In a separate post I like to mention this :
I feel we are on the verge of a new era in audiophiledom. Things start to happen in - or of a dimension which wasn't thought possible. I say this because *I* thought it was not possible, having worked on the sub-mm localization with phase angles and such.
As many of you will know, by now I "claim" that the 3D representation of recorded instruments, played back through loudspeakers, is not possible because of a lack of high frequency (without explaining further, my local positioning system (LPS) endeavors from pre-Phasure times worked in the 2.4GHz band, while here we have a max 20KHz band only). But I start to change my mind ...
The LPS lived by the grace of more than one frequency (a dirty trick), while audio, although comprising of say
100 frequencies we can hear, is still way too low in frequency itself to allow "re-positioning" by means of phase angle (and ITD vs ILD and whatever we humans work with). Very roughly, if 20KHz is 100 times lower in frequency than 2GHz, the theoretical accuracy of the "system" is 100 times less than the sub 1mm accuracy. Make that 1mm and the accuracy for audio would be 10cm which is insufficient for "pinpointing" and (re)building an image in mid air.
Can't follow ? or not agree ? then better stop reading.
If I mention a 100 frequencies (like they could spring from a few instruments, each expressing multiple frequencies like the easy example of a guitar or a violin which will have more sources of the sound of one string stroke) ... then these frequencies are certainly not all close to 20KHz. In other words, be they 1000Hz, then the accuracy again theoretically decreased by a factor of 20. 10cm now have become 2 meters and obviously nothing works.
However ...
What is 100% totally clear to me is that for audible frequency, again be that the guitar or the violin, but also the drum of any kind and name it except for a recorder (flute with sine tones), the extension of those harmonics (which theoretically are infinite if they only don't die out and could be captured in the first place (microphone range etc.) ... is now infinite for the audible range.
Yes. And if not today, then hopefully tomorrow.
What I mean can be heard in anything. Examples :
- Bells are in so many recordings (just think a general percussion arrangement), and they were not there a week ago.
- The scattering sounds which emerge in the lower frequencies, are in close to 100 "ambient" tracks. It just wasn't there before.
- Synthesizer glides (sweeps), denoted by similar squared sound, are in everything of that age (think two boys with their synthesizer). Did I say everything ? - it just was not there before.
- Clarity is infinite to begin with. So a human voice too extends to way more than you thought. It makes it more real. This is spooky, but was not there.
- The roar of low frequency is in about everything. But here too, this is about
higher frequency showing it. Mind you, such an oscillating sound (generated by an LFO) is always about more than one frequency (oscillating against each other).
Of course we know that each frequency we hear, also (close to) pure squares, is about sines (sinuses). Te more square, the more (infinite number of) sines are required to represent the sound.
What I sense happening is that all filtering that could attenuate the higher frequencies, has been banned. Or at least to a degree thought impossible. And that this is way beyond our current thinking, can quite easily be proven by selecting a "more rolling off" filter in all kinds of software (including XXHighEnd). Can we hear that ? a bit. Can we hear what I am talking about today ? with ears completely closed you still can.
The difficulty for me to bring this across, is and remains about it NOT being the
fundamental (!) high frequencies you suddenly hear. So no cymbal suddenly starts to hiss; no S'es are presented white and painful; no flute show a higher pitch (better : frequency) or anything. No.
It is the square-sound possibility which is exploited. I have heard so much of it now, that I am sure it is that. But what does it really mean, if at the same time no cymbal starts to hiss ?
-> That
all what is present for higher frequency in the data and which is laying (or layered) right on top of the lower frequencies which may be square and
usually always are square to some degree, - unless a pipe organ or something -,
can now be dug out of that.
It is not the higher frequency which now is louder (and thus the cymbal does not start to hiss), but it is the possibility to dig out the higher frequency of the lower frequency, which is thought to be the current eating one (while I for a very long time state that the higher frequencies eat more current than the low). Anyway, both the lower and the higher play together (in real life) and the higher are
required to make the sound square as it originally was, and possibly we must look at this as if the one does not eat the current for the other any more. Both now can play nicely along and it is caused by a cable. OK, shield configuration.
It hardly can be about a filter as such, because that would be audible in native higher frequency sounds (like from a cymbal). All right, they changed too (see yesterday's post) but the emphasis to square is a 100 times more apparent.
Without measuring anything, what about the thought that the shielding setup is not allowing to let escape higher frequencies which kind of "ride" on the lower frequencies. I am sure I am talking real BS now, but I seek an explanation. Lower frequencies may "push out" the higher frequencies because of a lack of space. What space ? don't ask me. But I see those higher frequencies kept in now. Kept by a shield.
Btw let's not forget that this is about analogue signal. This, while it is combined with the digital signal of the Lush^2 with very similar manipulation and of which we know for dead-sure that it changes the sound too. And vastly. But not much explainable (or at least not so easy).
Still there ? Good. What I was heading for throughout this post, was the sheer
fact of the way more present higher frequency in actually each fundamental sound, which not only extends to one audible native so-called fundamental because a synthesizer sound (etc. etc.) never is that - only sines have that ... meaning that the one sound already comprises of several mixed fundamentals, those each (!) having their own set of harmonics and THAT again is each comprised of sines ... and that this all leads to hundreds of frequencies from one sound only, them extending to way up out of the audible range. Like the 16.21KHz harmonic we can still perceive but the 32.42 one not any more. And now thus 100s of these for one sound. Make that one instrument if you like.
What I tried to reason is that my "100 frequencies all together" with which this post started out, now suddenly are 100s of them per sound. And careful, because a 10 finger play on a piano is 10 of those sounds. So we easily end up in the 1000(s) now.
... And what I see - or very much like to see from there, is that there *is* sufficient frequency to make sounds locatable in 3D from two speakers. I can just see it coming. That Angie hi-hat spooky example as the first.
And am I changing ^2 configurations at this moment ? no way. Nature is doing its job first (burn-in). So how knows ...
Peter
