Before it all got distracted by a typo, the point of the video was to 'show' what was actually happening and compare it with what your sub was doing. i.e can your sub compete with the bass-players thumb. Where funk bass differs from other genres is that notes are started and stopped (often on the same string).
Most of the harmonics of bass played fall way above the range of subwoofer.
Not sure what are you trying to achieve here.
Properly designed sub covers range 20Hz to say 120Hz (ideally upper cutoff selectable by plate amp) as flat as possible, with lowest distortion possible, with the least amount of unwanted noises in the port, if it has one. It should be properly adjusted, so that it does not overpower or otherwise is heard, not drawing attention.
It should not be heard, just felt as natural extension of main speakers.
Most deep sounding instruments like bass or organ have important harmonic content way above sub, so its a midbass to convey the musicality here.
There is no such thing as musical sub, only well executed or poorly executed.
Not sure what are you trying to achieve here.
Properly designed sub covers range 20Hz to say 120Hz (ideally upper cutoff selectable by plate amp) as flat as possible, with lowest distortion possible, with the least amount of unwanted noises in the port, if it has one. It should be properly adjusted, so that it does not overpower or otherwise is heard, not drawing attention.
It should not be heard, just felt as natural extension of main speakers.
Most deep sounding instruments like bass or organ have important harmonic content way above sub, so its a midbass to convey the musicality here.
There is no such thing as musical sub, only well executed or poorly executed.
Long ago, there was a debate about preferring east-coast or west-coast sound. I think east-coast meant AR and dry clinical sound that measures well. Some thought choice of speakers related to choice of music genre. And music genre brings along with it the twerking that the recording producers apply to that genre.
Of course if there's a preference, then the speakers (or recordings) must be adding their own voice. That doesn't sound like what you'd want, eh. But then, like it or not, speakers (and genres) do have individual voices. For subs, nobody makes a "well executed" sub (motional feedback excepted) and so you are always choosing the features you want and those you don't care so much about.
For some decades, I've been listening to electrostatic speakers. They are the ultimate bland and are fair to all genres, as far as I can tell, but not adding cosmetics either. And they have no "Q" because ANY motion of the speaker, other than on the recording, is a bad motion. There's no "optimum Q" except in a math-convenience sense: dead is optimum... until the music plays.
Of course if there's a preference, then the speakers (or recordings) must be adding their own voice. That doesn't sound like what you'd want, eh. But then, like it or not, speakers (and genres) do have individual voices. For subs, nobody makes a "well executed" sub (motional feedback excepted) and so you are always choosing the features you want and those you don't care so much about.
For some decades, I've been listening to electrostatic speakers. They are the ultimate bland and are fair to all genres, as far as I can tell, but not adding cosmetics either. And they have no "Q" because ANY motion of the speaker, other than on the recording, is a bad motion. There's no "optimum Q" except in a math-convenience sense: dead is optimum... until the music plays.
Last edited:
The method may not seem very scientific in a "white lab coat" sense -- no calibration of hearing sensitivity for starters but when adjacent semitones could vary by multiple dB, and notch filters required Qs of 3+, this was significant.
The big surprise was that adding series resistance -- bringing the damping factor way down, and increasing the effective Qtc -- actually reduced the comb-filtering effect of the room. So when people talk about their bass being so "tight" because of a high damping factor or ultra-low Qtc and/or Qts, I really don't know what they mean. A plucked acoustic bass is made of thin laminated bits of wood. It makes no musical sense for the room to rattle when a subwoofer tries to reproduce it.
The big surprise was that adding series resistance -- bringing the damping factor way down, and increasing the effective Qtc -- actually reduced the comb-filtering effect of the room. So when people talk about their bass being so "tight" because of a high damping factor or ultra-low Qtc and/or Qts, I really don't know what they mean. A plucked acoustic bass is made of thin laminated bits of wood. It makes no musical sense for the room to rattle when a subwoofer tries to reproduce it.
Last edited by a moderator:
Perhaps I did not make my point clearly because people are spouting random data as a response. The first video shows a bass player. The question being does your sub clearly respond to everything his fingers are doing. In Jazz Funk the bass is often the lead, the artist playing in excess of 10 notes per bar in a solo, starting and stopping notes.
On my main system many of the notes roll together, losing definition and masking the skill of the artist. On my desktop system I can hear everything the bass player is doing.
This track provides a good example.
On my main system many of the notes roll together, losing definition and masking the skill of the artist. On my desktop system I can hear everything the bass player is doing.
This track provides a good example.
I think those taking their time to post here believe you are not understanding their points.
Music reproduction systems do not operate according to musical parameters but acoustic ones. There is no knob (or circuit) that adjusts "clarity" only "tweeter level" or "harmonic distortion".
There is a big challenge converting subjective experience into physical parameters. That is true even for parameters that many might think are simple like loudness. Never the less, only physical parameters can be modified with a soldering iron.
The assemblage of control in a HiFi today are partly historical accident and flowing from ease of design in 1940. When in days of yore folks wanted a boost for "presence", they boosted the band 2500-6000 Hz. No good reason not to have a "Presence" knob today. In fact, in Toole's latest edition, he advocates for tilting tone controls, not separate bass and treble in the familiar curves... and as usual, his argument is a good one.
In so far as you can translate the shortcoming you hear (or fail to hear) into a physical or otherwise "material" parameter, it can be addressed.
BTW, whatever makes you believe that little plastic disc really has a double-bass encoded on it? Was the mic located that same place as your head would have been in a jazz club? Not that a mic hears like a human ear. Or that the speakers in the recording control room are good ones? Or that the recording engineer wasn't hard of hearing in the treble range after years of loud exposure?
B.
Music reproduction systems do not operate according to musical parameters but acoustic ones. There is no knob (or circuit) that adjusts "clarity" only "tweeter level" or "harmonic distortion".
There is a big challenge converting subjective experience into physical parameters. That is true even for parameters that many might think are simple like loudness. Never the less, only physical parameters can be modified with a soldering iron.
The assemblage of control in a HiFi today are partly historical accident and flowing from ease of design in 1940. When in days of yore folks wanted a boost for "presence", they boosted the band 2500-6000 Hz. No good reason not to have a "Presence" knob today. In fact, in Toole's latest edition, he advocates for tilting tone controls, not separate bass and treble in the familiar curves... and as usual, his argument is a good one.
In so far as you can translate the shortcoming you hear (or fail to hear) into a physical or otherwise "material" parameter, it can be addressed.
BTW, whatever makes you believe that little plastic disc really has a double-bass encoded on it? Was the mic located that same place as your head would have been in a jazz club? Not that a mic hears like a human ear. Or that the speakers in the recording control room are good ones? Or that the recording engineer wasn't hard of hearing in the treble range after years of loud exposure?
B.
Last edited:
IMHO a sub would need to be quite bad that it did not reproduce the individual notes being played here.
Noting that the difference between a "tight" sealed sub and a "bit mushy" second-order reflex system is a change in group delay from 10ms to maybe 30-40ms... The reflex sub delays the different frequencies by different amounts, so you don't get a nice sharp impulse response all the way from 30-100Hz, so it doesn't slap you in the face as cleanly... but it will happily reproduce all of those notes, each being 100ms+ long.
There's also not much truly-low frequency content in the recording, so IMHO either system is going to work fine here because there is little content in the spectrum with high group delay. You're only going to tell the systems apart with big kick-drum and slapped-bass recordings; things with broadband square-edge transients where the ear can detect any frequency-dependent time-smearing. The tones of a bass will come through either a sealed or 2nd order reflex sub very well either way, with little or no damage to their volume envelopes, which is what you seem to be talking about wrt notes running together.
Once you get to 4th and 6th order bandpass systems, yes they sound bad. They tend to have high Q to get high SPL, i.e they mechanically store energy for a significant duration (noticeable at the level of notes running together), which smears the volume envelope of the notes, and they tend to produce only narrower ranges of frequencies. Every damn bass note is approximately the same pitch with a soggy exponential decay, and they are incredibly fatiguing to listen to. People presumably buy them because they're physically small and very loud, so are impressive until you realise what is going on. A 4-letter brand starting with B is the exemplar here, often mounted to the roof of a bad cafe or played at top volume in my mother-in-law's kitchen; they turn every bassline into a monotonic MOOP MOOP MOOP noise that has more to do with the enclosure than the music. There is no more transient response anymore, just a bunch of 70Hz tones where even the envelopes are smeared.
Is that the difference you are trying to describe?
I don't believe I mentioned a double-bass. Francine McGee is electric bass.
Prejudice much?
If you use IIR eq and filters to compensate and adjust lower frequencies, then a good implementation of a well built closed enclosure will be indistinguishable from a good implementation of a well built vented enclosure. Any analog based filter and/or compensation, IIR filters and eq add delay, usually just as much and perhaps in some cases more, than a properly designed vented enclosure might.
If you have the computational power or perhaps time to wait for the sound to go through properly implemented FIR eq and filters they would also be indistinguishable.
I agree that poorly made vented enclosures are absolutely horrible, but using that as a basis of comparison against a properly engineered sealed implementation is not what I would call my personal preference. Pears and avocados...
I'm going to say that's a hard NO. At any time there are three systems in my room. Main, Desktop. and Test. In addition to this if I cite a track as an example, my experience dictates I've probably listened to the same track 1000 times on at least 50 different systems.
I am a little amused by those who believe sound reproduction is a perfect science - it is not and can never be. And whilst you continue to cling to free software as being the Oracle, you guys have no idea where you're trying to go. Yes, the software works fine so long you are at sea-level, it's not about to rain, and you don't move the sofa or draw the curtains.
I was the engineer. I am ashamed of this ****.
I am definitely prejudiced now, but a prejudice from direct experience in building, analysing and listening to these systems. I have built a 4th order sub, it was a waste of MDF - for the kinds of music that I like. We have all I think experienced the tiny, commercially-available speakers that I mention, and they really suck. I am not sure what you could do to a sealed design to make it sound anywhere near that bad; even cranking the box size down so that Q>1... except that a sealed design will be at 10x-100x larger in physical volume, which is why these things exist.
IMHO these high-order designs have a place for the efficient reproduction of low tones where transient response isn't critical, e.g. if you want to build an electronic organ or want to compete in a subwoofer SPL competition that is mostly continuous tones. They are demonstrably ineffective at reproducing broadband transients, hence my prejudice.
4th and 6th order enclosures are an attempt, via the mechanical storage of energy, to cheat Hoffmann's Iron Law of bass. And arguably they succeed at that: you can get high SPL at relatively low frequency from a small box without spending kW but there is a cost, which is that they are strongly resonant systems that achieve that SPL and LF extension through the storage of energy. That storage of energy means smeared transients and smeared note envelopes, which, to me, is "unmusical" per the thread; the enclosure becomes an instrument unto itself rather than a reproduction system.
I have not yet seen anyone fix a high-order enclosure via digital filtering. I have my doubts that it is possible to fix by a linear causal filter, but hey, I'm not gonna stand in the way of anyone who does think they can do it.
In general: I like smooth roll off myself, so not looking to argue about that.
I got caught up in, what seems to me, the opinion that any sealed enclosure is automatically superior to any vented enclosure, trying to point out that if you use the path of analog derivative means of compensating low end performance of a sealed design, you may in some cases end up with MORE group delay than if you made a properly designed vented enclosure, and you still would not get the biggest benefit from a vented design: cone control. That you can also let a bit of heat out through a vent is merely an unexpected bonus.
Simulate the whole network of filters and compensation it adds up, there's no free lunch in sealed boxes either:
a) You "gain" higher power consumption
b) Need more xmax
- and also end up with mostly the same group delay all things considered.
That you can use a box half the size of a compareable vented design is moot, because you still end up with point a) and b).
I used to share your opinion that closed boxes are superior, and actually set out to prove it through making a properly designed vented enclosure. After looking extensively at any and all variables for two years, I failed.
There's no free lunch, there's only: Pick your poison.
Everything is a compromise, perfection does not exist.
I got caught up in, what seems to me, the opinion that any sealed enclosure is automatically superior to any vented enclosure, trying to point out that if you use the path of analog derivative means of compensating low end performance of a sealed design, you may in some cases end up with MORE group delay than if you made a properly designed vented enclosure, and you still would not get the biggest benefit from a vented design: cone control. That you can also let a bit of heat out through a vent is merely an unexpected bonus.
Simulate the whole network of filters and compensation it adds up, there's no free lunch in sealed boxes either:
a) You "gain" higher power consumption
b) Need more xmax
- and also end up with mostly the same group delay all things considered.
That you can use a box half the size of a compareable vented design is moot, because you still end up with point a) and b).
I used to share your opinion that closed boxes are superior, and actually set out to prove it through making a properly designed vented enclosure. After looking extensively at any and all variables for two years, I failed.
There's no free lunch, there's only: Pick your poison.
Everything is a compromise, perfection does not exist.
Agreed, it is all tradeoffs for which there is no free lunch. While I am of the opinion that any good (Q=.5 to 0.7) sealed sub probably always sounds better than any ported sub of the same F3, that ignores the fact that the sealed version will probably require larger piston area therefore be more expensive and/or less sensitive.
If you put the same driver in two different enclosures (sealed, ported), the ported will reach lower and the "bad" group delay will all be below the corner frequency of the sealed version, i.e. there are only bad effects on the ported box where the same driver in a sealed enclosure is ineffective.
I am not sure of the group delay on Linkwitz Transform though. I suspect they can be close to min-phase but have not confirmed that numerically or by measurement, therefore equalising electronically does not introduce the same phase issues that a ported box does. But of course it doesn't actually improve the physical sensitivity down low at all, and the little 50Hz sealed box that's been LT'd down to 30Hz will not be capable of any greater SPL for having been transformed in that way. It's just high-frequency attenuation...
As you say, pick your poison. As I see it, the choice are approximately:
And while I'm sure you could build a 4th order box with similar group delay to a ported box, I suspect the volume+sensitivity+F3 would no longer be any improvement over the 2nd order.
So you can EQ a sealed box (at significant cost in drivers and power amplifiers) to achieve the same F3 as a ported box while - I think - retaining a better impulse response (minimal group delay) than the ported box, but I don't think you can 2nd, 4th or 6th order box to have the low group-delay and therefore impulse response of the sealed.
Hence my prejudice
If you put the same driver in two different enclosures (sealed, ported), the ported will reach lower and the "bad" group delay will all be below the corner frequency of the sealed version, i.e. there are only bad effects on the ported box where the same driver in a sealed enclosure is ineffective.
I am not sure of the group delay on Linkwitz Transform though. I suspect they can be close to min-phase but have not confirmed that numerically or by measurement, therefore equalising electronically does not introduce the same phase issues that a ported box does. But of course it doesn't actually improve the physical sensitivity down low at all, and the little 50Hz sealed box that's been LT'd down to 30Hz will not be capable of any greater SPL for having been transformed in that way. It's just high-frequency attenuation...
As you say, pick your poison. As I see it, the choice are approximately:
- sealed unequalised: large/expensive drivers with poor sensitivity (low EBP required to get low F3) but best-possible group delay
- sealed with LT: very high power and Xmax requirements and no improvement to peak SPL capability, but still probably(?) best-possible group delay
- 2nd order port: greater box volume and poorer group delay just above the achieved F3, but you can use a higher-sensitivity driver to achieve the F3, or use the same driver and get much lower F3
- 4th, 6th order: usually terrible group delay, but some combination of very low F3, tiny box and/or high sensitivity - Hoffman's nemesis!
And while I'm sure you could build a 4th order box with similar group delay to a ported box, I suspect the volume+sensitivity+F3 would no longer be any improvement over the 2nd order.
So you can EQ a sealed box (at significant cost in drivers and power amplifiers) to achieve the same F3 as a ported box while - I think - retaining a better impulse response (minimal group delay) than the ported box, but I don't think you can 2nd, 4th or 6th order box to have the low group-delay and therefore impulse response of the sealed.
Hence my prejudice
If one is a competent designer, there is no distinction other than cost. My "music" sub, I originally had two but down sizing means one, is a critical Q sealed, Peerless XLSS paper carbon. It darn near matches roll-off to room gain. I have no EQ on my sub. Don't need it. My HT I run a pair of 10's, push pull. Could not get the Q below .6 due to physical space. Still kept THD below 2% at 96 dB, 20 Hz. Now, does it shake the neighbors house with special effects? No. I can't stand that anyway. Some seem to think they need 120dB @ 10 Hz. I think I need my hearing and bowls in tact. But I could build you that and it be just as clean and tight as my living room sub. Probably 4 12's, but that is a guess. I don't care for 15's as control of the cone gets too hard. There is no good engineering reason to need different subs for SE and Bass as we are talking only an octave.
I accept someone may hear a 15 or 18 they like better than their current, but it most likely has to do with remaining within the drivers design that the size. It also may have even more to do with the capability of the mains and the crossover. All that "impact" stuff is not actually the sub.
I accept someone may hear a 15 or 18 they like better than their current, but it most likely has to do with remaining within the drivers design that the size. It also may have even more to do with the capability of the mains and the crossover. All that "impact" stuff is not actually the sub.
One thing to consider is that the shape of the group delay (not just height) could indicate whether or not the sub will generate ringing. With a 4th order roll off (vented), even a flat frequency response can have still show ringing in the step or impulse response. FR doesn't need a 'peak' as such, just a sharp corner from horizontal to vertical.
This is the same reason (one of) why Linkwitz-Riley and Bessel filters are preferred over Butterworth (above 2nd order) for XO slopes.
With that in mind, re-tuning an enclosure to get rid of group delay peaks can give curious results like a smaller box, and slight mid-bass hump. And may be the secret sauce behind the "fast" "punchy" bass of many commerical offerings, in spite of their other shortcomings.
These are my very cheap main speakers (sim is just 1 in halfspace, flat surface - no corner, not the pair), occasionally take them outside to liven up the neighbourhood (people are all smiles, never heard so much as a rumour about a bad comment).
Instead of boosting the lows I pad down to get even response.
Instead of boosting the lows I pad down to get even response.
