Thanks for this which fits to some degree although the wideband enthusiasts I have chatted to offered no technical explanation and were referring to something they could perceive. Perhaps they weren't referring to a particular aspect of the sound they could pick out and identify but more having faith that the overall "good" sound they were hearing was due to "coherence". I am still baffled by the attraction of wideband drivers but this adds to the pot of knowledge which may eventually lead to a better understanding. Thanks again.
Having said that it seems to be more of an argument against wideband drivers rather than for them. A wideband driver will go deeper into the mass controlled and stiffness controlled operating regions with their relative time lag and lead compared to drivers operating over a narrower frequency range. A multidriver speaker that is designed to minimise "coherence" as a feature would appear to be preferable? Of course signal processing could be used to address the issue in wideband drivers as well as multidriver speakers but wideband driver enthusiasts rarely seem inclined to use signal processing.
Of course the modest lag/lead in time of different frequencies with a wideband driver is not audible because of how the cochlea works. The modest lag/lead introduced by most reasonable crossovers isn't audible either but could be addressed if it was considered profitable to do so. At the lowest frequencies the relative time/phase shifts in some speakers can become perceivable (though it's importance is debatable) and so it isn't a complete non-issue.
There are those that do
. But I guess I could be more of an exception, my goal was to play with the variables of phase/magnitude to find my own point of view on the matter. Yes I've used multi-ways before, my current choice was driven by (many) more factors than one, as an example, I also wanted to minimise the influence of floor and ceiling reflections without having to use special treatment in those spots.Your words, not mine. As said, I specifically wanted to play with this subject and feel/think different about it. Did you notice most of the tests about audibility of phase have been done with headphones? Try a speaker in a real room but with phase/magnitude in tact at the listening spot.
Stereo pair of speakers as measured at the listening spot.
Why am I against testing this with headphones? Because in real every day life we "listen/perceive" with more than our ears alone! We feel the waves as much as we hear them. Having the phase linearity throughout the entire spectrum makes us "take notice and listen". It heightens our senses. So the absence of room influence to get that direct sound (not only to our ears but our entire body) does have an effect, an effect that I personally like very much!
You can keep your opinion of course, not trying to sway you. But I just took the plunge and tried it for myself, having had all options at hand to choose from and I've made my choice and on that basis simply don't agree with the bit of common knowledge you shared.
I'm more of a believer of what Griesinger has to say about human perception, especially his lecture about proximity:
A lot of my experiments were based on his papers and lectures. I tend to find more common ground in his material than the common wisdom of speaker manufacturers. Even in pro-audio phase linearity in mixing/mastering monitor speakers has made a giant leap forward. It seems to me the rest of this HiFi world is lagging behind.
But my speakers flawless? Neh... just another compromise. A set of compromises that I embraced to use as my personal lab test environment for all kinds of experiments.
Edit: If I would have used multi-ways, I would have used either a coaxial driver at the center or a very closed spaced WMTMW like the Synergy horns from Tom Danley. (and use DSP to play with phase linearity etc
) I'm not stuck on wideband vs multi-ways. I'm addicted to music. Music is emotion and passion, I just like to play with the variables to unlock that kind of emotion. So far I've been successful to achieve that in my tiny world, just aiming to please myself.
Last edited:
Post #104.
On SystemXXX serie Tannoy used analog allpass cells (in serie) to solve the issue seen in link in post 104.
You can see the circuit in first post here:
https://www.diyaudio.com/community/threads/tannoy-system-800a-crossover.336733/#post-5768150
No dsp, all analog.
Tannoy drivers used in the revolution XT 6, and some Kef drivers don't need compensation for Z axis. In big diameter B&C have some claiming same thing.
I would not want to listen in one of those! David Griesinger has taught me more than that! (lol)
It simply wouldn't work with Stereo, maybe with one of the newer surround formats. But the soundtrack of my life is locked inside the Stereo format, hence my quest to get the most out of it. (according to my own preferences, My-Fi)
Because in the absence of any reflection, the interaural crosstalk would mess with tonality as well as imaging and would show its perceptional flaws.
Some level of reflection is needed to hide that we are listening to 2 speakers instead of a sound coming from straight ahead. The tonal differences between phantom sounds and hard panned sounds would be a first queue. Move your head outside the exact sweet spot in an anechoic environment and you'll notice this imbalance change. Even at the reduced levels of early reflections in my room I perceived this as a flaw and found (my own) ways to circumvent this.
Some level of reflection is needed to hide that we are listening to 2 speakers instead of a sound coming from straight ahead. The tonal differences between phantom sounds and hard panned sounds would be a first queue. Move your head outside the exact sweet spot in an anechoic environment and you'll notice this imbalance change. Even at the reduced levels of early reflections in my room I perceived this as a flaw and found (my own) ways to circumvent this.
A simple test would be to listen to pink noise being panned L to R.
An other interesting test listen is the last track of the 5.1 mix of Beatles / Abbey Road. It a mono signal that starts in the Right Surround channel, and is panned all the way around to the Left Surround. Quite interesting.
An other interesting test listen is the last track of the 5.1 mix of Beatles / Abbey Road. It a mono signal that starts in the Right Surround channel, and is panned all the way around to the Left Surround. Quite interesting.
Pano started this long running thread about it, quite a while ago. It was timed quite right as I already had started experimenting with that subject at the time.
There isn't much I did not try. Most of it out of pure curiosity, did many a test much like the ones you describe here. I use my setup with Home Theatre without an actual center (but with surrounds) and that thread resulted in a long series of fun tests. What can I say, I'm curious by nature.
There isn't much I did not try. Most of it out of pure curiosity, did many a test much like the ones you describe here. I use my setup with Home Theatre without an actual center (but with surrounds) and that thread resulted in a long series of fun tests. What can I say, I'm curious by nature.
There are a couple of possible reasons for that, that I could think of. All having to do with how the two differing scenario's interact with the room.
(flawless) Speakers are fun, but speakers + room that work together is always better in my book.
What I would love to witness is having identical twins in the room, standing at stereo speaker positions and singing the same song in key... with eyes closed at the listening spot, would it sound like one person singing?
(flawless) Speakers are fun, but speakers + room that work together is always better in my book.
What I would love to witness is having identical twins in the room, standing at stereo speaker positions and singing the same song in key... with eyes closed at the listening spot, would it sound like one person singing?
These are experiments that measure different quantities. If you wish to measure the audibility threshold of relative phase shifts then that needs to be the only variable. If you use headphones then the HRTF is going to be different to a speaker in an anechoic chamber and so the quantitiy measured will be different although one could perhaps devise ways to map between them. If you use a room with reflections then that is going to be measuring a significantly different quantity. In some circumstances reflections can increase the sensitivity to hearing differences likely because the ear/brain is getting more chances to hear it.
Opinions about what? If by sway me you mean telling me the technical advantages (facts not opinions) of wideband drivers then that is exactly what I want to know. Please sway me.
If by common knowledge you mean published audibility thresholds in scientific journals then that is your choice. Arguing against them effectively with engineers and scientists is likely to be pretty hard going given the way science works but there is certainly plenty of rubbish in scientific journals given the way academia has evolved over the last few decades. Arguing against them with audiophiles is likely to be a lot easier. Particularly if they are opinions.
I've spent a lot of time on a 3-channel setup, trying out various LCR matrix strategies to help solidify the center and retain full envelopment.
My setup uses three identical speakers, including their subwoofers.
A standard "calibration" test for the 3-channel rig before comparing any matrices,
is to compare L & R both running mono, to C alone running same mono.
When I can't tell if I'm listening to center alone, or to L & R playing, I figure I'm good to go with matrix testing.
I've found that speaker placement determines how well i can get there (I don't use any room correction on any of the 3 speakers.)
So I'm in agreement with wesayso about a difference being due to room interactions.
You clearly are not on the same page as I am. Most of the studies into the audibility of phase rotation that I have seen have been conducted with headphones, not speakers. I don't agree with the results of those tests (using headphones) as I don't believe for one minute that they relate 'one-on-one' to listening to speakers.
If you look just that little bit closer you would have seen that part was targeted about your remark, stating: "The modest lag/lead introduced by most reasonable crossovers isn't audible either but could be addressed if it was considered profitable to do so." So, no, that part wasn't about wideband drivers in general.
I only disagree with the part where you stated that phase rotation isn't audible. My stand is that is is audible depending on the circumstances. There's plenty of text out there in this world in scientific papers to support that. So pick your poison, I don't go against science here, in fact I just try to use it for my benefit. As an Engineer I like to get to the bottom of it and not just assume things. Have you seen the video I linked from David Griesinger? Reading your text above it does not seem that way.
Last edited: