Wayne (or TV or anyone else), could you elaborate on "house curve"?

Discussion in 'Archived Threads 2001-2004' started by Steve Stogel, Feb 21, 2002.

  1. Steve Stogel

    Steve Stogel Supporting Actor

    Joined:
    Jul 19, 2000
    Messages:
    600
    Likes Received:
    0
    Trophy Points:
    0
    I read this in another post, and I was wondering if someone could explain this a little more. I just got finished EQing my sub, and I've gotten it (through much anal-retentive scrutiny) to about +/- 2.5dB. This house curve issue is not something I've read about, and I'm curious to learn more of it. Wayne mentioned he has a 9dB gradual increase between 100 to 32Hz. I just got my response pretty stinkin' flat, and I don't think I have enough EQ to get that kind of response (had wicked peaks from 40-63Hz). Any info would be appreciated. Thanks.

    Steve
     
  2. RichardH

    RichardH Supporting Actor

    Joined:
    Nov 28, 2000
    Messages:
    742
    Likes Received:
    0
    Trophy Points:
    0
    If you want accuracy, you're already there. The "house" curve I think is to boost the bass as you go lower to compensate for our ears' natural response curve. Do a search online for "Fletcher-Munson". You'll definitely need a lot of amp headroom to do that!! [​IMG]
     
  3. Wayne A. Pflughaupt

    Joined:
    Aug 5, 1999
    Messages:
    6,123
    Likes Received:
    40
    Trophy Points:
    6,610
    Location:
    Katy, TX
    Real Name:
    Wayne
    Steve,
    As you’ve noted from my other posts, “house curve” is a gradual rise on response as frequencies drop. The problem with flat room response is that it sounds overly bright and lacking “warmth.” Another characteristic of flat response is subjectively weak and insubstantial bass performance. Richard is exactly right that this is needed because of the ear's natural response to sound. Therefore respected equalizer manufactures like Ashly and AudioContol recommend a “house curve” for a natural-sounding system with adequate bass response.
    There’s really nothing magical about this. The truth is, any system that demonstrates smooth and natural-sounding octave-to-octave performance has a proper house curve, intentional or not, and it would be verified with full-range frequency spectrum readings.
    Dialing in the proper house curve is somewhat subjective, but there are a few "rules of thumb." For one, smaller rooms need more bass boost than larger rooms. However smaller rooms tend to get at least some of the boost for “free,” due to the effect of cabin gain.
    It took many hours of listening and tweaking before I was satisfied with the curve I have now. Not only do you have to determine how much boost to apply, you also have to figure out where to begin the slop and where to level it off (if at all).
    If it’s any help, my listening area is 6200 cubic feet. (When you figure yours, you need to include all other rooms and areas that are open to the home theater.) Since I’ve “been there,” it might help to use my results as a reference (a 9dB rise from 100Hz to 32Hz, and flat below that point), and begin by adjusting the maximum boost up a couple of dB if your space is smaller or down a couple of dB if your space is larger.
    If the upper bass begins to sound “bloated,” you might want to move the point-of-rise down to 90 or 80Hz, or wherever it sounds best. My sub is flat to 25Hz; depending on your sub’s extension you might want to change the leveling-off point as well.
    It might not be as hard to EQ as you think. If you let me know what kind of EQ you have (and how many bands are available if you have a parametric) I can make a recommendation.
    Regards,
    Wayne A. Pflughaupt
     
  4. Steve Stogel

    Steve Stogel Supporting Actor

    Joined:
    Jul 19, 2000
    Messages:
    600
    Likes Received:
    0
    Trophy Points:
    0
    Thanks for the info, guys.
    Wayne, I'm using a Behringer Ultra-Curve (Model DSP-8000). I haven't got a clue about this thing. It's my brother-in-law's. He was using it when he did sound at his church, I believe. While I really have no idea what I'm doing with it, it sure is fun to mess with. Right now my settings are as follows: 20Hz> +3, 25Hz> +3, 31.5Hz> 0, 40Hz> -14.5, 50Hz> -8.5, 63Hz> -5, 80Hz> -9, 100Hz> +3, 125Hz> +3. Now, I know I've got a few 3dB boosts, but I've got the master "slider" set to -3dB. I am hoping this basically cancels each other out (someone please correct me if this is not correct and I'm eating up my amp headroom), but since I've only got +/- 16dB, I wanted a little room to breathe at 40Hz (set at -14.5, so really -17.5 when you add in the master...right?). Here are the corrected numbers I got at my listening position:
    15 - 91.5
    17.5 - 90.5
    20 - 87.5
    22 - 91.5 (High)
    25 - 90.0
    28 - 87.0 (Low)
    32 - 88.0
    36 - 90.5
    40 - 89.5
    45 - 89.0
    50 - 90.5
    56 - 87.0
    63 - 90.5
    71 - 90.5
    80 - 89.5
    89 - 87.0
    As I mentioned, 40-63Hz (actually started at 36Hz) had huge spikes (about 15-18dB higher than my 20Hz spl). As you can see, I've gotten this thing pretty flat, but I sure am curious about this house curve.
    Richard, thanks for the suggestion. I'm going to do a search right now.
    Steve
    P.S. Wayne, I've got very little flexibility other than the Behringer. I'm using a Technics DD "ready" receiver and a Technics DD decoder. So the cutoff is 80Hz, and I have little knowledge about what kind of rolloff there is, et cetera [​IMG]. Thanks for any suggestions.
    P.P.S. BTW, I like the font you use, Wayne. Your posts always grab my attention and are easy to read (some of this may be due to the actual content in the post, too [​IMG]).
     
  5. Nick P

    Nick P Second Unit

    Joined:
    Feb 11, 2001
    Messages:
    270
    Likes Received:
    0
    Trophy Points:
    0
    Curses!! I've just spent hours and hours EQing to a flat response and now I have to do it again to get a curve! GGRRRRRR[​IMG] I am interested to find out more though.
     
  6. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    Nick,

    If you've EQ'd and used uncorrected RS spl numbers for your flat response, then you already have a nice 7-8 dB rise built in!
     
  7. Nick P

    Nick P Second Unit

    Joined:
    Feb 11, 2001
    Messages:
    270
    Likes Received:
    0
    Trophy Points:
    0
    Hey Rick, what kind of newbie do you take me for?[​IMG] Of course I used corrected values and it is a damn flat response too. I worked very hard to flatten it out. Now I have to "curve" it! [​IMG]
     
  8. Vince Maskeeper

    Vince Maskeeper Producer

    Joined:
    Jan 18, 1999
    Messages:
    6,499
    Likes Received:
    0
    Trophy Points:
    0
    There is one large conceptual problem that Myself and many others have about the concept of employing a "house curve" or variations on the Fletcher-Munson curve.
    Granted, the ears have a natural response that dips at the extremes of high and low, and one could certainly argue that it would seem logical to dial a system "flat" by factoring in our natural ear response. So just like when you dial you system figuring your room response, this type of curve also compensates for your natural hearing response.
    Great idea right? Well, not necessarily.
    The reason why is the idea of "creation of a relative environment".
    The idea of proper sound repoduction is not necessarily to achieve a flat response for the room or a flat response for your hearing- but rather the goal is to create a comparable response to the environment in which the material was originally prepared. To achieve this goal, we usually end up defaulting to a "flat response" environment because this is widely accepted as the condition of the control environment in which the material was prepared.
    To offer a better explaination- think of it this way:
    The engineer who mixed the material you are listening to had the same basic hearing response as you have (maybe slightly better utilized but...). The same issues of curbed performance at the outer edges of the spectrum come into play in his listening as it does in your listening.
    This fact will factor into the mix he prepares. He's is mixing on a flat system, or a resonably flat system. Since his hearing, like yours, would be slightly less sensitive to lower bass freq-- if he intends them to audible- do you think he would include more to comensate? Of course he will.
    The same would be true of midrange and upper mids, right? Since his hearing, like yours, would be slightly more sensitive to upper mid freq-- if he intends them to not be displeasing- do you think he would mix less upper mid to comensate? Of course he will.
    Now, enter a concept like a "house curve" into this equation. In your playback environment you have a curve attempting to compensate for a freq response associated with human hearing. What do you think the results will be when you playback material that was mixed and prepared on a flat system?
    Well, you'll be re-compensating with a curve the same compensation that was naturally made because the material was mixed by a human being. The engineer already would have made adjustments for the natural curve of human hearing- because they themselves would experience that natural curve- and he/she presented the material to you in the way it was intended to be heard when reproduced on a flat system.
    Again, remember- the idea of a flat system is not the norm because it is flat- but rather because it is a standard by which to equate the playback to the production environment. Similarly if it was decided that studios would mix with a giant hump at 250hz, it would be advisable for the playback system to replicate that hump- otherwise, the playback system would be lacking in response the engineer had when he prepared the master.
    As an engineer, I can say that I prepare my masters based upon a flat response system (or as close to flat as I can can provide). Naturally then, as a human with human hearing response, I compensate for lacking lows in my hearing by creating a master which relects more of those freq. I listen, and make it sound how I want it to sound-- and if you match the system properties of the system I mixed on resonably- you'd hear the results of that work the same way I did.
    So you simply match what you're hearing to what the engineer heard. You can match the system by making it flat, and you can match the hearing curve by being the same species.
    So I don't personally employ any house curve in my system-- and I have met many many engineers who don't believe in this type of curve either. As long as humans are mixing the audio you hear on a flat system, the compensation for your human hearing will already be built into the recordings.
    -Vince
     
  9. Steve Stogel

    Steve Stogel Supporting Actor

    Joined:
    Jul 19, 2000
    Messages:
    600
    Likes Received:
    0
    Trophy Points:
    0
    Thank you, Wayne and Vince.

    I think you have hit the high points nicely for both arguments. I really appreciate the input.

    Steve
     
  10. Nick P

    Nick P Second Unit

    Joined:
    Feb 11, 2001
    Messages:
    270
    Likes Received:
    0
    Trophy Points:
    0
    Vince,
    Would it be correct for me to sum up what you wrote by saying we should EQ our rooms "flat" because the engineers already compensated for low frequency rol off? So all my hard work wasn't for nothing?[​IMG]
     
  11. Steve Stogel

    Steve Stogel Supporting Actor

    Joined:
    Jul 19, 2000
    Messages:
    600
    Likes Received:
    0
    Trophy Points:
    0
    That's what I gleaned from Vince's statements. Basically, he's saying, "If you want to hear what the engineer wants you to hear, a flat frequency is the way to do this." And since I don't have the EQ to get a house curve, I'll have to settle for as flat as I can get [​IMG].
    Steve
     
  12. brucek

    brucek Second Unit

    Joined:
    Dec 29, 1998
    Messages:
    335
    Likes Received:
    0
    Trophy Points:
    0
    I've heard both sides of this discussion before and Vince certainly makes his point very well, but I'm afraid I'm not completely convinced. I'll stick with the house curve theory.

    I've always found systems that are equalized flat, "sound" to me, just that - flat. I certainly understand Vince's argument, but my feeling is that there is not much chance my living room (where my HT is located) will ever even remotely approach a "like environment" of a studio where an audio mixing engineer does his work. The acoustics of a well designed studio have little to do with the acoustics of my living room. They are carefully designed to remove as much color and reflection etc as possible, and might be viewed as almost sterile to most people if they listened to any music there.

    In fact any studio pictures that I've seen have the engineer listening "near field" to further remove the studios room effects. It's my understanding that most studio monitors are intentionally rolled off.

    I know that the engineer listens with the same human ears that I listen with, but he's listening in the studio.

    Now play that studio generated CD in my living room where the room acoustics and listening distance are much different and the sound that I will perceive needs a bit of a boost in the bottom and sometimes in the upper end also.

    brucek
     
  13. Sonnie Parker

    Sonnie Parker Second Unit

    Joined:
    Dec 11, 2001
    Messages:
    407
    Likes Received:
    0
    Trophy Points:
    0
    This is very interesting!
    Just this week I had to re-eq my sub. My recliner broke (heehee...with some help) and I had to move some furniture around.
    I set up somewhat of a house curve with a gradual 10db slope from 20hz to 80hz (my x-over point). It's +-3db from 18hz to 31.5hz. I actually believe I like the sound better this way. The only thing is I haven't compared apples to apples yet. I need to eq it to flat and switch between presets to be fair. But, it appears as if this new curve will be better. The furniture moving did make a difference, not much, but some, from the main listening position.
    I believe it should be at least a suggestion to try the house curve and see if you like it. Makes sense if it sounds better to you then you'd use it (and vice-versa).
    Wayne, it seems you know a good bit about the house curve. Is there anyway I can talk you in to writing up a detailed information piece concerning the house curve. And maybe even a few graphs with different examples of what curve you might use in a particular size room. Yours would be a good example I'm sure. I would be more than happy to put the info onto a page and use my web space for the page. I would like a link to it on my BFD page. Any information I can gather about sub eqalization is greatly appreciated.[​IMG] [​IMG] [​IMG] It was suggested to me that you might not mind doing this. If it's too much to ask then I respectfully understand.
     
  14. Wayne A. Pflughaupt

    Joined:
    Aug 5, 1999
    Messages:
    6,123
    Likes Received:
    40
    Trophy Points:
    6,610
    Location:
    Katy, TX
    Real Name:
    Wayne
    It would be nice if all engineers were as conscientious as Vince. But unfortunately this is not the case. Anyone who has more than a half-dozen CDs or DVDs knows that program content is all over the map. Brucek is right, a studio control room bears little similarity to the home environment. All a recording engineer can do is make his product sound as good as possible, but he ultimately does not know if you will be playing his CD in a car, a small bedroom, large living room or disco club.

    Personally I have never EQd a room where the system sounded good once measured-flat response was achieved. Therefore I would be truly surprised if Vince’s system does not exhibit a house curve, despite his claims to the contrary.

    Anyone with a SPL meter and a test disc with broad-band pink noise can verify whether or not they have a house curve: Set the meter to A weighting, which only reads frequencies from 500Hz to 10kHz, and take a reading of the pink noise signal. Then switch the meter to C weighting, which will add frequencies from 32Hz to 500Hz to the equation. If you truly have flat room response, the meter reading will remain the same, or perhaps increase only 1-2dB. But most likely you will see the meter’s reading increase by 6-15dB (maybe even more). This means bass response in your room is higher than the upper frequencies – in other words, you have a house curve. You may not have intentionally created it, and it may not be “perfect,” but it is there nevertheless.

    Steve, if you’re still with us, the best and easiest way to dial in a house curve is with a shelving EQ filter. A shelving filter has a designated turnover frequency (the “hinge” point, if you will) and with boost applied, response gradually rises below that point and levels off.

    Some parametric EQs include shelving filters; these are the best because they allow you to dial in any turnover point you want.

    If not, a parametric filter can be configured to essentially create the desired effect.

    Alternately, the bass and treble controls on receivers are shelving filters, so you might try using the bass control (if it works for the sub). However, they often effect response all the way up to 1kHz. Some receivers give a selection of turnover frequencies; if so experiment between 60Hz and 100Hz.

    With all the above examples, you would apply the adjustments on top of any EQ you already have used to flatten response.

    Steve, for a guy “just playing with it,” you’ve dialed in remarkably flat response - within a 4dB window despite those “wicked peaks” you’re (needlessly) worried about. I noted at the Behringer site that the current model of this EQ has parametric filters in addition to the 1/3-octave filters. If yours does, try setting a 1½ or 2-octave filter centered between 28-36Hz and boost between 4-6dB.

    Notice I’m giving a range for all settings - again, this is mostly by ear. You have to experiment. And you can’t go just by measurements.

    If there are no parametric filters, you can try tweaking the 1/3-octave settings. Based on the reading you provided, try the following changes (i.e., added to your current settings):

    20Hz > +5dB

    25Hz > +2dB

    32Hz > +4dB

    40Hz > +2dB

    50Hz > (no change)

    63Hz > -2dB

    80Hz > -2dB

    You might have to re-set the master level to restore a proper blend between the sub and mains.
     
  15. Steve Stogel

    Steve Stogel Supporting Actor

    Joined:
    Jul 19, 2000
    Messages:
    600
    Likes Received:
    0
    Trophy Points:
    0
    I am still very much with this thread, and I appreciate all of the wonderful information. Thank you all.

    Steve
     
  16. Thomas_Berg

    Thomas_Berg Screenwriter

    Joined:
    Feb 28, 2001
    Messages:
    1,422
    Likes Received:
    0
    Trophy Points:
    0
    Location:
    Dallas
    Real Name:
    Thomas
    i really think that the pros here (Wayne/Vince) should get together and write up an EQ/Calibration FAQ with two levels: Basic and Advanced. that way, every time there's a question regarding something like this you can just post a link instead of having to be redundant.

    btw- i've been really learning alot in these past few weeks and am anxious for my BFD to arrive this week. (it should've been here LAST TUESDAY though!!) thanks for the info guys!
     
  17. Sonnie Parker

    Sonnie Parker Second Unit

    Joined:
    Dec 11, 2001
    Messages:
    407
    Likes Received:
    0
    Trophy Points:
    0
     
  18. Vince Maskeeper

    Vince Maskeeper Producer

    Joined:
    Jan 18, 1999
    Messages:
    6,499
    Likes Received:
    0
    Trophy Points:
    0
     
  19. gene avallon

    gene avallon Stunt Coordinator

    Joined:
    Jun 13, 2001
    Messages:
    187
    Likes Received:
    0
    Trophy Points:
    0
    can anyone tell me if the pink noise on the strykes basszone test cd vol. 1 is broadband?I did the test like wayne said and I had a 10db rise.should all other speakers be off? I had the mains off only.
    thank you
    gene[​IMG] [​IMG]
     
  20. gomez_a

    gomez_a Stunt Coordinator

    Joined:
    Nov 14, 2001
    Messages:
    119
    Likes Received:
    0
    Trophy Points:
    0
    I would really look forward to Wayne, Vince, Sonnie working up their sub EQ sites - this stuff is really interesting, and perplexing - like the "chicken or the egg first" conundrum. Anyone wanting to EQ their subs should be aware of this . . .

    I hope I'm summarizing Wayne and Vince's posts on this thread correctly . . . . I can see both sides of this argument, on one hand you have the "flat/sterile" recording environment, which if the sound engineer did not correct/compensate for, then it appears that a "house curve is needed", but then as Vince mentioned, the sound engineer is human and he may engineer the sound with his version of a "house curve", therefore, a house curve is not needed.

    I wonder if this boils down to knowing how the sound engineer recorded tthe material - but this is impossible to know unless you know him personally, or the CD case mentions it (not likely)

    I guess the BFD has a memory feature that let's you save a flat and a house curve setting whereby you can flip between the two, right?
     

Share This Page