferred. But since over sampling the entire system will halve the processing capability of the hardware, this solution is costly in a workstation environment where processing power is at a premium. Some digital EQ designs address this problem by up sampling before the EQ and then down sampling at the output to match the system-sampling rate. But although this is more cost effective than running the whole system at 2FS, the up sampling and down sampling processes are themselves a possible cause of error and quality loss. 

By employing novel coefficient generation techniques, the Sony Oxford EQ plug-in produces a fully de-cramped and symmetrical EQ response without resorting to inefficient or error prone over sampling techniques. In fact the method also allows the EQ to simulate the responses of an analogue EQ with the centre frequencies above the Nyquist frequency (i.e. 26KHz for the GML option), all at normal base band sampling rates without any change to the performance of the rest of the system.