| Columns Retired Columns & Blogs |
October 25, 2010 - 10:34am
#1
The House Curve
| Loudspeakers Amplification Digital Sources | |
Analog Sources Accessories Featured | |
Music Columns Retired Columns | |
Show Reports | |
Features Latest News Community | |
Resources Subscriptions |
© 2024 StereophileAVTech Media Americas Inc., USA
All rights reserved
What makes you think audio reviewer's hearing sensitivity would measure as a ruler flat response? As to this new fangled room analysis and equalization I predict that eventually speakers will come with a calibrated microphone you place in the listening chair whilst the speaker's software analyzes the response and adjusts it's output accordingly. Velodyne and some other sub-woofer designers are already half way there.
The problem will be in convincing customers that even after such compensations are made acoustic room treatment will still be required for best results.
We do that already. And, yes, post-EQ tweaking is common although I prefer to build the target curve into the EQ process.
Kal
Who is 'we'? The manufacturers, Stereophile's reviewing staff?
Target curve? Surely a flat response at the listening position is the only valid target. I'm missing something here.
Some manufacturers (McIntosh, Audyssey, NAD, TacT, for example) and me.
Yes, I had the same predisposition that a perfectly flat response was the only valid target but there is evidence to support a curve that is subtly tilted down. This is discussed in Toole's book (and other places). It is based on the differences between a concert hall and a listening room in acoustics and source-listener distance.
Note that we are talking about adjustments on the order of +/-5dB over the entire spectrum and most systems cannot be confined within that envelope to begin with, so this applies mainly to those that have been or can be made flat anyway.
Kal
I don't.
Two different people's hearing will likely have two different frequency response curves. However, they will not only hear differently in the listening room, but in the concert hall as well. What I believe the article is saying, is that measured flat response in the listening room will sound less accurate and therefore less like the concert hall, to all listeners, than some "house curve" with a boost in the bass and a cut in the treble.
I'd like to know why.
Kal, you seem to have answered my question while I was typing. Thanks. I'll have to go find that book now. What was the title?
http://www.stereophile.com/reference/boo...omsi/index.html
Actually, "house curves," like most tweaking, represent an ad hoc attempt to accomplish the same thing and a tacit acknowledgment of the issue.
Kal
DR. Kal,
I am hoping this is to the point, but do you find the need for "different " curves based on varying loudness levels in your rooms? Would you consider one curve acceptable for say an spl of 65-70db and the need to change it when you might go up to 80-95db base on some changes in the power response of your speaker systems?
I am wondering if this is an effect at all for one to consider?
It is a real effect, of course, as illustrated by Fletcher and Munson. I do not adjust the curve but, then again, I can set the volume at what ever level I want. Both Audyssey (Dynamic Volume) and Dolby (Dolby Volume) have systems which will apply automatic compensation for sub-reference-level settings.
Kal
On the surface I find these ideas make sense until you start thinking about what happens in the concert hall to our hearing sensitivity over the full frequency range when differences in the volume produced by, say, a live symphony orchestra can be extreme. An additional puzzle is the behavior of a range of speakers at low, mid & high volumes. I find very few speakers I've used over the decades sound as dynamic and real at low volume as they do at higher ones. My Usher BE 10's however, to my ears in my room, are as impressive at low volume as they are at higher ones. I've no idea why this is the case. One other design I've heard recently that also manages this trick is the Aaron HMF -600MkII.
By the way, at $4,400 Aus these are an absolute screaming bargain.
http://www.aaronhifi.com.au/category.php?cat_id=32
That is not an issue, of course, since it is inherent in the live sound and in the recorded source. Turning down the volume knob is, as you will agree, not the same as having the musicians play quieter.
Now, there, I agree with you and it is such subtleties that evade the technology at present. However, these automatic compensations for different playing levels are not aimed, imho, at serious listening but, rather, at providing some help when you wish to listen under conditions that constrain you from setting the levels where they belong.
Kal
Possibly the important point we can learn from this thread is that IT SHOULD BE.
No, it shouldn't.
The two are different things.
Think of it this way, Diss. If you have a bass and a guitar playing on the hi-fi, and you turn down the volume, the bass will diminish more quickly because of the Fletcher-Munson effect (and possibly other factors). If however, the two musicians were to gradually play more quietly, the bass player would compensate, because of the Fletcher-Munson effect on his own hearing, and not play quite as quietly as the guitar. Make sense?
Yes but more than that. Performers play sounds at different levels as called for by the music and they do so by playing their instruments differently. Unless the instruments are purely electronic, that will change the attack, the overtone structure and the decay.
Even if there was some way to mimic these changes electronically, it would be even more of a distortion of the music than merely compensating for the F-M curves.
Kal
I agree with everything you said, Kal. I wasn't thinking of an electric bass and guitar, but acoustic, like Ray Brown & Laurindo Almeida's "Moonlight Serenade." A good performer will also compensate for the other factors you mentioned, therefore I stand by my statement, while agreeing that more it at work here than Fletcher-Munson.
There's a couple of things you're not taking into account, Kal.
First, by doing partial-loudness analysis, it's quite possible to reproduce the same loudness ratios at lower total loudnesses. This is, indeed, not the same as the players playing more softly (for reduced loudness) but can in fact create the sensation of players playing loudly, but with less overall total loudness.
This creates a sensation of a quieter version of loud playing, but without the missing bass and somewhat muffled higher frequencies.
For listening under "don't wake the baby" conditions, this is, while not perhaps ideal, a better experience in my opinion than having all the bass and some of the treble disappear.
This kind of processing is not entirely simple, there is quite a bit of processing and careful adaptation to the exact listening setup required as well. It can, however, be made to work very nicely when done carefully and with good measurement of the playback arena.
Interestingly, when this is done, it is also possible to compensate for environmental noise masking, and some other things I won't mention at the minute.
So, am I to assume that the goal of reproducing music performed at a low volume and music recorded at higher volumes but played back much lower on a domestic audio system cannot be accomplished in a manner faithful to the music ? Given what I hear from my collection of audio junk I'd disagree. I know the human ear responds differently at low volumes but on my system anything like a 'loudness switch' ( sorry for using a dirty word!) would be pointless. My impression is that what I hear from my speakers at low volume does not change the feeling ( can I use such an unscientific term here?) of the music at all.
What I want to get to the bottom of is why some speakers are able to do justice to the music at low volumes when most obviously don't . What's going on here?
Speaking from personal experience, the large percentage of pipe organists will have a colleague play whilst they sit at a distance from their instrument during rehearsal to hear effect of various stop settings etc are at different volumes.
Sadly though, not all musicians, Fletcher M effect or not, alter the balance of their performance to compensate for lower volume playing. I've attended a number of concerts where it's just too obvious that the performers assume that if it sounds alright at their location that's all that matters. There's the question of distance to the listener, especially at higher frequencies, we need to consider as well.
Excuse me using another rude term liable to upset the audio purists, but isn't it time defeatable tone controls started re-appearing on high end electronics? Recently I read a review of an amplifier that, as volume was decreased, instituted a graduated (logarithmic?) 'loudness' compensation. Can't remember the brand but I'd love to hear how effectively such an approach actually works with a range of speakers.
Reading back all that I'm aware I've used a number of very subjective terms but I don't think even Harry Pearson has the vocabulary to describe these questions.
Well, it can be approximated but there will be discrepancies.
I cannot say because I cannot account for personal perceptions.
I cannot say because I cannot account for personal perceptions.
Here's where measurements may tell us something as I sure don't expect you to account for my personal perceptions - even I can't. I'd like however to hear what you thought of Ushers' BE 10 speakers in this regard to see if your subjective reaction was the same as is mine and a number of others who've commented favorably on their behavior at low volumes. They do insanely loud rather well too, in fact they can go way too loud for me.
I'd like also for you to hear the Aaron HMF-600MkII's in this regard but the chances of a pair of those turning up in the US is very unlikely.
It's just occurred to me that if progressive frequency 'enhancement' for ever lower volumes was built into an amplifier, to compensate for the Fletcher M curve, such a strategy could come undone due to different sensitivity etc in a particular speaker being driven. The argument for active speakers just got stronger I imagine.
I have only heard the BE-10s briefly at shows so I cannot say anything in this regard.
Your point is well-taken but well known. In the old days, such compensation was done and suffered as you would have predicted. Today, we are smarter and the compensation is more intelligent and is only implemented after system calibration at reference levels.
Kal
Sorry, I'm way off topic here.
Contrary to advice I give anyone who ever asks, I didn't hear the BE10's on my system before buying them. I stumbled across them in a Hong Kong showroom and instantly fell in love with their 'naturalness'. A big jump from my previous speakers, Soundlab ULP x's. I've only come across one review of them in a UK magazine.( Hi-Fi News?). I'd love to have Stereophile or Absolute Sound pass an opinion on these just to see if my enthusiasm isn't an auditory hallucination. To my ears, on my system they make some stats sound plastic by comparison. Any chance Stereophile could get Usher to lend them a pair ? The only weakness I've found in my room was an overpowering lower bass which I've cut off at 45hZ handing that job over to a pair of Velodyne DD15,s which, when positioned carefully, minimize standing waves at the hot seat and at my desk further down the room.
Maybe that's your answer. Could the excess of low bass be partially compensating for the Fletcher-Munson effect at low volumes?
No, the same effect is obtained cutting the Usher's off at 45hz and giving the lower bass to a pair of Velodyne DD15's. There's no excess of low bass using that strategy and that's how I'm running the BE10's in my room. Mind you, this could indeed be a room problem as I've had other speakers produce the same bloated effect. One reason full range speakers have real problems in my experience is the fact that if you position them to get the soundstage etc correct then the bass is wrong. Position them to get the lower bass right and the soundstaging falls apart. My solution is to hand the sub bass over to a pair of sub-woofers, placed optimally to reduce standing waves at the listening position. This allows the main speakers to be positioned for best imaging etc.