Well yah.... that keeps it in phase.
Left the other way, not so fine......at least that’s what my ears tell me!
I would advise to be cautious when it comes to audibility of phase response. It is not a yes or no question, a tweeter mounted in the kitchen whilst listening in the living room is a clearly audible phase error. One at least has to say "phase errors up from 200 Hz with no more than 0,1 ms our better 60 ° referring to the individual frequencies you look at" are audible - or not audible. Opinions depend and whether you listen to a good accurate speaker or not, whether your recording is smeared or not, and whether one has a chance to make a direct A/B comparison or not. And there are some AES articles that deal scientifically with the subject, their resumés mostly are: "not a very big deal, but audible". As mentioned earlier. the time domain measurements look much better than without linearization, see measurements on our page.
The mechanism of how to linearize a speaker both in Phase and Frequency response is described on our website at The HEDD Lineariser(R) - Heinz Electrodynamic Designs . As it needs a well measured impulse response of the speaker to be linearised unfortunately only HEDD users can make an A/B comparison, it would be nice if dome more people could report their experiences insofar. As QUAD founder Peter Walker never stopped to say: the proof is in the listening.
The mechanism of how to linearize a speaker both in Phase and Frequency response is described on our website at The HEDD Lineariser(R) - Heinz Electrodynamic Designs . As it needs a well measured impulse response of the speaker to be linearised unfortunately only HEDD users can make an A/B comparison, it would be nice if dome more people could report their experiences insofar. As QUAD founder Peter Walker never stopped to say: the proof is in the listening.
Humans absolutely can detect a change in absolute phase, but it's signal dependent. Any significantly asymmetrical waveform can easily be detected when phase is inverted so long as the total system group delay is not severe. Most speakers are not severe enough to scramble mid-band group delay to the point that asymmetrical signals polarity cannot be detected.
There are two kinds of "phase" being discussed here though. One is actually "polarity", where a signal is inverted or non-inverted. The other phase issue is actually group delay, a change in phase vs frequency. Both can be easily audible given certain conditions. Polarity inversion I just described above. Group delay becomes audible when harmonics of a signal fall in the audible frequency range up from low to upper-mid-band, and group delay is significant in that frequency range. High frequency group delay becomes progressively more ambiguous.
Something like the HEDD Linearizer should be audible in many situations, especially if it is used in a room with acoustic properties that do not overwhelm the results. There are several other systems using FIR filters to accomplish similar effects.
However, absolute polarity is a bit of a red herring, as there is no way to determine "correct" without a reference very early in the recording process. Theoretically you want positive air pressure to produce positive speaker diaphragm movement, but there are numerous steps in the recording and production chain that can confuse absolute polarity provenance. Often listening and comparing the results of phase inversion will result in a preference, but that preference doesn't necessarily correlate with true absolute polarity.
There are two kinds of "phase" being discussed here though. One is actually "polarity", where a signal is inverted or non-inverted. The other phase issue is actually group delay, a change in phase vs frequency. Both can be easily audible given certain conditions. Polarity inversion I just described above. Group delay becomes audible when harmonics of a signal fall in the audible frequency range up from low to upper-mid-band, and group delay is significant in that frequency range. High frequency group delay becomes progressively more ambiguous.
Something like the HEDD Linearizer should be audible in many situations, especially if it is used in a room with acoustic properties that do not overwhelm the results. There are several other systems using FIR filters to accomplish similar effects.
However, absolute polarity is a bit of a red herring, as there is no way to determine "correct" without a reference very early in the recording process. Theoretically you want positive air pressure to produce positive speaker diaphragm movement, but there are numerous steps in the recording and production chain that can confuse absolute polarity provenance. Often listening and comparing the results of phase inversion will result in a preference, but that preference doesn't necessarily correlate with true absolute polarity.
Is that connected with the way the ear perceives higher frequencies, ie the process "phase locking"?
Right. With a trick lab signal, people can detect a change... on a good day. Even tiny changes of all kinds, I suppose.
But with music, maybe not a chance.
And "detect" might mean a slight change in sound; but that isn't the same as walking into a room and saying, "Aha, somebody has reversed the polarity, I can tell it is wrong" even with trick lab signals. And 'detect" also has no bearing on "prefer".
Again, as jaddie correctly points out, there's a phantasy out there that there exists some "original" or "true" sound that we simply lift off our old vinyl disk and try to play it pure back.
B.
Thanks for that jaddie,
That would explain a couple tracks I’ve got in my playlist that sound ‘off’ just as if they were recorded out of phase....but I wasn’t sure it was even possible.
Ben as someone who’s been engrossed in this phase issue thing for going on a yr now.....I agree that someone is not going to come into a room without a baseline reference and say ‘wow that’s out of phase’ (well maybe some could, but not me) but what I can tell you is that after 100’s of hours testing on the same system you kindly become intimate enough to say whether it sounds correct or not......especially if your playing around with xo that’s known to affect phase.
That would explain a couple tracks I’ve got in my playlist that sound ‘off’ just as if they were recorded out of phase....but I wasn’t sure it was even possible.
Ben as someone who’s been engrossed in this phase issue thing for going on a yr now.....I agree that someone is not going to come into a room without a baseline reference and say ‘wow that’s out of phase’ (well maybe some could, but not me) but what I can tell you is that after 100’s of hours testing on the same system you kindly become intimate enough to say whether it sounds correct or not......especially if your playing around with xo that’s known to affect phase.
It may be that one channel has been inverted, try flipping one of the speaker leads
There is a long history of debunking "golden ear" claims using silent A-B switching, esp double blind. Prof. Stan Lipshitz (an AES award winner) is famous for that - esp when he shamed a certain high-end manufacturer about telling digital from non-digital. To do A-B testing, the two signals must match within a fraction of a db in freq response. If not, anything you hear just might be due to FR.
Since that is nearly impossible to manage, my conclusion is that there are nearly no legitimate observations or tests ever unless somebody compulsive like Stan does it. That sure applies to phase claims even from (or esp from) physicists.
Many if Toole's tests are for preference or otherwise not needing to be blind. That's not the same as "I can hear it" claims that have to be tested blind.
Granted, all of us must make decisions about sonic qualities and can't do it blind. The knack is to structure our test to be trustworthy. I have recordings that I've played hundreds of times (maybe thousands) going back 60 years like my Pop Science test record with a sweep with cricket chirps and others.
B.
Last edited:
It doesn't take a "trick lab signal" at all. Male human voices make it quite obvious.
There are asymmetrical musical wave forms, just not as many. Some horns, double reeds, percussion, etc. Yeah, much more than a chance, but less obvious that the human voice.
Again, it doesn't take "trick lab signals". Many natural and acoustic instrumental sounds are asymmetrical. Yes, "detect" means to discriminate a change. I never said that anyone could tell if polarity was reversed, and I thought I made it clear that preference is ambiguous.
I think I did say that.
Yup, big problems there.
I don't think I'd say it that way. The ear is less sensitive to harmonic content as the fundamental goes up. 3rd harmonic of 7kHz is at the edge of the ability to detect it.
There's no such thing as "phase locking" in human hearing, at least, not as I use the term. "Phase locking" is where a phase comparator is used to control a signal that can be adjusted to lock to another in phase, actually, any arbitrary phase relationship. I.E. a phase locked loop.
Not sure what that refers to with regard to hearing.
"Recording out of phase" is something else entirely. We speak of signals "out of phase" as one channel relates to another, or as one microphone signal relates to another. In any recording there is generally always some out of phase information, and that's not a problem. When sounds panned center are out of phase, that's a big problem. Out of phase bass...nope, not a good
Terminology: "out of phase" relates to the polarity relationship of two signals, typically channels. Almost anyone can be trained to instantly detect out of phase speakers. I hear it walking around retail stores all the time. Polarity inversion of both channels, that's something else entirely, and not clearly detectable, and preference is usually random.
I think already answered. The HF components are not too high as to be affected by usual hearing loss.
2000 Hz was my standard for a long time (as minimal because it is difficult to go higher). Now it has changed to 5000 Hz (it related with my decision to leave high order crossovers behind). It is related with drum instrument, never observed with the guitar.
It is called absolute phase. One option is (usually) better sense of imaging and more relaxed sound, the other option is better bass definition. Given clear differences (i.e. listener can describe the difference well) I think most will prefer the later. Well, who doesn't like bass
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.