i knew we can't really tell pitch at very low frequencies and now you're telling me also at very high frequencies
so what are the frequencies at which we can most accurately judge what we hear ?
so what are the frequencies at which we can most accurately judge what we hear ?
Although young humans with normal hearing can hear sounds with frequencies between ∼20 and 20,000 Hz, only repetition rates between ∼30 and 4000 Hz elicit a pitch sensation that is salient enough to carry melodic information (Attneave and Olson, 1971; Pressnitzer et al., 2001).
So 30Hz to 4000Hz appear to carry the most salient pitch information from research.
Theres a satisfying correlation with the piano range from 27.5 Hz to 4186 Hz.
So 30Hz to 4000Hz appear to carry the most salient pitch information from research.
Theres a satisfying correlation with the piano range from 27.5 Hz to 4186 Hz.
Evolutionary biology would speculate we are a product of ancestors who could perceive speech really well to survive aggression, attract a mate and co-operate within the tribe
Getting back to audio this frequency perception theory may predict that an audio driver that can
seamlessly deliver 200Hz - 5000kHz would be appreciated for vocal and instrumental integrity.
Getting back to audio this frequency perception theory may predict that an audio driver that can
seamlessly deliver 200Hz - 5000kHz would be appreciated for vocal and instrumental integrity.
this is why i don't believe in 3-ways.
in a 3-way the woofer has to reproduce bass and that bass then modulates the vocals and other important instruments.
you need to cut off right at about 250 hz which is more or less the divide between things like bass kick and things like vocals and other instruments.
if you look at music in FFT you will see they master it that way - they kind of try not to overlap the two so it doesn't turn into a mess and we should design speakers the same way they mix it.
i would then put another crossover at around 80 hz so the kick fundamental can be handled by an extra powerful driver in a separate enclosure you can put in room corner or such.
apologies to all the classical and jazz people ... they must be wondering - but what is bass ?
in a 3-way the woofer has to reproduce bass and that bass then modulates the vocals and other important instruments.
you need to cut off right at about 250 hz which is more or less the divide between things like bass kick and things like vocals and other instruments.
if you look at music in FFT you will see they master it that way - they kind of try not to overlap the two so it doesn't turn into a mess and we should design speakers the same way they mix it.
i would then put another crossover at around 80 hz so the kick fundamental can be handled by an extra powerful driver in a separate enclosure you can put in room corner or such.
apologies to all the classical and jazz people ... they must be wondering - but what is bass ?
This is the same conclusion i have. Mid (250hz and up) then bass (250hz to 60) then sub.
But if done right your loudspeaker should play everything right jazz, industrial, opera, drum&bass or dubstep.
Acoustic instruments can produce a 'large' amount of low end.
But if done right your loudspeaker should play everything right jazz, industrial, opera, drum&bass or dubstep.
Acoustic instruments can produce a 'large' amount of low end.
80Hz is the chest-thumping pants-flapping frequency, kick drum induced. It is not that low. I can't see how putting a xover frequency there would make things better. It will only lead to more group delay. Using a FIR filter at these frequencies requires a lot of taps.
well my design goes back and forth between 80 and 60 hz for sub XO ...
using a subwoofer essentially centers the woofer voice coil in the gap by offloading the lowest frequencies to the sub. which is especially beneficial if you're using a push-pull woofer like JBL Differential Drive or Eighteen Sound Tetracoil which have a symmetrical BL curve ...
i mean if you listened to the music i posted as a sample of what i listen to there will be fairly strong fundamental around 40 hz that is about 6db louder than everything else ... it's obviously not like that with classical or jazz
using a subwoofer essentially centers the woofer voice coil in the gap by offloading the lowest frequencies to the sub. which is especially beneficial if you're using a push-pull woofer like JBL Differential Drive or Eighteen Sound Tetracoil which have a symmetrical BL curve ...
i mean if you listened to the music i posted as a sample of what i listen to there will be fairly strong fundamental around 40 hz that is about 6db louder than everything else ... it's obviously not like that with classical or jazz
Last edited by a moderator:
What is bass? that'll be the 18Hz from the canon of 1812, or the 60Hz ( approx ) of kettle drums. A problem with tweeters that rarely gets mentioned is the constantly changing temperature, hence resistance of the voice coil, most of the time a tweeter will be doing very little ( ie cold ) then with a couple of cymbal crashes they're suddenly awoken and heat up, increasing the resistance, messing up the balance and crossover. This is why I like the idea of a small full range ( or two wired in series, with a cap across the lower one ) augmented with a sub under each speaker.
Nah, i understand all you're saying quite well. We were just hung up, misunderstand terminology in play.
As I use Keele's formula for pattern control frequency all the time, for sizing MEH builds. (what you call directivity control)
And have built floor to ceiling line-arrays...... and MTMs...and yada 🙂
BTW, i agree with your simplified 3 part speaker assessment. Bass, loud, and vocals.
My test for quality of vocals has become intelligibility (can i accurately hear the lyrics) and the ability to separate multiple or backing vocals.
In essence, vocals' quality has become my test for accurate tuning.
I used overall tonality to tune for a long time; but now i view overall tonality as the "coarse control tuning knob", and vocal intelligibility as the "fine control knob"
I mostly agree. Taps are not an issue by themself, the latency they induce can be though. There is way to lessen this if needed as long as you don't have to monitor something realtime.
I prefer 60hz for sub to put a xover. Some satellite and norms use 80hz though ( Thx). This keep location issues out of concern and allow some freedom for 1/4wavelength location of sub.
60hz gives even more room for location choice and is usually well below Schroeder frequency so integrate well in most rooms ( and group delay is dominated by room at those freq so even 50ms can be hidden by the room behavior anyway).
it's just that all instruments except vocals are arbitrary.
some guy makes a violin and declares this is what violin must sound like. but that's just his personal opinion. in fact the violin itself is a social construct ...
i listen to electronic music because it doesn't cling to arbitrary primitive instruments made using same technology as what is used to make shovels. even music made with synthesizers in the 80s and such i don't like. i prefer music made entirely in the computer from start to finish - never crossing over into analog domain ... except for Vocals.
human voice and 1" electrostatic microphone capsule are the only valid instruments in the physical domain.
everything else must be digital from start to finish.
very few masters can make acoustical music not painful to listen to ... most of it sounds horrible in itself regardless of what you play it on ... yes all the instruments are "natural" but nails on a chalkboard are natural too ...
Every crossover point introduces a phase shift. Steeper filters introduce larger phase shifts. More precisely, 90 degrees per filter order. Each additional filter order makes the filter slope 6dB per octave steeper. At 60 or 80 Hz, you would probably want at least a 4th order filter. That means at least a 360 degree phase shift.
Group delay is a way to look at the slope of the phase shift and is measured in milliseconds. The lower the frequency, the larger the group delay. It has been demonstrated that large group delays are audible.
well all dance club systems use subwoofers.
on the other hand nobody has ever blown a subwoofer by listening to opera.
??? If they added the harmonics to the music it would be harmonic distortion, the kind most people think of when you say distortion. This test is only about hearing noise (unrelated test tones) under the music. It has nothing to do with harmonic distortion. Its a stupid test. The only thing it shows is how much you can hear a constant 60hz hum (or other single freq. tone) under the music. Which is completely dependent on the music. I can detect a 60hz tone 80db under a flute solo, and just destroyed there graph. And so what.
