Question about equlizing with a BFD (2 Viewers)

[Greg Yeatts]

I have used a BFD for the last year or so to tame some room modes. In particular, I have a 6bd hump at 35 hertz and a +12db hump at 47 hertz. These have been present in any sub I have used, so I know its a room mode. My problem is that I cannot get the +12 db hump at 47 hertz flat without using -20 db or so cut at 47 hertz. Why is it taking more than 12db of cut to counteract a 12 db hump. The other problem is that if I do add in enough EQ to make the hump completely flat it sucks the life right out of the music.

Could this 12db hump be an artifact of my measuring process. I am using sine waves and a rat shack SPL meter. My guess is that the constant sine wave energizes the room mode in a way that music, which is transient in nature, does not. Should I just adjust the eq to subtract 12 db from the hump and not worry about the response being flat?

BTW, the 6db hump at 35 hertz hump is easy to equalize with no penalty in sonics as a result of the equalization.

 

[TimForman]

"I feel your pain."
I could give a better answer if I knew how wide the peak is. The problem I had was getting the bandwidth correct. I was going too wide to address a narrow peak which throws everything off. I finally got the "octave thing" through my head which simplified the process considerably. I also made the mistake of using too many filters. Your goal should be to use as few filters as possible.

 

[Greg Yeatts]

You should get a copy of spreadsheet designed by Anthony Gomez on the FRD Consortium that helps you set the BFD. It is a great resource. You measure your sub and put the measurements into an input worksheet. The sheet them graphs your results and you can actually apply eq to the spreadsheet (using an input interface that uses the BFD's peculiar input format). The input results, eq curve and projected final curve are all graphed. It's a very usefule tool. It worked fantastic except for that one huge peak. That is not the fault of the spreadsheet.

 

[TimForman]

I did just that but I still noticed some things I wasn't happy with. Of course, everything changed when I added a second sub. Still, I found I had much better results just using 4 or 5 filters with very narrow bandwidths as low as 1/60th octave. I also ran test tones with all the speakers running. That was quite revealing too. It's an interesting effect when you hit 120 db (adjusted) with everything running. Probably not healthy though (for me, the equipment or the house). I was wondering what your personal criteria is for variance. I figured if I was +/- 6 db I was doing exceptionally well (I have no wall treatments yet). I understand your concern about the 12 db hump. That's probably a little too much but as I said, in my experience overtweaking just caused more problems.

 

[Allen Ross]

for music tune it to your likings, most likely you want something like a house curve, some one will find you a thread about it. basically it has a nice little boost that will sound great for music but too boomy for HT, so use to presets one for music with a house curve and one for HT thats flat.

 

[Wayne A. Pflughaupt]

Greg,

The other problem is that if I do add in enough EQ to make the hump completely flat it sucks the life right out of the music.

A lot of people find out the hard way that ruler-flat response just doesn’t sound good. Allen is right that a so-called house curve generally sounds the most natural. A house curve is a gradual rise (or slope) in response at the lowest frequencies – something like what you had when you had the 47Hz hump equalized “incorrectly.”

You might want to take a look at this link to learn more, including how to determine the slope your room needs:

http://www.hometheatershack.com/foru...ed-how-do.html



Regards,
Wayne A. Pflughaupt

 

[TimForman]

A lot of people find out the hard way that ruler-flat response just doesn’t sound good.

Agreed, plus given the reality of rooms probably unachieveable. I concentrated on the big peaks and let the rest just happen. As long as there's no exaggerations it sounds good to me.

 

[RichardHOS]

I've always wondered why flat doesn't sound too good. ??

With the acoustic treatment of most mixing/mastering rooms, and the limited extension that many (not all) popular near-field monitors used to mix so many albums, you'd think that it is less likely that a studio has a "house curve" than a high-powered sub in a typical home listening room.

If anything, you'd think that mixing/mastering engineers would emphasize the low frequencies to (1) compensate for a perceived lack of bass from their own playback systems and/or environments or (2) to compensate for the lack of bass they would expect from a typical listener's car/boombox/home system. Following that logic, it would seem that a high powered home audio/HT system that truly achieved flat response would actually have too much bass due to the way the album was mixed.

So why this seemingly inverse phenomenon? Why do the rec. engineers mix/master albums with what we perceive as "too little" bass? Do they set up a "house curve" in their studio on purpose, just as many here are saying is needed for "balanced" sound (not that I disagree)? Do they skim the bass intentionally to prevent damage to lesser playback systems?

I'm not disputing at all that this condition exists... I'm just curious from a recording viewpoint why it does.

 

[TimForman]

I'm not sure engineers have that much discretion. Don't they have to stick to what the artist/producer wants? There are many modern recordings where the bass is exaggerated (Hip-Hop, Techno). My understanding in movie sound tracks that the same restrictions apply, although they do allow for the LFE to be handled separately, but the main track still has to include the full bass signal intended by the creator. But, I'm no expert and my eyes start crossing about halfway through those AES papers.

 

[RichardHOS]

Yeah Tim, I could certainly see the producer or label having their finger in pot as far as how the final mix is done. But wouldn't you think that for most music they would hype the bass if anything?

I'm trying to figure out a legitimate reason why they would intentionally skim the bass such that you have to EQ your sub "above" flat to produce what is perceived to be an equal or pleasing balance.

 

[Wayne A. Pflughaupt]

Richard,

I'm trying to figure out a legitimate reason why they would intentionally skim the bass such that you have to EQ your sub "above" flat to produce what is perceived to be an equal or pleasing balance.

They don’t skim. It’s a very complex situation that it has to do with way the human ear hears and the psycho-acoustical effects (I think that’s the right term...) of the listening environment. Actually, you nailed it when you said “perceived” balance; keep your eye on that ball as you read the following.

Starting with the ear, it’s a well know fact that the ear does not have “flat” response. Its perception of low and high frequency extremes varies with sound intensity (i.e., sound pressure level). And while this is not something you often see discussed, I’m convinced that everyone’s ears hear a little differently.

The next situation is the listening environment – your car, house, etc. There are two major acoustical factors here, time/distance, and cabin gain.

The rule of thumb for the first factor, time/distance, is that as sound travels through the air, the high frequencies attenuate and roll off at a faster rate than the low frequencies. (Note: The designed dispersion pattern of the high frequency driver will also play a role here. Tweeters with wide dispersion will have greater high frequency attenuation with distance.)

Perhaps you have noticed that speaker manufactures offer their frequency response specs at a very close distance – typically one meter (which is roughly three ft.). However, if you listen to them at that distance you will find they sound thin and overly bright.

Put some distance between you and the speakers and there is some natural high frequency attenuation - which makes them sound more balanced and “natural.” So “right out of the hole” the ears are looking for a slope in response – reduced highs (the inverse of bass boost) – in order for the sound to be perceived as balanced (there’s the “p” word).

The next factor is an acoustical phenomenon known as “cabin gain,” which has to do with the amount of bass reinforcement that a room gives to the speakers. The rule of thumb for cabin gain is that smaller acoustical spaces naturally boost bass. This is why you easily get tremendous amounts of bass in a car – a tiny acoustical space. As room size increases, more “woofage” is needed – larger or more drivers, more amplifier power etc. - to fill it with the same SPL level and extension you got for “free” in a smaller room.

Thus it would seem that the time-distance factor and cabin gain factor are “fighting” each other in achieving properly balanced, natural-sounding room response: For instance, in a small room you will obviously be closer to the speakers, increasing perceived high frequency response. Yet you will also get exaggerated bass response due to increased cabin gain. In large rooms it will be just the opposite.

However, if you adjust a system in both a small room and a large room to where each sounds balanced, you will find that when you actually measure broadband response (20Hz-20kHz), the smaller room will have a steeper, more exaggerated bottom-end slope. This phenomenon is known as a “house curve” – so-called because the amount of slope needed to achieve the perception of balanced sound (that, as noted, the ear already demands) is dependant on the room in question (and again we have the “p” word).

Thus a house curve is merely a balancing act between the room’s cabin gain and the distance you sit from the speakers.

Hope this makes sense – hope it wasn’t overwhelming...

Regards,
Wayne A. Pflughaupt

 

[RichardHOS]

lol, you know someone is up too late posting when they start explaining what a meter is. Thanks Wayne, but I think I had that one covered already.

I wasn't confusing the steps of the recording process... I do some tracking and mixing myself, which is why I mentioned acoustic treatments for both mixing and mastering, and nearfields only for mixing. It still stands that either the mixing or mastering engineer could boost or cut bass at their respective steps in the production process, but that is really beside the point I suppose.

I'm sorry if I find your two explanations (or is that a two-part explanation?) quite lacking. It really seems very simple... if the typical home listening room needs a house curve to sound good, then why don't studios master the album accordingly? And when you consider that the vast majority of cars and home systems probably have inadequate bass, it hardly seems likely that they EQ the album they way they do to avoid excessive bass if a few outrageously equipped automobiles.

Most mastering houses have several rooms, or at least several monitors in a room to judge the sound by. Are you suggesting that it is common for a mastering studio to be so much larger than a typical room that they 'unintentionally' mix the album with less bass than would be needed for our somewhat smaller rooms (assuming that size/bass relationship holds true)?

I guess what I'm asking is pretty simple: if trained mastering engineers thought an album needed a "house curve" to sound "right" when played back, in a room similarly sized and with relatively flat speakers, why not just master the album that way to begin with? If our ears perceive a lack of bass, wouldn't theirs as well?

Or, as I've tried to fish out... if the bass does sound lacking to them, and they are doing that intentionally for some unknown purpose, then what is it?

 

[Brian Fellmeth]

Richard's logic here is inescapable. If someone thinks a house curve sounds better, it must mean they like more deep bass than the recording engineers/artists do.

 

[BruceD]

Richard,

I guess what I'm asking is pretty simple: if trained mastering engineers thought an album needed a "house curve" to sound "right" when played back, in a room similarly sized and with relatively flat speakers, why not just master the album that way to begin with?

I would tend to agree with Wayne, it's not so simple, and would ask you how could anyone master a song with a "generic house curve" to meet the needs of all the different "house curves" of all the different sized rooms out there?
ANSWER: Impossible.

 

[Wayne A. Pflughaupt]

Richard,

Richard's logic here is inescapable. If someone thinks a house curve sounds better, it must mean they like more deep bass than the recording engineers/artists do.

A house curve isn’t about getting “more” bass. It’s about achieving response that sounds flat in your room (remember the “p” word?), as opposed to merely measured-flat .

The fact is, if you believe your system has balanced response, you have a house curve - whether you intentionally calibrated for it or not.

You don’t have to take my word for it; you can easily check for yourself if you have an SPL meter. Play a CD and switch the meter between A and C weighting. As you probably know, A-weighting rolls out at 500Hz and therefore doesn’t “see” low frequencies. If you truly have flat response (no house curve), the reading on the meter will not change when you switch to C weighting.

If you want something a little more scientific than a CD with music, you can use a pink noise signal from a test disc. If you only have single test tones, a low and a high frequency will suffice – say, 50Hz and 1kHz.

Regards,
Wayne A. Pflughaupt

 

[RichardHOS]

Bruce and Wayne,

I'm not arguing that a house curve might not sound flatter than it measures (i.e., may sound better and more balanced to us than a flat response), simply wondering what the technical reasons for that are.

There seem to be only two possibilities: (1) engineers master the recording with less bass than what sounds flat to them because of one or more playback environments they are targeting where, for whatever reasons, flat sounding bass response in their room wouldn't correlate well to good sound in that environment, or (2) they do master the recording to what sounds flat to them, and for one or more reasons our playback environment is different enough from theirs to yield less bass than what they hear.

I could easily buy the first argument... I just knew what those "reasons" were. I would have guessed that the tendancy was to hype bass, not cut it, for the benefit of people lacking bass in their systems. Maybe they don't want to damage those systems, or figure the biggest playback environment is a high-powered car system with tons of cabin gain?



Perhaps it is the second explanation. Wayne has hinted at this. He suggested that there is a proportional relationship between room size and perceived amount of bass. By extension, that should mean that adding narrow and broad band bass absorption to a room would simultaneously reduce the SPL of bass in the room and increase the perceived SPL. Strange effect. Perhaps, the explanation lies in the significant amount of absorption found in most mastering studios. If Wayne's suggested relationship holds true, then that absorption of bass in the mastering room would result in less reverb, less SPL, and more perceived bass... thus when they master the recording it has less bass than is needed in our boomy reverberant rooms.

That would indeed be a weird relationship, but one I could probably accept. If that is the case, then the way to good sounding bass is clearly proper room treatment for a reduced RT60 in the bass frequencies... something that might be hard to achieve.

 

[Stephen Dodds]

It also depends on the volume you listen. The ear is less sensitive to bass frequencies at lower volumes, so a house curve can help there, acting like the Loudness button on some preamps.

Of course if you start playing a reference levels you will have too much bass.

Steve

 

[RichardHOS]

Which brings up another question for this somewhat meandering thread. Are there EQ's (digital I would presume) that use F-M curves such that the gain of each band is variable depending on the volume setting? Not sure if such an EQ would have to be tied to the preamp gain for information on gain setting, or if it could just take an average for each band from the material being played.

I know there are dynamic EQ's, but I've been under the assumption that those included a dynamic compressor and weren't "dynamic" in the sense I'm describing.

 

[Pete Mazz]

The human ear certainly has a "house curve" that does indeed change with volume. It is a Fletcher-Munsun curve:



However, if you listen to a live unamplified performance, there is no EQ being done to it. Why would you introduce one to a recording of that event, assuming it was recorded w/o any EQ? To sound "natural", we should shoot for flat in-room response, and not try to EQ for the non-linear nature of our ears.

Pete

 

[Wayne A. Pflughaupt]

Why would you introduce [equalization] to a recording of that event, assuming it was recorded w/o any EQ?

That’s a pretty big assumption. You can pretty well rest assured that a recording of a live event has been processed and equalized by the time gets to the store shelves.

I’ll say it again: If your system sounds balanced to your ears, you have a house curve. It doesn’t necessarily require equalization. It can be done say, with the tweeter level control that some speakers have. Or by adjusting the receiver’s tone controls.

It can also be done simply by your choice of speakers. For instance, if you brought home a pair of speakers that sounded thin, or bass-heavy, and took them back and exchanged them for a pair that sounded more “balanced,” you were “tweaking” your house curve.

So even without equalization there is already a natural slope in the mains. Factoring in the sub) it makes little sense for everything below 100Hz to be flat-lined. It simply doesn’t sound good to most of us.

Regards,
Wayne A. Pflughaupt