Eq limitations

December 24, 2021
 by Paul McGowan

Equalization is the act of increasing or decreasing the amplitude (loudness) of specific ranges of frequencies.

The most common form of EQ was once the ubiquitous bass and treble controls found on consumer audio equipment. These knobs or sliders allowed one to reduce or increase the amounts of both frequency ranges. (in my experience they were almost always cranked up on high).

Today, a few brave souls are using EQ to help adjust low frequencies anomalies in the room—mostly with the addition of a low cost easy to use DSP product like the MiniDSP. For $100 this device can take the analog output from your preamplifier and be used to feed a separate bass amplifier in a bi-amped system. Once inserted between the preamp and woofer amplifier, a simple and intuitive user interface can be used to adjust the bass response of your woofer.

Any loudspeaker with separate binding posts for woofer and tweeter can enjoy the benefits of biamping and EQ control using this simple setup.

While I wouldn't recommend trying this for anything other than a woofer, it can be helpful in smoothing out the peaks and bumps caused by room modes and standing waves.

While peaks and bumps are easy to eliminate fixing the dips and valleys caused by the room are almost always impossible to fix. That's because the loss of bass frequencies in a room are caused by cancellation (just as peaks and bumps are caused by addition). And unfortunately, regardless of how much more loudness you pump into the system at those frequencies, the cancellations just keep doing their thing.

So, should you go down this EQ road, just be mindful the most valuable improvement you can hope for is the reduction of peak and boom, not the pickup of missing bass.

Subscribe to Paul's Posts

44 comments on “Eq limitations”

  1. MiniDSP is pretty standard for subwoofers. For streaming you can do this sort of equalisation in Roon. An advantage of integrated systems is that they can incorporate DSP, like DIRAC. Devialet has a thing called Sweet Room, where you just create a set of adjustments in a .txt file (each with frequency, ±gain and Q factor) and copy the file to the system SD card. You can do two sets and separate left and right, and switch between them on the fly from the remote, or turn off the adjustments completely.

    I agree, you don't want to try adding to the room and personally I would only use this stuff if you have a serious problem.

  2. Back to the room being the main culprit. Or maybe the bane of the audio system listener?

    I would guess if one is going to jump into the deep end then it would make the most sense to pick up a good mic and software and find the peaks and valley’s 1st, along with the suck outs and the additives. At least then you’ll know what you have. (An acoustic MRI if you will)

    From there you can start to figure out what things you have control of.

    Otherwise let some algorithms do it for you and be satisfied.

  3. So the only way to fix those dips is experimenting with alternative speaker and listening positions!? Or to work with well placed and dsp-controlled subs!?

    Merry Christmas from Cologne, Germany.

    1. Moving the listening position as well as the subwoofers is the only practical means of reducing the dips. There are means of building traps that can absorb much of the unwanted bass that can in some cases cause the dips, but that's hardly a practical solution.

    2. As Paul suggests, the dips are caused by TOO MUCH BASS reflections. The same phenomenon happens at higher frequencies where it is usually referred to as "comb filtering". Ethan Winer writes extensively on that part of the spectrum.

      This makes it difficult to reduce bass holes in the room instead of merely shifting them around by bass speaker and ear placement. I am a certified room tuner, and the best I can do for heavy walled, featureless rectangular boxes is tune for a single seat - which goes against my principle that music should be shared.

      To flatten a listening area for friends & family you have to change the architectural acoustics, and for bass that takes big changes - a lot of surface area, a lot of volume, and/or large diffusion features. For example, a bass diffusor needs to be at least 1/4 wavelength deep, or 1.7m for 50Hz. I have done this successfully in a basement room that already had a fireplace corner alcove, closet, and recessed entrance that were 1.2m deep and on the same wall, adding furniture that was 50cm deep to approximate a Gaussian diffusor shape. Note that this extended the entire long wall of the listening room (8m), and floor to ceiling.

      For my residence and live recording studio/venues, I always use bass leakage. You need large areas of windows, skylights, open doorways and open plan architecture including arches and stairways. Sprung floor and ceiling help, especially if they are just subfloor and flooring with resilient connection, open beam with no sheetrock or plaster.

      It is possible to trap the most egregious spatial resonances with membrane absorbers, but they have to be tuned by room measurements and it still takes a lot of surface area, like a half wavelength square - so 12m2 for the 50Hz example, and it has to optimized in location as well. Once again, I suggest a well instrumented professional.

      A high tech approach is to combine three or more subwoofers using placement and DSP to fill in the holes. This takes yet more expertise, and there is a tradeoff of phase cancellation if the subwoofers are more than 1/2 wavelength from the treble speakers at the crossover frequency.

      Note that in my professional experience building and installing sound systems, there have only been two cases where the optimum placement of treble speakers was the same as the optimum placement of bass speakers unless the room was a gut re-model designed by a professional acoustician and not subject to any architect or interior design oversight.

      I have worked on 100+ gut re-models and new construction for residences over US$5M, and only one architect understood acoustics well enough to make a music room. I made interior designers cry when I told them what good sound required. Look your partner in the eyes and ask, "How much do you love me?"

      1. acuvox,
        Once again, interesting stuff & thanks for posting.
        Yes, I agree that music should be shared; if only people
        could learn to SHUT UP when the music is playing.
        Women, in particular, seem to want to start a new topic
        for discussion exactly when the guitar solos start...WHY?
        Under the above-mentioned circumstances all that is
        required is that the frequency range be as flat as is
        possible right in the single-seated 'sweet-spot'.
        All the very best to you & yours for 2022.
        Cheers,
        Martin

        1. I recall hearing about a study that showed most people only recognize songs by the lyrics - they can't identify instrumental versions of pop hits. Of course they consider the guitar solo interruptible - there are no words!

          My research with conservatory trained musicians (the only people who still know what music sounds like) shows that you need flat phase and waveform fidelity at the sweet spot, not merely flat frequency response. It is possible to have flat frequency and rotating phase (all pass filter), but if phase is flat, frequency must be flat too.

          1. acuvox,
            Usually, always it's the lyrics that are
            the emotional hook for me in a song.
            However, there are many guitar solos
            that will emotionally out-rank the lyrics
            of other songs, & to talk over them is
            just totally blasphemous.

  4. Certainly agree with Paul’s comments above when put in context. I switched, however, from never using tone controls and an almost disgust for equalisers of the 80’s and on…

    So, in my experience even going back to the early 1970’s with quite humble kit my tone controls were left level, as their use always seemed to also have an adverse affect. As I improved my kit over the decades this continued, I guess I lived with a few room issues… then used a bit of corner bass acoustic measures, as the kit got more powerful.

    Now I’m using a $10,000 AVC designed and built in Japan in limited numbers, and apparently using some of the best discreet components as found in PS Audio kit. It has room equalisation that I limit to below 500Hz and have found I can get better sound while moving full’ish range tower speakers, almost as high/wide as FR30’s if not as deep, closer to walls/corners. Still with a bit of room acoustic correction I’m getting the best sound I’ve heard and got a bit more useful room space, important as like most folk I don’t have a separate ‘music room’.

    I don’t consider myself brave, as Paul puts it, more willing to use kit designed by folk that listen to music and then tweak their high end kit even further, all based on how it affects the sound. Which is also why I follow Paul’s musings as I know that’s what he and the PS Audio designers obviously do… Maybe one day I’ll afford the full kit now available from them, I always like to aspire to such quality - but like a house in its own grounds, Tesla car, high end cameras and very few other areas, for many of us they are and will remain aspirations…

    Merry Christmas everyone - or Happy Holidays if more acceptable… 🙂

  5. Mike makes a very good point, if you're serious about your audio you should definitely buy a microphone and a spectrum analyzer. Even the low cost ones do a pretty good job. It's a real eye opener to mic your room and find it's not just low frequency that is affected by room reflections. At higher frequencies moving the mic as little as 3 or 4 inches can give dramatically different amplitude readings. Whether you want to call it comb filtering or constructive/destructive waveform interference room reflections can make even the best equipment sound awful.

    Merry Christmas and Happy Holidays to all!

    OHT

    1. You can get RTA (spectrum analysis) apps on your phone, but you really need phase response, transient response and polar pattern to start teasing out the problems.

  6. pre amp, 2 mono blocks, cd transport, dac, streamer, turntable, cassette deck, subwoofers, main speakers...
    And as if this were not enough now we have to add another device, this DSP thing ?
    Man, life of an audiophile is not easy (and certainly cluttered in the living room !).
    When does it ever stop.
    Maybe SntbcwS was right with his one box pizza solution.
    Maybe it's time to acknowledge that he is the only visionary on this site.

    Now, about yesterday Fat Rat, the cricket thing.
    I was not talking about population but about countries. And when I stated that about 95% of the countries don't have a clue about cricket that is pretty much correct, give or take a few percent.
    That some of those cricket countries happen to be very densely populated (e.g. India) does not take anything away from that 95%.
    And when I wrote "let Brits and Australians have their own little party", obviously I was referring to the 2 persons on this site talking about cricket on a (more or less) regular basis.
    The Australian is you, the Brit is the pizza man. I thought that was easy to understand. I was wrong.
    So, given the 95%, I don't consider cricket to be a world sport.
    Do you think anyone on this planet considers Chinese to be a world language ? Don't think so.
    And yet, 1.4 BILLION people speak Chinese. Much, much more than people who speak English.
    Think about that.

  7. Off topic, but bear with me on this.

    I ordered a cd, btw, nothing to do with PSA or Octave, from the USA on 29/11. It happens to originate from Denver and arrived at their mail centre on 2/12, but it didn’t get to L.A. till 19/12. How so? Did they send it by Little Donkey? Just discovered it arrived London Heathrow on 23/12 but I don’t expect to see it before Christmas 🙁 Thought that might be of interest to anyone waiting for mail.

    Now, according to the lyrics of a song, “It’s a long way from L.A. to Denver”. The song being “Starwood in Aspen” by John Denver, but it just made me wonder, you know, just saying.
    The almost on topic bit. Is there any link with the title of that song and the naming of the new speaker series? Is PSA populated by fans of the talented musician?

    Compliments of the season to all.

    1. Our postal service seems to have lost a beat...or 3. On the third we sent our daughter in Washington State a package, It went from Mississippi to Mobile Al, to Birmingham Al, to Atlanta GA, and arrived in Memphis TN on the 12th where it sat 6 days. It finally arrived in Seattle on the 22nd. Maybe we should bring back the Pony Express.

  8. It’s hard for me to visualize when this will ever stop.I agree with you once again jb4.

    I would need to be an ultra wealthy person to start from scratch by hiring a a great Audio Consultant to design and oversee the construction of my listening room and install the system of my choosing. Finally, the consultant tunes the room to the equipment and my satisfaction. A good friend of mine in New York, Frank Huang would be my choice to do this job when I hit Mega Millions or Power Ball. It’s going to happen next week… I can feel it in my bones. Yeah…right :-/

      1. jb4

        I already had a personal listening room when I lived in New York. I had contractors in use double 5/8 inch sheet rock with some kind of material in between. I had bookshelves made that held my vinyl records books and other effusive material, a couch on the other side wall and a large area rug between the speakers in the listening chair. The room dimensions we’re pretty close to the golden ratio and the only thing that this was used for was my listening pleasure and entertaining guests that visited.

        I have my current system set as best I can right now and I am not going to start using new technologies that may only make things worse Since the system sounds very good at present. i’m not looking for perfection only quality listening experiences.

        That’s why I stopped being a critical listener for the most part and just sit back now and let the music come into my soul. if you don’t already do this give it a try, it makes life much more pleasurable.

  9. If you want to do this properly, you need to get a processor from a Swiss company called Illusonic. They cost a fortune, they come and do the calibration, but it is the last word in room correction.

  10. The problem with using EQ to fix room modes is that EQ can only correct frequency domain problems. Room modes are also time domain problems - the room will "ring" at its modal frequencies. The only real way to fix room modes is with proper room design and acoustic treatments. EQ can certainly help, but it's no replacement for proper room design.

    1. TIME correction is essentian: chopping the initial waveform to protect against deferred bass ringing.., FAIL

      DSPeaker from Finland does bass TIME adjustments. Winner

  11. Altec had a system they called “Acoustavoicing”. It had a 1/3 octave equalizer as default (24 band).
    They worked with local subs to set it up with tools like pistols, claps, mics and pink noise loud enough to really load the room.
    Once set, you could opt to have a panel lock-out on the EQ, but it became an integral part of the system.
    The feed from the mixer was sent to the EQ and power amps attached to the speakers.

    If further issue needed addressing, they offered room treatments, but by then, most customers were tapped out.

    In my 20s I picked up some Altec Santiago speakers (15” woofer, horn mid/high)
    These were nice walnut sealed cabs with the amp internal.
    I fed them with an SAE pre with an internal 5 band EQ.

    All my home or mobile gear has been bi or tri-amped since then.
    Today I use no EQ, other than the integration of a sub with 3-way speakers.

  12. Two things. MiniDSP digitizes the signal. Or you can use it as your DAC plus for the EQ and crossover. They have some models that can do that brilliantly, like the SHD.
    Second. There is a lot of software that you can use for EQ. Many software packages take the signal from REW readings to generate the proper modifications.

    Or you can get speakers that “self-correct”.

    However, some very low frequencies are hard to adjust, as Paul said. But you can achieve quite a lot! Using a UMIK microphone plus REW is essential to objectively improve sound in a room. Your ears aren’t good enough.

    Or you hire Mitch from accuratesound.ca that can do it for you remotely.

    Anyway, Merry Christmas to those that celebrate it.

        1. Already on it. There's nights I just wanna listen and then some nights you just want to tweak the bajeezuz out of things... ;-). Is it a listening session or an adjustment setup improvement session..? One is beer, one is Scotch...
          Because that's half the fun of having this cool gear.

  13. I wish I could say I don't need/want this stuff, but some of my favorite recordings sound like they were mastered by a guy with a deep sea diving helmet over his head. And lets face it - if you NEED EQ on a particular track/album, the addition of another piece of gear isn't going to take anything away...
    I use a DBX PA2 EQ with active X-over, room EQ with mic, balanced inputs/outputs, driver delay, tons of memory settings for different recordings and my favorite - with the app if I NEED EQ it has a draw feature - you can sweep your finger along the screen left to right (low to high) to quickly adjust EQ. It is also switchable to full range or whichever subwoofer crossover setting you prefer. This way I can run my mags full range with the REL on great recordings or for rockier stuff cross them at 50 60 or 75 Hz to a different sub to eliminate overdriving the panels. The autoEQ wasn't all that great but I only tried it once. But the on-the-fly quick draw EQ is gold when you need it.

  14. It would seem the MiniDSP would be a much more simple approach if it were incorporated directly within a power amplifier, though this would present its own complications. At the extreme end of this direction we have full frequency digital EQ built in to active speakers with a separate amp for each driver. Such brands as Avantgarde Acoustic, Eikon and PranaFidelity come to mind. Or as The Steven has often suggested, you can put all the goodies into a pizza box as with a Devialet.

    Merry Christmas, may your energy provider bring you clean electricity throughout the coming year. 😎

  15. Exactly right what Paul says. For fixing the dips you need to move speakers or listening position or both. For fixing the peaks (which is even more important) you need analog or digital EQ/DSP.

    Which is why I don’t understand how one can buy/use a sub without EQ/DSP. I wouldn’t even buy a full range speaker anymore, I can’t at least adjust in bass.

      1. Well completely non-adjustable speaker equipment is fine…in case one’s flexible to fix all the upcoming serious frequency response problems with room treatments…or live with them.

  16. I took AFAIK the first uni course in Thiel-Small-Keele theory in 1975, and one of the factoids taught by AES Fellow J. Robert Ashley was that ANY AND ALL EQUALIZATION IS TIME DISTORTION.

    To differentiate between signals of different frequencies, you have to combine samples from different times. Low pass filters sum over time, high pass subtract (differentiate). This is inherent in the Math, it does not matter if the filter is mechanical, analog, or digital.

    This time smearing is a loss of phase information, information that is essential to re-create musical waveforms. Every time some gearhead cites the Fourier Theorem or the Nyquist Criterion, remember that you have to include the phase vector as well as the frequency spectrum to re-construct even scalar signals, and filters throw out the phase vector.

    Audio chains include a series of filters including the invention of the Electrodynamic Loudspeaker, so people who learned to hear music through speakers are phase deaf wrt music. OTOH, we all hear hours of acoustic speech daily from childhood so our phase perception on speech has very well developed neural circuits, accounting for mysteries like blind bicyclists, the one ear test, and the cocktail party effect.

    In reality, the human transduction of sound into neural impulses is in the time domain, not the frequency domain - so we are listening to a transform that throws out 90% of the raw spatial information when we listen to audio, and therefore do not develop the neural circuitry to decode phase in our music processor, which is a separate brain region from the speech processor.

    So, equalization is distortion - except when it is a perfect phase conjugate of a previous filter, like the RIAA curve.

    You can't make a scalar filter that corrects for vector filters like room acoustics - that is like trying to stop a river with a sword!

    You have to let the bass out!

Leave a Reply

© 2022 PS Audio, Inc.

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram