Interesting Read From a Recording Engineer about EQ


 

On another note, I find it strange how many audiophiles or audio enthusiasts are averse to using EQ to achieve audio bliss with a product that they feel is just a tad too bright or harsh, or needs a little boost or cut in the low end, midrange, etcetera, in order to be ideal for their preferences.

I understand that using EQ doesn't make sense if the headphones, IEMs, or speakers will require A LOT of extensive EQ to get them dialed in. But I find that most products are within a +/- 3dB range (and usually MUCH less) in a certain frequency or two to get them dialed in, which is perfectly safe and achieveable with the majority of products we use.

I also find it a bit ironic that we as listeners don't want to consider using EQ, but if you realized how extensively at least some type of EQ or spectral balance shaping techniques were used in the mixing & production process of 95% of the music that we listen to, you would be shocked.  😛

For multitracked music, strategic EQ and tonality shaping with levels are used by mixing engineers to place the various instruments and vocal(s) withing the soundstage from front to back, i.e. Depth and Layering. EQ is also used to bring vocals forward in the mix (or push them back). IOW, EQ is a Very Powerful Tool in the mixing and mastering engineer's toolbox.

Yet as audiophiles seeking audio nirvana and the best possible performance from our gear, most of us are afraid or averse to using ANY EQ whatsoever. 😕

In the recording studios these days, you will find that nearly all engineers are using and embracing at least some type of monitor speaker/room correction software (which is primarily EQ-based). In addition, several of the companies that design and manufacture highly regarded studio monitors, such as Genelec for one, are incorporating these types of measurement/analyzer/correction systems built into their products, which helps each engineer to achieve their absolute REFERENCE system for their mixing and mastering work.

Again, these are the PROFESSIONALS that are responsible for producing the music that we are listening to on a daily basis. They spend many thousands of dollars on both analog and digital outboard rack-mount EQs or DAW plugin EQs to complete their daily work. 

I first learned the true importance and the amazing power of EQ in my journey as a high-end car audio enthusiast. In a car audio system, there are at least two major factors related to the proper use of EQ that are major obstacles to achieving a tightly focused center image that does not wander or drift dependent on the immediate frequency of the vocal or instrument as the song plays, in addition to achieving a perfectly linear and balanced Left-to-Right Soundstage.

The "Depth To the Stage" (where it begins in front of you), and "Depth Of the Stage" (how deep or forward it projects beyound the speaker positions) is also highly dependent on proper Independent Left & Right EQ, which is absolutely necessary in order to produce a smooth, even, and linear frequency response that corresponds to our preferred Target Curve, AT OUR LISTENING POSITION.

# 1. In our vehicles, our Main Listening Position (MLP) is not perfectly centered between a pair of loudspeakers. This presents a huge problem, or should I say, multiple problems. Just try sitting 3 feet to the Left or Right of Center in your home setup, and also closer to the nearest side speaker as well. The SPL from the closer speakers will be significantly higher, AND, any reflections that you hear from the nearest side speaker will arrive much earlier than the opposite side speaker. I think it's obvious that the results will not be ideal, right?  😛

# 2. In a vehicle, we don't have a properly sized or acoustically treated "room" to play those nice loudspeakers in! The highly reflective near-field environment of a vehicle wreaks havoc on the frequency response at our MLP from our otherwise perfectly flat and neutral high end speaker transducers. And all of the reflections that we hear will be EARLY reflections, meaning there is not enough of a delta in time for us to perceive or differentiate the multitude of reflections from the direct sound coming from the speakers. This produces horrible comb filtering and a very erratic frequency response from our otherwise flat and neutral high-quality speaker drive units. That's not a good start!

# 3. In addition, each of the speakers in our vehicle cannot be ideally placed on a common baffle, and they usually end up being spread around the interior in seemingly random and non-ideal locations within the vehicle. Again, Each Speaker will be affected differently, dependent on its immediate environment or location (nearby reflective boundaries).

# 4. There is another factor that is not so much related to EQ per se, and that is the need for independent "Time Alignment" or digital delay for each speaker so that each one arrives at our off-center listening position in perfect sync, just as they would when sitting in the "sweet spot" equidistant between our home loudspeakers, or headphones/IEMs.

Most high-end car audio systems will have a "front stage" speaker set (equivalent to our front "mains" speakers) consisting of a Left and Right set of Tweeters, Midrange, and Midbass drivers, along with one or more subwoofers that are usually placed somewhere out of the way in the rear of the vehicle.

Because we sit Off-Center to each Left and Right group of speakers, one side is relatively On-Axis to our listening position, while the group of speakers on the opposite (near) side will likely be severely Off-Axis to our listening position. This will cause a massive difference in frequency response between the Left and Right sides, and this, in turn, will destroy any chance of achieving accurate image placement and soundstaging, as well as a pleasing, balanced, realistic and lifelike spectral balance (tonality). 

Try playing a full-range mono Pink Noise track in a non-EQ'd car audio system and quickly adjust the Balance control from full Left to full Right. The massive change in Frequency Response from the Left group of speakers will be COMPLETELY DIFFERENT from the Right group of speakers, and this will be readily apparent at our off-center listening position.

To help in this regard, we try and use a combination of speaker drive units that can play an ideal, particular passband with low distortion, and most importantly, without "beaming". Minimizing any beaming (the narrowing of dispersion as frequency rises) helps the On-Axis and Off-Axis frequency response from each speaker to be much more even and linear. We achieve this by using the appropriate Crossover Network Filters on each driver that protect the drivers while keeping them from playing into a frequency range where they will start to "beam". Ideally, we want to maintain a very even and smooth DIRECTIVITY across all of the drivers. Carefully chosen Crossover Filters are also another way to control the summed or overall Frequency Response...IOW, Crossovers can effectively function as "EQ" as well.

Remember that our off-center listening position in a car makes the On-Axis and Off-Axis Left and Right frequency response quite different. But minimizing beaming and optimizing for smooth directivity helps to even out the differences in FR between the left and right sides as much as possible. HOWEVER, we will still need LOTS of EQ to optimize and balance this Left vs. Right frequency response. This is due to the highly reflective environment of the car. There is really no way around this in a vehicle listening environment. 

Still, EQ can only do so much. Because of the unique interior dimensions of each vehicle, (in addition to obstacles such as large center consoles, transmission humps, the steering wheel, etcetera), there will be Cancellation Nulls at several particular frequencies that cannot be boosted or corrected using EQ. For example, if you try to Boost these cancellation nulls, the speakers will just be working much harder without actually producing any more output at the null frequency, and you risk damaging the speaker due to overexcursion or heat.

HOWEVER, EQ can be used to Reduce PEAKS in the response, and/or to Lower the Peaks On Either Side of the Null, which effectively smooths out the overall frequency response so it becomes less of a distraction or irritation. The SAME EQ technique can be used on Headphones and IEMs. The internal chambers and cavities in the cups of headphones or IEMs can produce cancellation nulls, as well as peaks in response at particular frequencies. Most modern designers do a very good job at minimizing these effects, but there will always be some amount of resonance or null at one or more given frequency. Use EQ to tame these as much as possible.

So in a vehicle, to achieve a lifelike, realistic frequency response that has excellent imaging and a realistic soundstage, we ABSOLUTELY NEED a multi-channel DSP that provides fully INDEPENDENT Left & Right Time Alignment (digital delay), with LOTS of P-EQ bands, and fully adjustable Network Filters/Crossovers for Each Channel. Some such DSP units made for car audio are the miniDSP C-DSP 8x12 & Harmony (with optional DIRAC Live), or various other units made by Audiotec-Fischer Helix or Brax, Audison Forza, Gladen/Mosconi Aerospace, Zapco HDSP-V, etc. miniDSP makes smaller 2-Channel or 4-Channel DSPs that achieve the same goals in home audio systems.

Without using extensive EQ in a mobile audio system, there is simply no way to achieve anything close to our home audio speaker system or our headphones & IEMs. But that same EQ can ALSO be used by us as well as studio engineers to effectively shape the characteristics of the sound to our tonal preferences, and also to improve dynamics, impact, soundstaging, realism, and overall musicality.

AND if nothing else, PLEASE REMEMBER THIS: Using proper EQ will have a MUCH LARGER EFFECT on the "CHARACTER" or Spectral Balance of your Headphones or IEMs than ANY type of "Upgraded", "High-End" CABLE!!! STOP buying expensive cables to "EQ" your goddam Headphones and IEMs!!!

Using EQ effectively is usually MUCH LESS EXPENSIVE, and often times it is FREE, as it's built-in to nearly all DAPs. And EQs or EQ plugins are available for nearly all computer-based music playback software or network streamers.

I'm not saying that Cables Do Not make a difference in SQ. I'm just saying that using EQ effectively is a much more LOGICAL approach, and nearly always less expensive. EQ is also INFINTELY ADJUSTABLE, whereas you will be stuck with the specific characteristics or properties of whatever cable you choose, and it may not be exactly what you were hoping for.

I also have no problem spending good money to choose a "better cable" for it's improved physical or cosmetic properties, such as better construction, less tangling, better comfort, ideal length, the appropriate connectors, and microphonics, etc.

But using CABLES to "EQ" your headphones, IEMs, or speakers makes absolutely no sense as long as the original cables are decent and usable. Engineers in a recording studio DO NOT immediately reach for a different cable when the spectral balance of the sound needs to be altered. They reach for their favorite EQ! 

END OF RANT

128x128lordrootman

IF you decide to use equalization, DO NOT equalize the in-room response of the left and right channels independently. You want the first-arrival sound to be identical from both channels for a variety of reasons, including image stability. Apply the exact same EQ to BOTH channels.

It is totally OK if the left and right speakers' in-room responses (which are dominated by the reflections) are not identical. If you EQ the two channels DIFFERENTLY based on their IN-ROOM responses then you no longer have MATCHING first-arrival sound from the two channels.

fac mentioned the Palette by Mark Levinson. That was indeed a superb unit, VERY intelligently designed. Each knob controlled BOTH channels simultaneously, so the EQ was applied equally to BOTH channels, such that the first-arrival sound from both channels always MATCHED.

Duke

speaker manufacturer

Post removed 

What Duke is speaking about is exactly right.  The challenge of room sound now is this common "solution" of DSP room correction.  It's typically taken from one [mic] position.  Room correction "looks" at what's at the microphone, which is a SUM of everything in the room- direct, reflected, resonances, etc.  They work by introducing EQ to make a change to this sum, as seen at this one microphone position.  Any change these devices make to system EQ will affect BOTH direct and reflected sound and these room correction solutions focus on the sum only.    But the direct sound might be fine- you like your speakers right?  This direct sound may not need correction.  This could be thought of that your speakers sound great but the room doesn't.  The core question here is why change the speaker direct sound based upon the reflections?  Aunt you chasing your tail?  The reflections are still different from the direct because the same room effects are still in play.  What you have done is adjust the sum, which may make it sound better, but might make imaging worse or make other issues worse.   Every room is unique.    

One work around is multiple microphone positions: taking multiple scans or readings of the direct+ reflected sound by moving the mic around and taking multiple snapshots of the "room" and then averaging them all together so the changes are not based upon just one position.  This doesn't solve the direct vs reflected issue, but it does reduce the number of indirect anomalies that might push the correction too far in one direction or another.  Again, we are back to the core weakness of this electrical room correction process of changing the direct sound based upon what the reflected (indirect) sound is doing.  

Changing the room acoustics is the most effective solution, as now you can attack and fix just the indirect sound by changing the surfaces or boundaries that are the source of the indirect sound.   In recording studios, it's the "sound of the room" that makes that studio popular or desirable.    East West in LA has a large tracking room that can hold a small orchestra or large band and was the room Frank Sinatra did "Come Fly with Me" in.  This room sounds amazing and gets booked all the time because everyone loves the sound of it so much. You will not be able to replicate this room anywhere, it is a one of the kind room.  Capitol had two rooms that were unique, musicians loved to record there!  This is the nature of why commercial studios work, the sound of the rooms. 

There are some esoteric exceptions to this set of principles, but for most of us it's an issue of making our rooms sound better to make our speakers in the room sound better.  Duke is right: whatever you do to the direct sound must be consistent.  It may be easiest to understand if the L/R volume is different, the image will shift to the louder channel.

Brad 

One work around is multiple microphone positions: taking multiple scans or readings of the direct+ reflected sound by moving the mic around and taking multiple snapshots of the "room" and then averaging them all together so the changes are not based upon just one position.

As someone who has used this approach at home and in actual theaters I’ve come to the very strong belief that this approach in the average home is bullocks.

The issue has to do, in my mind, is how different speakers intended for wide coverage are from speakers intended for the home. In a motion picture auditorium we need people seated far left to hear speakers in the center and far right. This rarely is the case in homes, and this approach tries to create wide-field speakers from speakers that are not wide-field.  There is also nothing inherent about multi-point measurements that removes room reflection issues very well, at least not as well as close-microphone techniques will.  It just averages them all out. 

I know it's completely against accepted dogma, but my experience in comparing results refutes the dogma.  Measure in one place and use a light touch. 

There’s also the issue of incremental adjustments. Making multiple measurements, adjusting, and repeating is a PITA with a single microphone. It also doesn’t help bass EQ as much as people think. Speaker placement and bass traps do much better to help in room bass than multi-point measurements and applying an averaged EQ. Fix your speaker placement, fix your room acoustics and then EQ’ing your sub from one point can yield most excellent results.

Having also tried ARC systems and compared them to hand tuned EQ settings I’ve come to realize that Floyd Toole’s commentary, that room correction is actually speaker normalizing, and if you want your speakers to be normalized why bother buying any particular brand of speakers??

In the home, I think the best advice is to use the EQ above the bass sparingly, and avoid tuning your speakers to an arbitrary EQ curve, which is what ARC systems try to do. If you have a HT system, use your L and R speakers as the reference in a single point, and match your center and surrounds to them. Leave your L and R alone unless there are glaring EQ anomalies.

Despite Duke’s comments, I’ve had very good results with imaging improvements after making subtle adjustments to match the L and R speakers to each other. Even more so when adjusting a center channel to match the L and R.

Eric: That sure is a good example of "your mileage may vary". While Dukes method is the industry norm, especially in pro, there are exceptions and stories that support those exceptions.  I still think Dukes advice is the "best practice" starting point for most end users. Getting end users to consider the room has more infliuence on the sound of the speaker than nearly anything you hook up to them is not easy to do.