Jump to content
IGNORED

How low can you go?


Recommended Posts

Bluesman,

 

The slight mistake you make in referring to instruments not playing exactly together is that it is not about that. What it *is* about is the one key from a piano (etc.) producing it's whatever sound which is formed by the harmonics of the whole instrument and *that* needs to go together at the exact same point in time.

 

Peter

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment

2c and just saying :

 

Our D/A converter can operate in two electrically different modes; one which is the most normal (hence it is so in any DAC) where 1000Hz will show a phase angle difference between left and right of 0.02 degrees (at 352.8 input) and one where the phase angle difference is 0.00000 degrees. The former presents a totally fuzzy sound relative to the latter.

Of course the higher the frequency the more the phase angle difference is in the normal mode.

But surprisingly, what the 0.00000 mode does is creating the "crazy" bass separation between left and right. Really so; I never looked at the pahse angle difference at e.g. 100Hz but just device the 1000Hz number by 10. So that is 0.002 degree and on this very long wave this is the most easily audible.

 

To keep in mind : In my sitiation there is zero smear because of ringing filters, because my filters do not ring a bit. This is crucial for observing this (the time difference is now 100% accurate).

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
Bluesman,

 

The slight mistake you make in referring to instruments not playing exactly together is that it is not about that. What it *is* about is the one key from a piano (etc.) producing it's whatever sound which is formed by the harmonics of the whole instrument and *that* needs to go together at the exact same point in time.

 

Peter

But they don't, Peter. Because of the different masses of the many strings and moving / vibrating parts in a piano, the same key will produce a different spectral splash each and every time it's hit. Players aren't perfectly consistent either, so the force with which a note is initiated also changes the temporal relationships among the harmonics within it. The differences are not grossly audible, but they're measurable and easily demonstrated with a Fourier transform display. There's simply not enough temporal consistency in the relationships among the components of that note for a tiny shift in physical positioning of a driver to affect the sound - there's more variance in the source than there could ever be in the transducers reproducing it. And I don't know how you'd be able to tell if the sound was "accurate", as the notes themselves do not sound identical even if striking the same key, fretted string etc repetitively.

 

This is true for every instrument in isolation and more so for combinations of them. When you factor in the intermodulation among multiple tones generated by multiple instruments, I don't see how it's even possible that shifting a subwoofer 1/8" could have an audible effect on sound quality apart from room factors. This is the kind of belief that requires blinded testing to validate - unless a blinded listener can reliably and repeatably tell me that a driver was shifted a fraction of an inch from listening to its reproduction of music, this simply isn't happening.

 

Yes, I've tried all this and more. What I discovered was a bunch of hype and little substance. I made many recordings on my Crown SX724 with Sony condenser mikes of my Martin D-28, my Yamaha G1J grand, my Selmer alto sax and my Getzen Eterna trumpet to use as test sources. With no difference at all except a change of strings on the guitar, friends told me with certainty in level-matched blinded tests that I'd changed the bias on my output tubes, replaced my caps and resistors, changed the crossover frequencies in my electronic crossover, etc etc etc. No one ever considered that what they're hearing may be variance in the source. And this was true for every instrument. I changed the mouthpiece on my trumpet from a Bach to a Denis Wick. I changed reeds in my sax. Each and every time, the sonic alterations were perceived as electronic - and even when I brought out the instruments and played them live in the same room in which I'd recorded them, there was disbelief.

 

Take it for what you think it's worth. Best regards -

 

David

Link to comment

b-b-b-but David,

 

How can you ever think that it is "just allowed" to time shift e.g. the lowest octave of a piano by means of some digital processing (that is what it comes down to) because all the vibrations in a piano are messy to begin with ? So say we shift the fundamental somewhat, but the harmonics stay put because dealt with by a different driver. And you think we won't hear this ? the whole instrument will change !

 

I'm afraid all you prove (not explicitly and for sure not negatively intended) is that your test procedures were off. But for example (nothing specific about your "procedures") *I* would never ever say that subwoofers can be positioned by means of moving them 8 inch, 8 meters or 8 feet. That will never work. Also it is easy to prove that ONE sub will never work and I can personally testify it is a hell of a job to tune two of them. And then it needs analysers to do it in the first place. So all who claim this can be done with any good result ... not agreed (at all).

 

What I try to say is that most speakers are not properly time aligned in the first place. So, just that one box with woofer and say mid. Do you perceive that ? probably not. Unless ... unless you have the experience that they *are* properly time aligned, then you will know. So what this will come down to is that when not OK you can mangle with almost everything and hardy will perceive a difference because not right in the first place. But low and behold when woofer and mid behave like one piston and *then* something like a sub is brought in; it will destroy all over (assuming overlap at the low end) and you won't get it right. Not by normal audible means.

And to name something, with a normally fast woofer and also fast mid, the both will be out of order for something like 3 ft to begin with. I am not talking impulse response, but phase aligned. Can only be solved by putting the woofer one cycle behind minus that 3 ft. Still speakers never show such a distance. Strange.

 

All 'n all, the source (sound) must remain consistent, no matter how strange the instrument is organized. And if we can't agree on that, well, then not !

;-))

 

Regards,

Peter

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
"most speakers are not properly time aligned in the first place" and "all the vibrations in a piano are messy to begin with".

 

All 'n all, the source (sound) must remain consistent, no matter how strange the instrument is organized. And if we can't agree on that, well, then not !

;-))

 

Regards,

Peter

Yes, yes! We do agree. Where we differ is that I believe the variance in the source is so much greater than the variance in the system that tiny changes of a few degrees in phase from fundamental to harmonics can't possibly render so imprecise a sonic image with precision. One may or may not hear differences (personally I do not, and I'm a professional musician with a normal audiogram who's been playing for 60 years and performing for pay for 50). But even if you hear those differences, they're not more accurate - they're just different. They may be more pleasing to some, but they're just different - they're objectively neither better nor worse.

 

I'll use my favorite puzzler: you prefer the sound of a Steinway to that of a Baldwin. You hear a piano recording over someone else's system that sounds like a Steinway to you. Can you possibly know whether the instrument was a well recorded and reproduced Steinway or a badly recorded and reproduced Baldwin?

 

As for my testing methodology, recording the same solo guitar or trumpet passage with no changes at all except the strings on the guitar and the mouthpiece on the trumpet, then playing them randomly at matched levels to listeners in the same positions in front of the same playback system who don't know which is which seems to me to be pretty sound science (no pun intended....well, maybe a little).

 

Hey, man - we're discussing, not disagreeing. Discussion is how we all share ideas and learn, and I hope you've learned something from me just as I've learned from you. If there were one correct answer. everybody'd have it right.

Link to comment
b-b-b-but David,

 

How can you ever think that it is "just allowed" to time shift e.g. the lowest octave of a piano by means of some digital processing (that is what it comes down to) because all the vibrations in a piano are messy to begin with ? So say we shift the fundamental somewhat, but the harmonics stay put because dealt with by a different driver. And you think we won't hear this ? the whole instrument will change !

 

I'm afraid all you prove (not explicitly and for sure not negatively intended) is that your test procedures were off. But for example (nothing specific about your "procedures") *I* would never ever say that subwoofers can be positioned by means of moving them 8 inch, 8 meters or 8 feet. That will never work. Also it is easy to prove that ONE sub will never work and I can personally testify it is a hell of a job to tune two of them. And then it needs analysers to do it in the first place. So all who claim this can be done with any good result ... not agreed (at all).

 

What I try to say is that most speakers are not properly time aligned in the first place. So, just that one box with woofer and say mid. Do you perceive that ? probably not. Unless ... unless you have the experience that they *are* properly time aligned, then you will know. So what this will come down to is that when not OK you can mangle with almost everything and hardy will perceive a difference because not right in the first place. But low and behold when woofer and mid behave like one piston and *then* something like a sub is brought in; it will destroy all over (assuming overlap at the low end) and you won't get it right. Not by normal audible means.

And to name something, with a normally fast woofer and also fast mid, the both will be out of order for something like 3 ft to begin with. I am not talking impulse response, but phase aligned. Can only be solved by putting the woofer one cycle behind minus that 3 ft. Still speakers never show such a distance. Strange.

 

All 'n all, the source (sound) must remain consistent, no matter how strange the instrument is organized. And if we can't agree on that, well, then not !

;-))

 

Regards,

Peter

 

 

Pete....while I often enjoy your posts, in these particular discussions I have absolutely no idea what you're trying to share. We can't talk about time alignment of drivers in an enclosure who's patterns of directivity don't match....there's absolutely no value in it as the low freq spectra is primarily Omni and therefore it's energy becomes a product of the drivers direct energy and the room......That being said, aligning the drivers acoustic centers to the listening position across the spectrum can ONLY be achieved through DSP or the use of a 100% cardoid speaker system......impractical or virtually impossible.

 

If you can get near constant directivity down to 500hz, that's an outstanding accomplishment. There's no commercial home audio speaker that does this and few pro applications that can within a small environment.

 

All you can hope for in the lower octaves is a somewhat smooth frequency response....proper phase quadrature/alignment or alignment within time as many put it.....that's an exercise in futility.

 

I'd add that it's best to not refer to drivers as 'fast' or 'slow' as neither is a function of the driver, but instead the alignment which really relates to impulse response. If the driver 'rings', that doesn't make it 'slow'.....the fundamental still happens just as 'fast' as another. These unsuited applications of verbiage that audiophiles tend to like to throw around don't translate very well...or often. I bring this up as you clearly understand impulse and step response in principle so if you look at the function mechanically, you'll understand what I'm talking about.

Link to comment

Mayhem,

 

It is funny, but apparently we landed on matters which are suddenly more beyond general knowledge that ...

without words we all understand ?

But this is the point - without too many words it should be obvious. But read between the lines, because it just is. Ehm, for those with the knowledge and obviously not for all. Few examples :

 

Can only be solved by putting the woofer one cycle behind minus that 3 ft. Still speakers never show such a distance. Strange.

 

This small sentence tells it all ... DSP of course ! How to do it otherwise. It also suggests that no normal high fidelity speaker contains DSP and thus they "have to be" wrong. Btw, yesterday I started adding text about getting away with half of the length because of twisting the phase 180 degrees) but I scratched that because from one come the other and that too I assumed known (for those who want to read anyway).

 

And to name something, with a normally fast woofer and also fast mid, the both will be out of order for something like 3 ft to begin with. I am not talking impulse response, but phase aligned.

This really requires more explanation and I didn't want to go into it because it is a fairly large subject in itself. But the sheer fact that I mentioned *and* fast *and* not impulse response, should tell something. Of course, when it's not explained one can say "ok, the guy is dizzy today".

Ok, I am always anyway. ;-)

 

Aligning drivers by impulse response is useless because this is (relatively !) hardly used in music. What is used is continuous "tones". So, a bass playing - assumed is plays in the woofer and in the mid - must be phase aligned foremost. And of that is detrimental to impulse response - bad luck. Just different thinking / approach.

Assumed you can agree with this (not that speaker manufacturers think like this), what's next is the, say, response speed of the driver. So, a woofer will always have lag compared to the mid. This is not really impulse response (though it sure is related) but just that there's more power needed to move the woofer which thus implies lag (more than a mid). It moves at the same speed (like at 260Hz in the suggested XOver region) but is behind somewhat. So, must be put forward because of this.

 

We can't talk about time alignment of drivers in an enclosure who's patterns of directivity don't match....

I gave one example of me never trying to move a sub to get things right, but this (quote of your text) is another. And many more; Piles of prerequisites exist to get it right, or say the best possible. And to give the most vague example : one of the prerequisites is to move out the sub in the first place.

Maybe it helps (and I hope not the contrary) if I tell that the speaker I talked about in post #15 of this thread is not a stupid hobby project; it is a project of many many years with explicit goals and what ended up in this commercial speaker (first pair was shipped a month ago; 3 playing a homes at this moment).

So I started out with some controversial software, then a controversial DAC to emphasize my ideas and now this. :-)

 

Regards,

Peter

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
If you can get near constant directivity down to 500hz, that's an outstanding accomplishment. There's no commercial home audio speaker that does this and few pro applications that can within a small environment.

 

To be clear :

 

When I put up a sine test signal of something like 26Hz (as low as possible while it still is the most clearly audible) it is totally audible whether this signal springs from the left or the right speaker. And min you please, this is pure sine so actually the most difficult. When this would be an upright double bass there is super pinpointed positioning but now aided by the higher frequencies of the metal strings.

 

This *is* related to the radiation pattern which is why I prefer horns in the first place.

 

I tell you about this bass separation because I don't know of any DAC outside the Phasure that can do this except for the Alpha which comes close. So there is a very fair chance that (at your place) even the upright bass with its higher frequency harmonics will show from the dead middle let alone sines. All bass does. But not so here - not at all.

This is important, because it is a reference to physics which apparently can do it.

 

Btw, this source of the sound is not to be confused with directivity itself; both are not the same (the latter can be guided while the former is as (good as it) is).

 

The subject is way more large because in the end it is all about the quality of "the wave" and (my famous hobby horse) the better that quality the less standing waves; this starts with the software (always bit perfect) - but never mind.

Never mind ? well, it is crucial because no concert level bass would be able to play in any (decorated) room if this is not arranged for. So to be highly controversial once again : imagine fairly continuous bass in the low 30Hz region playing at close to 100dBSPL in a room of 36x24x10ft (12x8x3m) and totally nothing goes wrong anywhere (walls are all glass/concrete, floor carpeted). With "anywhere" I also refer to each corner no matter where.

Now, I am pretty confident that everybody will say "can't be" and "smoked too much". But that's a good thing, because now you know how different things can be to what "we" are used to.

But the difference ? this bass is undistorted and that does some tricks.

 

More tricks ? Mentioned 100dBSPL is according to @1m standard. And at 6 meters ? ~96dBSPL. So think about that too and how two folded implications spring from this; 1. So what about back bouncing back wall ? (not detrimental) 2. how about bass directivity (quite a lot - but that's horns).

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment

Or.... Align the **** speakers and subs. THEN treat the room. THEN fiddle with DRC if you feel the need. If you can't get the first job done right, the rest won't make up for that loss. Ever. Correct direct radiation will make most room problems seem petty in comparison.

Mac Mini 2012 with 2.3 GHz i5 CPU and 16GB RAM running newest OS10.9x and Signalyst HQ Player software (occasionally JRMC), ethernet to Cisco SG100-08 GigE switch, ethernet to SOtM SMS100 Miniserver in audio room, sending via short 1/2 meter AQ Cinnamon USB to Oppo 105D, feeding balanced outputs to 2x Bel Canto S300 amps which vertically biamp ATC SCM20SL speakers, 2x Velodyne DD12+ subs. Each side is mounted vertically on 3-tiered Sound Anchor ADJ2 stands: ATC (top), amp (middle), sub (bottom), Mogami, Koala, Nordost, Mosaic cables, split at the preamp outputs with splitters. All transducers are thoroughly and lovingly time aligned for the listening position.

Link to comment
Or.... Align the **** speakers and subs. THEN treat the room. THEN fiddle with DRC if you feel the need. If you can't get the first job done right, the rest won't make up for that loss. Ever. Correct direct radiation will make most room problems seem petty in comparison.

 

No such thing as aligning the subs to the mains if we're talking the last few octaves. Midbass?......go right ahead if you're using bass horns or dipoles.

Link to comment
To be clear :

 

When I put up a sine test signal of something like 26Hz (as low as possible while it still is the most clearly audible) it is totally audible whether this signal springs from the left or the right speaker. And min you please, this is pure sine so actually the most difficult. When this would be an upright double bass there is super pinpointed positioning but now aided by the higher frequencies of the metal strings.

 

This *is* related to the radiation pattern which is why I prefer horns in the first place.

 

I tell you about this bass separation because I don't know of any DAC outside the Phasure that can do this except for the Alpha which comes close. So there is a very fair chance that (at your place) even the upright bass with its higher frequency harmonics will show from the dead middle let alone sines. All bass does. But not so here - not at all.

This is important, because it is a reference to physics which apparently can do it.

 

Btw, this source of the sound is not to be confused with directivity itself; both are not the same (the latter can be guided while the former is as (good as it) is).

 

The subject is way more large because in the end it is all about the quality of "the wave" and (my famous hobby horse) the better that quality the less standing waves; this starts with the software (always bit perfect) - but never mind.

Never mind ? well, it is crucial because no concert level bass would be able to play in any (decorated) room if this is not arranged for. So to be highly controversial once again : imagine fairly continuous bass in the low 30Hz region playing at close to 100dBSPL in a room of 36x24x10ft (12x8x3m) and totally nothing goes wrong anywhere (walls are all glass/concrete, floor carpeted). With "anywhere" I also refer to each corner no matter where.

Now, I am pretty confident that everybody will say "can't be" and "smoked too much". But that's a good thing, because now you know how different things can be to what "we" are used to.

But the difference ? this bass is undistorted and that does some tricks.

 

More tricks ? Mentioned 100dBSPL is according to @1m standard. And at 6 meters ? ~96dBSPL. So think about that too and how two folded implications spring from this; 1. So what about back bouncing back wall ? (not detrimental) 2. how about bass directivity (quite a lot - but that's horns).

 

Aaaaah.......I get it now. Your explaining what 'audiophile' bass should be.....and how to deliver it.......either massive bass horns or large format dipole bass arrays using multiple large surface area drivers with high linear travel and massive current to drive them without added harmonic distortion?......well yes, then I agree. Not practical, but gets the job done.

 

Always a compromise though Pete. Those systems are BIG and and heavily dependant on not so convenient placement. Even average dipole bass systems need to be fairly large and to hit your target levels at 30khz......a sealed box subwoofer or two would be required to supplement the lowest octave. That's the problem with dipoles. And neither of these types of systems can provide the time or phase alignment you suggest. Again, only advanced DSP can do that.

 

Agreed.."..properly done horns and waveguides are the way to go.....not convenient or family friendly but......

Link to comment
No such thing as aligning the subs to the mains if we're talking the last few octaves...

 

[Edit: I've only had to use the supplement method in my system, I forgot that the timing handoff of using an HPF doesn't mandate any large mains-vs. sub driver offset] Start with the subs just a hair closer to your ears than the the mains' bass driver. It's one of the most pleasurable activities in audio; you just hear the bass snapping ever closer to perfection.

Mac Mini 2012 with 2.3 GHz i5 CPU and 16GB RAM running newest OS10.9x and Signalyst HQ Player software (occasionally JRMC), ethernet to Cisco SG100-08 GigE switch, ethernet to SOtM SMS100 Miniserver in audio room, sending via short 1/2 meter AQ Cinnamon USB to Oppo 105D, feeding balanced outputs to 2x Bel Canto S300 amps which vertically biamp ATC SCM20SL speakers, 2x Velodyne DD12+ subs. Each side is mounted vertically on 3-tiered Sound Anchor ADJ2 stands: ATC (top), amp (middle), sub (bottom), Mogami, Koala, Nordost, Mosaic cables, split at the preamp outputs with splitters. All transducers are thoroughly and lovingly time aligned for the listening position.

Link to comment
Those systems are BIG and and heavily dependant on not so convenient placement.

 

Depends on what we call big ? In my case we talk about 5ft high, 3.1ft wide and 2ft deep. And convenient placement ? the things won't care. I have them 2ft from the wall behind and then to think this is open baffle ...

 

Even average dipole bass systems need to be fairly large and to hit your target levels at 30khz

 

Though this open baffle this is not exactly a dipole of course. And straight to 19Hz without audible distortion (88dBSPL reference level).

Yes with DSP.

;-)

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...