• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

30 Years of Digital and the 92% Solution

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Perhaps you ought to place your disclaimer ("The ideas expressed on this website are the opinions of the author and therefore their value and truth should be determined by the readers for themselves.") at the TOP of that page.

Anyway who cares about random "mistakes" in absolute phase when the music is phase-random to begin with? I mean random random is still random!

The very first line of that article is nonsensical. It reads: "Reproduced music is in absolute polarity when its compressions and rarefactions are in sync with the compressions and rarefactions of the original performance." So what? If the mics were moved just marginally certain frequencies would be inverted onto the recording anyway. If you move your head a tiny amount during playback of the most "phase-perfect" recording on the most "phase-perfect" music system (even if you wasted your time calibrating it all with a microphone like some of the manufacturers get you to), the polarity is going to invert into your ears at certain frequencies. MUSIC SOURCES ARE PHASE-RANDOM. Even the sound from a single violin is phase random as it reflects off different surfaces of the instrument. Apart from the tiny blip during changeover, I cannot hear so-called 180 degree "absolute phase" (polarity) reversals and the approach of certain highly regarded digital equalisers and crossovers to achieve alleged phase perfection is nothing but marketer's emphasis of the irrelevant. The only place that phase "correctness" is audible is at the crossover frequencies of your speakers. What you hear is the summed amplitude response. If phase is out around the crossover points, the levels will suffer. Even if linear phase digital crossovers are used, so what? The source on the recording is random. A LW 24dB/octave crossover for example can invert polarity in the middle of the midrange band (compared to the input signal). If you press a 180 degrees flip button on your CD player the crossover output is gonna be inverted somewhere else instead.

Even with headphones (no crossovers) you and I will not perceive a polarity reversal on a played music source. Forget triangle waves.

Why should we read your commercially motivated scientifically deficient nonsense? You are selling "CD dampers" and throwing the word "Platinum" about for goodness sakes!

Oh, I forgot: The ideas expressed in this post are the opinions of the author and therefore their value and truth should be determined by the readers for themselves.
 
Dear Ian,

You've misunderstood what acoustic polarity is about. No matter where you sit the leading edges of the acoustic wave that your ears pickup as sound has the same polarity. 12db x-overs do inverter polarity by 180 degrees (+90 degrees and -90 degree relative to each other) in the at their x-over frequency, therefore to avoid cancelation at of the highpass and lowpass drivers output at an around the x-over frequency the highpass and lowpass side of the x-over speaker's drivers must be connected in opposite electrical polarities which makes the highpass and lowpass drivers 180 degrees out of phase with each other in the center of their passbands. However, 4th order x-overs put their highpass and lowpass drivers 360 degrees out of phase at the x-over frequency so no reversing of the electrical polarity of the highpass and lowpass drivers is required even though compared to lower order x-overs there's more group delay. Therefore, if possible no x-over or first order x-overs (6db, that are +45 degrees and -45 degrees that adds up to zero degrees) is the only transient perfect x-over.

It seems to me that you haven't read the entire monograph. Many much more technically knowledgeable and astute music-loving audiophiles and audio engineers than I am haven't raised the objections that you have.

Respectfully submitted,

George S. Louis
 
Dear Ian,

I suppose it's too much to ask that you do some homework on what absolute polarity is all about and how crossovers function. You don't have to take my word for the truth of what I state, simply ask any audio engineer or electronic engineer for their opinions, and if they can prove me wrong I'll listen corrected. But none so deaf as those who will not hear and in the land of the deaf the one eared man or woman is king or queen.

Respectfully submitted,

George S. Louis
 
It appears you were making a similar argument five years ago when you revived this seven year old thread by making posts #16 and #17:
http://www.diyaudio.com/forums/everything-else/54596-audibility-absolute-phase.html

As said earlier in that thread, the claim of audibility might or might not be true (or more diplomatically, it might be true for some and not for others), but since it's easy enough to check that polarity is maintained through the whole signal path, it's done as a matter of course "just in case."

Maybe ten years ago there was a claim on rec.audio.pro of polarity audibility with a 20Hz sawtooth wave. A .wav file was generated with a couple seconds of "positive" sawtooth and a couple seconds of "negative" sawtooth. I downloaded and listened, and while I could tell the discontinuity at the point where it changed from one to the other, I couldn't hear the difference.
 
Hi George
I have no problem with the idea that absolute phase reversal is audible, but I'm surprised at your finding that it is reversed 80 to 90% if the time. I would have expected that some labels make an effort to get it right these days, and that the rest would be random.

I notice that you came to this conclusion after "determining the polarity of over 3,500 CDs", but I'm curious as to how you did that. Was it purely subjective, as in "this polarity sounds better than that polarity", or were you looking e.g. for telltale known waveform "fingerprints" of certain instruments?

On a related note, there's some interesting examples of asymmetric waveforms here: Asymmetric musical waveforms. Of particular interest is the violin example about 2/3 of the way down the page. I can't help wondering if that was due to an editing accident, or due to the violin actually producing a reversed waveform.
 
Reading through that, I found this interesting comment:
[snip]
The most amazing difference was on a Paco Pena solo guitar piece, where switching polarity actually changed the tonality of the guitar. I could liken it to the difference between strumming up or strumming down. I guess the greatest benefit of getting the absolute phase correct is with instuments with lots of leading edge information and simply mic'd recordings.
That seems to tie up with speculation here about the inverted violin waveform:
An interesting possibility is that in one case the note was ‘up bow’ and the other was ‘down bow’ since we may expect the direction of bowing to be significant.
 
It's notable that instruments have different tone at different angles and distances (which makes microphone placement so important in a good recording). It would be interesting to do a polar plot of waveform (or characteristics such as asymmetry and crest factor) versus angle of the instrument with respect to the microphone.
 
Dear Music-Loving Audiophiles:

At the link Polarity Think Piece: A Speculation Regarding Perception of Detail | UltraBitPlatinum.com to Polarity Think Piece: A Speculation Regarding Perception of Detail, I describe how I test by listening and more about the audible differences between in and out of absolute polarity. I also describe a totally objective test of the audibility of relative polarity with music samples.

Respectfully submitted

George S. Louis
 
The subject was thoroughly covered decades ago by Prof Dick Greiner. Bottom line: minor effect with certain test tones, pretty much inaudible with music. That assumes low distortion speakers; if they have high second harmonic, the differences become larger.

The CD mat and squirt on stuff he's peddling is highly dubious for digital media, and no way I'd ever use it on LPs without a LOT more test data (remember all the potions and treatments that ended up destroying records?). Not to mention the $100+ charge for two ounces of something that probably cost less than the bottle it's packed in.
 
A couple of years back I was hunting down differences between two bits of kit that i had that were only apparent with a couple of pieces of music. Turned out that one of the cd players was inverting phase. The upshot was that I had a friend burn a cd with the two pieces of music that I noticed a difference with and he recorded segments, some inverted some not and we tested them with him at the controls, so a single blind test.

Result I could pick them out 9/10, time after time, but with most other tracks I couldn't discern any difference at all. When I swapped from my ES14's to a pair of Kef referneces 203's I couldn't discern the differences even with the test tracks. (The Es14's have no crossover, just a cap for the tweeters).

Thought it might be an interesting reference point, no fancy foo of any kind was required. ;-)
 
Administrator
Joined 2004
Paid Member
I don't get it. :scratch: Every single CD player I've tested, and most DACs are wired for correct polarity. I check and double check this when building, using or modifying any digital source.

How do I check? Easy - with a sawtooth wave. I generate a sawtooth waveform as seen below and play it from the digital source. It's then easy to see on an oscilloscope if the polarity is inverted or not. For this, the sawtooth is ideal. Much harder with sine, triangle or square waves. :)

So if >90% of all CDs play in reverse phase, I don't blame the digital players. I find it hard to blame the recordings, tho I don't know how to check this. George, is it possible you just like reversed polarity better?
 

Attachments

  • sawtooth.png
    sawtooth.png
    4.9 KB · Views: 138
The subject was thoroughly covered decades ago by Prof Dick Greiner. Bottom line: minor effect with certain test tones, pretty much inaudible with music. That assumes low distortion speakers; if they have high second harmonic, the differences become larger.
that's very interesting SY, I'll search for more data.
there's the Chesky test CD which has an absolute polarity test and I think I can hear the difference as described (distance of trumpet varies). I remember reading once an explanation about this but can't remember the source. it mentioned something about the auditory receptors functioning as half-wave rectifiers.
I've checked my system for correct absolute phase. it doesn't hurt.
but I don't think there's any point obsessing about it as you're basically at the mercy of the recording engineers. are we supposed to check this for every recording by analyzing the waveform? what if multiple instruments were recorded out of phase wrt each other?
I think the most reasonable approach is: wire your system for absolute phase and be happy with it.
 
Last edited:
Or use speakers with low second harmonic and don't worry. :D Problem is, a vast majority of recording have polarities all over the place- what may be "right" for one instrument will not be "right" for another. Prof Greiner's conclusion was that it's worth doing right because it's so trivial, but has little real effect with low distortion transducers and even less with multimiked recordings where the engineer either paid no attention to absolute polarity or used whatever polarity on a given mike got him the "best" overall sound in the mix because of overlap and interference effects with other mikes.

BTW, kudos to George Lewis for using the term "polarity" instead of "phase."

edit: Greiner's paper was "Observations on the Audibility of Acoustic Polarity" JAES Volume 42 Issue 4 pp. 245-253; April 1994
 
SY, could you please point me to anything detailing prof. Greiner's study? looks like I can't find that myself.

I can't find a legal downloadable copy, but if you have a university library anywhere near you, you can get a copy there; JAES is a major journal. He did a shorter, simpler version in Audio magazine about a year later, but I don't have that reference to hand.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.