Site Links
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Categories
This utterly Fascinating to me . . .
P. Bateman at work again and he makes a lot of sense concerning FR and Phase importance for those using DSP and likely passives as well - it just makes so much sense.
"OK, so at this point in the discussion, I've proposed that:
1) our perception of high frequency location is determined by amplitude
2) our perception of low frequency location is determined by phase
3) in the midrange, it's both
By the way, if anyone's curious, there's an actual anatomical reason for this. Basically, the median diameter of a human head is 14.4cm. 2327Hz is 14.4cm long. Because all frequencies above 2327Hz are shorter than the diameter of our heads, we can't perceive their phase easily. Because all frequencies below 2327Hz are longer than the diameter of our heads, we can perceive their phase easily.
This means that at high frequencies, amplitude differences are everything. At 5 khz, a 3dB difference in the left and right speaker will be incredibly apparent.
This means that at low frequencies, phase differences are everything. At 500 hz, a 90 degree difference (34cm / 6.7") in the left and right speaker will be somewhat apparent.
This means that at high frequencies, phase differences are (mostly) unimportant. At 5 khz, a 90 degree difference in the left and right speaker will be unnoticeable; that amounts to moving the speaker 1.7 cm. Nobody will notice that.
This means that at low frequencies, amplitude differences are unimportant. At 500 hz, a 3dB difference in the left and right speaker will be almost unapparent. "
More here if interested --> https://www.diyaudio.com/community/threads/cbt-with-crosstalk-cancellation.405434/
Comments
Interesting.
Yet this seems to be (limited to) relative to phase and amplitude differences between (left and right) speakers? As phase differences (and amplitude) between drivers of a speaker (and woofers and their PRs or ports) are problematic - positive and negative interference across the shared frequencies.
Genius insight of P Bateman isn’t needed here, only Wikipedia article.
https://en.m.wikipedia.org/wiki/Sound_localization
Read under Human Auditory System - Duplex theory.
But we are so privileged to have his intellect which is of the highest caliber pontificate on all aspects of audio.....
This hobby attracts those... outside the bell curve.
At the very least, I read far enough into his threads to see if I accept his premise. I remember he had some fascinating posts on JBL waveguides. And also some kooky ones. He's chasing the same dragon most of us are, and if he's going to invest his resources to trying the next weird thing... well, I'm glad one of us is.
In this case...I'm not sure I accept his premise. I think the results are going to be VERY program-material dependent. And very room dependent.
Very well articulated, Dirk - Had he not posted what he did I would not have received the feedback and greater understanding [dcibel's link] because I didn't know what to look for nor know _what _I was looking for.
But relocating the frequency divided drivers around the listening position just reminded me of one of my very 1st posts here about doing just that with my 3 way stereo speakers, seeking a similar effect back in my late 30's.
What is old becomes new again in the light of physics expressed as mathematics translated into readily understood concepts for those within the bell-curve [per Dirk].
This conclusion appears to contradict my experiences with my Kowaxial speakers a few years ago. See link below which shows the judges and my comments on soundstage imaging during the MWAF 2018 competition. I think the judge that scored me "2" for soundstage/ imaging is entirely accurate. After the competition, I set up my speakers at home and was able to duplicate his findings. The soundstage completely collapsed in the position where this particular judge was sitting. My initial thinking was that the very narrow polar glitch at 8kHz, which occurred at exactly 15.5 degrees horizontal, creating a relative phase difference between the two channels. But what you seem to be saying is that the soundstage collapse must have been due to the SPL difference alone, and had nothing to due with relative phase difference? See attached comparison chart.
Link to judges comments:
https://techtalk.parts-express.com/forum/tech-talk-forum/1382742-2018-mwaf-score-charts-and-comments/page2
Measured glitch at 8kHz (15.5 degrees horizontal verses 12 degrees horizontal):
I don't doubt your measurements but I do question your interpretation of the data in light of P.Batemans thread where he says that discernable directivity at higher frequencies is due more to SPL than phase. Correct me if I wrong and why but your 15.5 deg sweep uncovered a dip in SPL and not a noticeable phase shift.
Do you still have the Horizontal polar plots to share?
Looks to me that the judge just wasn't sitting in the correct position to be able to adequately/authoritatively review/"judge" your speaker's imaging.
We cant design speakers that make everyone happy in every location and universal room.
You are probably correct. I simply jumped to the conclusion that this was due to relative phase differences and not SPL differences. As I recall, when the soundstage collapsed, it was replaced by the type of strange, phasy type sound that you get when you accidently connect your speakers up 180 degrees out of phase. The phase shift through the glitch region at 8kHz would be substantial, but is not shown in this particular graph. I'll see if I can find a graph that also shows the phase shift as well.
Phase and amplitude go hand in hand when combining audio from multiple sources, either between drivers or between L and R channels. That's why the Wikipedia article uses the terms phase for low frequency, and the term group delay for high frequency. I would say Mr. Bateman is quite incorrect in point #1 in the OP . Forget about 90 deg phase difference at 5kHz, do you really know it to be 90 degrees difference, or is it 450 or 810 or 1170 etc. as well as realizing the complex constructive and destructive nature of wide band audio, natural sound is not a singular sine wave. It's no wonder why construction backup beepers are slowly migrating from single tone obnoxious noises to white noise generators.
No argument here from me because we [I] build speakers for a specific listening environment most of the time and that is all that matters - we can control the performance of the drivers within a reasonable listening position [LP].
https://falstad.com/ripple/
Looking at that graph, my 1 ½ cents (guestimate) would be regardless of left vs right phase issues one would hear (judges / listeners) a significant enough dip in one side? Amplitude difference if not phase as well over this frequency range.
I give our hearing ability a lot more than what may be measured.
Yup, still have them (0-90). But they are from the first version of the crossover, which was later modified to reduce the amount of baffle step compensation applied by about 2-3dB. This version sounded very dark and bass heavy.
Here is another graph, showing on-axis verses 15H, with phase curves. Note how the 15H phase curve shifts dramatically at roughly 8kHz. Roughly a 140 degree phase shift!
I know we can't design speakers that make everyone happy, but that is what we are being asked to do when we enter a competition, such as MWAF. We have to design for a listening window, not a listening position. And we have to properly guesstimate the room size and treatments (RT60 maybe 500ms or so) to get the tonal balance right. Sometimes they throw us a curve, with a smaller room and lots of damping material on the walls (RT60 of about 200). It is not an easy task.
2nd graph - phase; are the colors respective to their axis measurement?
On axis = RED and 15* off axis is BLUE in both sweeps?
Thanks for finding and posting these as it aids in the discussion/understanding.
Yup.
What is the Mid to Tweet XO point, please?
I had a pair of JBL 6.5" coaxials here in my little drum room that had a serious void/dip at 10Khz but these 10" Eminence CX-10 don't have that issue at all even in the same position.
Maybe some of this is related to the cone/surround?
No doubt phase is involved or the shift is a symptom - I just don't know . . .
Another thing to note. The 8kHz dip on the blue 15 degree off axis curve is not quite as deep as it was on the first comparison graph that I posted above. Those were two different measurement sets done several months apart.
The first graph compared 12 degrees verses 15.5 degrees off axis. The 2nd graph compares 0 degrees verses 15.0 degrees off axis. As I recall, that small 1/2 degree shift made quite a difference in the sharpness of the null.
I think a more controlled environment needs to be involved in order to figure it out.
The bigger picture here is determining how much phase effects localization in the general population's hearing above 2400 Hz for the positioning of tweeters.
4kHz @ 12dB/octave electrical. It's the Dayton CX150-8; the tweeter dome has a small 1.5" diameter waveguide that does not transition smoothly into the surface of the woofer cone. I suspect that that is the source of the glitch. I was going to experiment by taping a small, felt donut into the discontinuity and re-measuring, but I never got around to it.
I think we can set aside your coaxials and mine for that matter concerning the discussion revolving around P.Batemans thread posted elsewhere at this point.
The salient points I am interested-in are those in the OP:
1) our perception of high frequency location is determined by amplitude
2) our perception of low frequency location is determined by phase
3) in the midrange, it's both
Our minds and youthful hearing are miraculous but at what point in frequency do we lose complete auditory ability to detect phase changes?
Once we know this we can design listening spaces rather than just deceased lab simian containers and truly put our efforts where they matter - in our homes where we utilize the technology most frequently.
Thought:
A bird chirp - can you tell how far away and direction it is more accurately than a dog barking or a diesel truck idling?
Hmmmm . . .
Yes it is fascinating
Check this out Steve:
I first heard about this some 25 years ago:
https://en.m.wikipedia.org/wiki/Crystal_River_Engineering
https://en.m.wikipedia.org/wiki/Head-related_transfer_function
The Acoustitron !!! 🤯 Lol Thats awesomeness