Yes, I believe that would be correct. While I'm not sure about the 1K number, I'm sure it's not hard to get good enough for all humans using say, 35-50 ohm cables, or three to four parallel runs per speaker. That even allows going up a gauge as you do have more copper.
I'd try to keep ITD errors down up to 10khz give or take. I always like to stay at least an order of magnitude better than spec, three if possible.
It's fun giving a presentation where the questioners are concerned about you even making target...and then your next slide shows measured performance three orders of magnitude better. Classic deer in the headlights..
Jn
Recent research 2019 "Smallest perceivable interaural time differences." for ITD thresholds establishes higher than 2uS - about 18us for untrained listeners
Last edited:
Oh yes, Polk audio cable that would fry 50% of all audio power amps. '-) That was the FIRST speaker cable that really sounded different from zip cord that I ever tried. Later, the manufacturer added a termination (RC I think) that made it much more easy do drive.
In fact, it might be a good idea to resistively load a cable at the loudspeaker end with either the series RC or just the R, as suggested by Dan. I think of it as two regions of operation, the audio range up to 100K, and the RF region above 100K. After all most speaker cables are picking up RF from the air as well. It makes sense to DAMP the RF resonances with a resistor of reasonable value, like 75-120 ohms, depending on the cable.
In fact, it might be a good idea to resistively load a cable at the loudspeaker end with either the series RC or just the R, as suggested by Dan. I think of it as two regions of operation, the audio range up to 100K, and the RF region above 100K. After all most speaker cables are picking up RF from the air as well. It makes sense to DAMP the RF resonances with a resistor of reasonable value, like 75-120 ohms, depending on the cable.
Watch out for the trucks..
Please learn the salient points of the discussion.
Perhaps read the links posted, what a novel idea.
Jn
Last edited:
If you recall, zobels and resistors were discussed the last time this topic engaged, somewhere in the JC preamp part two.
Jn
Thanks for the link, interesting read.
What I find strange is they use all kinds of weird stimulus, band limited Gaussian, pure tones, cosines summed tones. Cool and all, but what about sounds we actually listen to in real life? Female vocals, drums, horns...
I wonder if things we as humans have evolved to hear are easier to localize?
Has anyone done a three speaker setup just to compare an actual source position to a stereo pair that is delayed inter channel?
Jn
Pathological conditions such as outright oscillation at certain levels should be in another discussion.
Watch out for the trucks..
Please learn the salient points of the discussion.
Perhaps read the links posted, what a novel idea.
In fairness, the discussion has been something of a moving target, with interested and careful readers posing guesses about the actual subject being discussed. I'm a bleedin ficky, but even brainiacs are finding it heavy humping.
Much thanks, as always,
Chris
Yes that is a common problem found in a lot of psychoacoustic research - it's similar to the problems found in audio measurements - the signals don't represent real world conditions. At least there is pressure within psychoacoustic research that recognises this shortcoming.
Something I forgot to take into account is that we also perceive the sound envelopes of HF sounds & these envelopes will be perceived as LF so ITD applies to the envelopes of HF sound.
This recent paper (2015) goes into this aspect in a bit more depth "Sensitivity to Envelope Interaural Time Differences at High Modulation Rates."
I'm not sure you can say that it has been a "moving target" rather that people refuse/refused to discuss what JN has clearly laid out preferring instead to argue about TL effects at audio frequencies
Rayma also laid it out fairly logically here
The real question is just how audible this might be?
To do listening tests, I think you'd need a buffer amplifier with (high power) adjustable
purely resistive input impedance, between the load end of the speaker wires and the speaker.
There are too many variables introduced by the speaker to allow that to enter the fray.
Then experiment for audible differences with matched vs unmatched loading on the wires.
Last edited:
My difficulty is that nothing was clearly laid out; it began as a guessing game posed to PMA, with no clues at all about the direction to be taken.
Most folks here have a reasonable familiarity with transmission lines and reflections, etc. But that's only very periferally the point of the discussion, despite all early appearances.
To judge someone harshly for being unable to guess which number from one to ten you're thinking of, seems unfair to me. And, yes, I'm holding JN to a much higher standard than I would ordinary yahoos.
All good fortune,
Chris
Last edited:
Speaker guys are mostly subjectivists. With speakers, where so many compromises are made, you cannot brag with how good it measures when everyone can hear clearly how bad it sounds. With amps, you can see so many that admit they cannot hear but are pretending that well measured amp sounds so great because they all do sound similar anyway. I think the objectivist of speaker world is Earl Geddes.
What symmetry?? When a music is recorded in a room, the record should contains the information of the room from the reflections that get into the microphone. In speaker reproduction, assuming that the 'image' information is already in the record (or engineered to be), best imaging is when there is minimum reflections in the room. The closest reflective material is often the floor between speakers and listener, which is rarely addressed
. In amplifier reproduction, the details (which contains the room information) should be intact, without phase distortion.Balloon popping in a non-reverberation room has no imaging at all
YouTube
Last edited:
No, it began with ScottW mentioning about JN's ITD "hobby horse" to Simon700, & asked JN to spell it out again which he did, AFAIR. PMA just doggedly confused matters with off-topic arguments.
So then I am an extraordinary yahoo
My postings in this topics carry several levels.
1. The t-line analysis demonstrates that a 5 or 10 meter cable will slow down the step response of the system. (Note that equivalency between t-line and LCR analysis has been confirmed as identical)
2. Load impedance variation above and below the nominal cable impedance impacts the settling time up into the regime of localization capability of humans.
3 Magnetic to mechanical conversion devices carry with them an inherent non linearity in their transfer function, allowing that transfer function to be modulated by power signals through position, velocity, and acceleration. It is that transfer function modulation that will alter the dynamic impedance the speaker presents to the amp through the cables.
4. As a stereo program relies on each channel having time and amplitude independence based on program, the possibility that the transfer function of each speaker will modulate independently is there. In fact, that independent modulation is a requirement, as that is the only way to introduce ITD in this way.
5. Does the aforementioned set of conditions have the level of effect required to alter soundstage as discerned by humans.
As an engineer, I break this problem into smaller more manageable sub problems. And all of what I just mentioned I have discussed at length in this recent discussion. I in fact elaborated more at length on the magnetic stuff, as I know that is a weak spot for all.
Pavel is not guessing. He emphatically stated that the primary analysis I provided was flawed. He did so using his learnings (which I also learned) in regard to sub wavelength t line analysis which was that it was taught as inappropriate. That teaching however, is a rule of thumb, not a rule of physics. At that time, T-line theory was taught for engineers to understand waveguides, t-lines, stub tuning, quarter wave transformers, neat stuff like that. Never was sub wavelength t-lines considered.
Pavel asked for proof that the t-line stuff exists. It has been provided in the writings, analysis, and documented tests of Bateman that Zung was so kind to provide us. I hoped appendix two would be available as it provides even more on the design and building of the reflection bridge, but sufficient evidence was provided nonetheless. I do not recall who sent me it, but I do not know if I can provide it on forum.
Pavel did take the time to do tests, I thanked him for that effort. However, he did not consider the problem as I laid out logically.
Bonsai simply ignores what has been provided and continues to yell at the tractor trailers.
And that is only the first section in the analysis, the part I would consider bog standard. As I mentioned to Demian, I've no clear engineered approach yet for actually measuring the non linear time variant full frequency span impedance variations on a fully hard music driven device..that is where the real brainstorming is needed. Sine waves need not apply.
Your trying to hold me to a higher standard is much appreciated, you speak well, mean well, and thank you for eloquently stating your concerns. I take them very seriously.
You must know how exhausting this post has been to me...extraordinary, equivalency, aforementioned, inappropriate.. More multi syllabic words in this one post than I have used this entire calendar year
Cheers, John
Last edited:
Oh I almost forgot that comment..
Scott? Hobby horse? Really?
You Will pay for your insolence.
Jn
Between friends really? All in fun, at least I didn't accuse you of being a crypto-nazi. Honestly I sat here today sort of regretting I said that but this thread is so all over the place one gets carried away (even to the point of thinking copper is a superconductor).
Um, that was tongue in cheek, you know me. I learned a long time ago that life is too short, it should be about fun. We still need a tongue in cheek (TIC) emoticon. I'm seriously thinking of using that dang Santa waving one as my TIC ..
Yah, I saw that comment...what was that all about? Nah, don't wanna know.
Wait, you mean copper isn't a super? I didn't see that memo. I miss all the good ones.
Jn
They're getting closer to room temp these days, Bybee et al will leave out the giga Pascals of pressure needed, details, details.
- Status
- Not open for further replies.