Interesting thought, and it agrees with my experience. I think I have heard very good midrange and treble from speakers quite often, but very rarely have heard what I consider to be very good bass. One of those times taught me a lesson, and that is the room may have an effect, but the basic problem in most cases seems to be the speaker itself. The demonstration was a string bass, played on a large bass reflex box with an 8 inch driver in an acoustically lousy living room (small, boxy, reflective). The revelation was that it sounded just like a real string bass in that lousy room.
I think the room is used too often as an excuse for lousy-performing speakers - bass-wise.
I also feel that the notion that quality of bass underpins critically the entire experience is correct. Just a hunch, really, as it must be.
Earl, Ginetto mentions,
and I have noted over the last 40+ years that a lot of speaker systems have difficulties in reproducing choral ensembles particularly loud passages. ( Best described perhaps, as a sort of "tearing paper" sound, sometimes loud, more often soft, intruding). What is interesting, is that I have also sometimes noted the same sort of unpleasant characteristic of sound in the live situation when in, or close to, the ensemble.
My conclusion is that this subjective experience is usually a combination product of distortion, both linear and non-linear, in hearing and speaker systems.
The "culprits" I suspect are "wobble", vibrato, and singer's formant which has frequencies between about 2 and 5kHz and even higher. (There are pictures here: https://www.google.com/search?q="si...JPNDpoAS0r4LgDQ&ved=0CDsQsAQ&biw=1366&bih=624 )
A choral singer's formant is not usually as prominent as an opera singer's but it's there, nonetheless, and "wobble" is the variation around the pitch that all voices have and vibrato is the voice turning on and off. Under normal circumstances these are not noticeable and are even pleasant, but when the ensemble sings loudly, and if the speaker system has badly implemented cross over, or cone breakup, or diffraction products, then they can really bite. Previously pleasant wobble, zing (formant), vibrato and accompanying disharmonies, all part of the normal choral texture, start sounding unpleasant as bits of them are emphasized.
I'd use something like this as part of evaluating speakers
It's really high class performance and fairly decent live recording but if speakers are problematical then we ain't going to be singing halleujah and the soprano who comes on afterwards, whose voice has real zing (good formant) isn't going to sound as beautiful as she looks. The trumpets aren't going to sound very pleasant, either.
Haendel Hallelujah from Messiah , Collegium 1704, Václav Luks - YouTube
Very good points.
In my experience its a variety of sibilance, and that's down to intermodulation distortion in electronics, not speakers. I agree that operatic singers are the hardest to reproduce cleanly. Whilst working on my DAC I thought I'd got this problem licked but it turns out that on a particular operatic recording I could still hear it. So I worked away at reducing it in the electronics, through additional power supply filtering and changing of the grounding layout.
In my experience its a variety of sibilance, and that's down to intermodulation distortion in electronics, not speakers. I agree that operatic singers are the hardest to reproduce cleanly. Whilst working on my DAC I thought I'd got this problem licked but it turns out that on a particular operatic recording I could still hear it. So I worked away at reducing it in the electronics, through additional power supply filtering and changing of the grounding layout.
Part of the reason that operatic recordings can expose shortfalls in electronics is that when the singer hits a big note and holds it for a sustained period, that this is a particularly high stress moment for the amplifier, the power supply at this point is working close to the limits of its capabilities. I've never had gung-ho rock or such-like cause audible problems, but on a sizable number of occasions the amp has started glitching, from the internal protection mechanism cutting in, or shut down completely from thermal overload, when a soprano has been working the air waves hard ...
Believe or not, I had this happen with a Nellie Melba recording, done in, ohhh, the 1910's - the SPL's were up, she hit the big note and kept the power coming - next minute, clicking and popping as the amp went into survival mode ...
Believe or not, I had this happen with a Nellie Melba recording, done in, ohhh, the 1910's - the SPL's were up, she hit the big note and kept the power coming - next minute, clicking and popping as the amp went into survival mode ...
So only circumstantial evidence then. Nothing scientific.
Frank - I am not a big fan of classical music, but the Messiah is a piece that I try and get to every few years. To me it represents all that classical music can do. Religion has a way of bringing out both the best and the worst in mankind. It's a curious thing.
Frank - I am not a big fan of classical music, but the Messiah is a piece that I try and get to every few years. To me it represents all that classical music can do. Religion has a way of bringing out both the best and the worst in mankind. It's a curious thing.
Actually a quite wrong claim - science is founded on observations. Measurements are secondary.
I've found that even mid-fi electronics are satisfactory and will produce very good sound if speaker systems are well designed and built. Some artifacts of speakers can really fool people into thinking they've got electronics problems. Oddly, often the problem isn't even nonlinear - diffraction, for instance, can sound like sibilance, depending on dimensions.
I do remember early digital gear had some nasty sounds and opera singers did give it grief but, and I have no proof, I suspect it was the analogue part.
Could be the fact that I'm using very low-cost speakers on my hot-rodded electronics. But every time I've isolated an annoyance in the sound, there's been an electronics fix for it. Haven't had to upgrade my speakers so far to date. I don't rule it out though longer-term but all the low-hanging fruit has been distortions added in electronics for the past 3 or so years.
And likewise ... doesn't excuse one from fixing silly, obvious weaknesses inside the speaker carcase - this is something I always do as first thing - but if the driver leads can be fed clean signal then the subjectively irritating artifacts go away ...
"Circumstantial" as in correlation doesn't necessarily mean causation? Perhaps so. But, I've found that speakers that have problems with big choral pieces don't give satisfactory results with other kinds of music over any decent listening period.
It's not scientific of me to say this but I've been at the hi-fi thing, like you, since the 1950s and one sure thing in terms of evaluating speaker systems is their performance with choral music. It's not the only thing, but it's a biggy for me.
I'm pretty sure this is the reason:
http://www.ncvs.org/ncvs/tutorials/voiceprod/movies/ringfrm.mov
It sounds nasty as hell but the principle is clear, I think. That hump between 1500 Hz and 5000 with the 20 dB gain centered around 3000 is not chopped liver. The women can deliver some equally surprising stuff, as well.
Your speakers manage it just fine but there are plenty that don't.
I was speaker more to the guy who said that the ear is very sensitive to "distortion" in the mid band. I don't think that it is true that we are very sensitive to nonlinear distortion in the midband. Our hearing is very sensitive to other things like group delay and frequency response and I suggest those are the culprits. But everyone wants to hang all these problems on "nonlinear distortion" and I haven't seen anything that corroborates that.
Yeah. It seems reasonable since I sometimes can hear the same phenomenae in live performance (under particular circumstances) and the only nonlinear thing there is my hearing.
Earl I was talking to a mic designer from Pearl Microphones and asked him if HOM's could be a problem in those big 1" capsules. By HOM's here I mean the "drum head" resonances as demonstrated by for example Hans Jenny (and others). Ever see anything?
The higher order resonances in a diaphragm do affect the frequency response, but not the group delay. That because the waves don't propagate along different paths as they do in a waveguide. The higher order resonances in a diaphragm basically sets the upper limit on frequency response because the first HOM causes a null in the response. This is why the diaphragms are tensioned so high, to push this mode up in frequency.
Why that doesn't make sense to a number of people is that they make changes to the system which have zero impact on group delay, FR and such things - and hear, for them, very noticeable or even dramatic qualitative changes. So, they "hang" the causes on non-linear behaviours - because what else is there?
No, the cause of the audible change is nonlinear distortion, vs. linear distortion. A classic example is MP3 encoding - nonlinear distortion of the original signal - and everyone agrees how dreadful that sounds ...
How do you know it's not the timing errors the mp3 compression algorithm creates that makes it unlistenable
.