Yeah, linear phase and linear frequency response are not exactly the same topic and it’s better to keep them separate. Also, filters can be without phase distortion (using DSP).
All zero crossing points (including the starting point) change upon phase shift. In fact, the difference of the zero crossing times between the original sine wave and the phase shifted one IS the phase shift (if you measure the time in radians). There is, therefore, no tilting or skewing of the waveform to the left or right happening.
I was going to say this too. The whole waveform moves - so all the zero crossing points and all the peaks move. The way Paul drew it, the frequency changed.
@@gerritgovaerts8443 I believe you're judging too hard here. In front of the whiteboard, one can sometimes get lost a bit. :-) On several occasions i found myself writing "less correct than desirable" things on the whiteboard for students, too late realizing that a "more simple explanation" turned into an incorrect one. As the famous (Einstein?) quote says: "Everything should be made as simple as possible, but not simpler." I think, Paul was making a legit attempt to explain phase without using almost any of the relevant math, such as e^(i phi(t)) = cos(phi(t)) + i sin (phi(t)) and introducing phi'(t)=phi(t)+phi_0. Maybe two good old overhead slides with sines, overlaying and shifting them in t-direction would have been better, though.
Linear phase is a property where the phase response is a linear function of frequency. The result is that all frequency components of the input signal are shifted in time (usually delayed) by the same constant amount (the slope of the linear function). To go from phase to time domain requires a first derivative. So linear phase equals a constant delay. Paul - I'm a big fan but your explanation was not among your best.
Man , this is so wrong , that zero point DOES shift ! If that is your level of understanding , I'm afraid it's way below what is expected from even an average acoustic/electric engineer. This is really basic stuff and I mean , it can't get even more basic
Sorry Paul, but phase shift does not 'tilt' the sinus function; that would involve non-linear effects unrelated to non-linear phase. A linear phase response means that all inputs frequencies are delayed the same. That is it! For a flat amplitude response, a linear phase response is a non-minimum phase response for that particular amplitude (the amplitude aim for most loudspeakers), unless the phase is zero which a special case of linear phase and that would be the minimum phase response for a flat amplitude. A Linkwitz Riley cross-over is generally (in total, the LF part and the HF part added together) an allpass (flat amplitude response) filter with NON-linear phase, and thus input frequencies are NOT delayed the same. As a consequence you cannot recreate a square wave for example (should you want to). But a LINEAR phase all-pass filter would not have this time-distortion. It requires DSP to make though (unless you go first-order and unrealistic drivers). So in short; linear phase means all signals are delayed the same, but there is a delay, which could be an issue when watching a video along with the audio.
Great! You've told Paul something he likely has known for decades. Would you mind explaining it to the rest of us, in a simple way that even the least educated of us can understand?
@@HareDeLune Draw a sine wave with some frequency. Your vertical axis is amplitude, horizontal is time. Consider this your input. Now draw another sine wave, same frequency but shifted to the right compared to the input. Consider this your output signal. The output is lagging (although it is 'in front' of) the input, in that it peaks at later times than the input. There is a phase difference between the two signals, and while a phase difference does not necessarily come from a time delay, that is what we are considering here. Read out the time difference between the two signals on the horizontal axis. Now make two new sine waves but this time at a lower frequency, so the the wavelength is longer than before. Shift them so that they have the same TIME delay as the first sine waves. What you will see is that while the delay is the same, the PHASE difference is lower for the lower frequency pair than the higher frequency pair. So for a constant time delay, you will need different phase differences at different frequencies. Plot this phase as a function of frequency, and you will get a linear curve. Linear phase; constant time delay. That means there is no temporal distortion of any signal made up of any combination of sine waves, as any signal is only time-delayed, but retains its shape. Can you hear if a filter response is linear phase or not? I would say yes, as I could hear a difference for a pair of Kii3's (blind test, Bruno Putzey did the switching between the two states), but the track was also suited for pointing out the difference. Much work would suggest you cannot hear any difference for most music. See more in my blog post on the topic: www.acculution.com/single-post/2017/08/23/009-time-alignment-in-loudspeaker
@@Rene_Christensen I really didn't believe you would actually respond with anything constructive. Thank you for 'putting your money where your mouth is', so to speak. I would also like to point out that you are the only person to do so thus far. Reading your reply while half asleep at 2:30am, I was very distracted by your consistant mis-spelling of 'sine wave' as "sin wave", and kept thinking you must be plotting a hellish signal. 😆 Trying to follow your explanation in my head at this hour, you lost me at "...plot this phase as a function of frequency..." Perhaps I could have followed along if you were using a whiteboard in your own video, or perhaps you got a bit more technical at that point. Too sleepy to tell. At any rate, respect to you for coming up with something I could mostly follow, even when half asleep, lol.
@@HareDeLune I was planning on making a video on this topic anyway so stay tuned (acculution channel here on youtube). I will try and get the sin/sine right ;-)
This video is so totally wrong. I can point to a bunch of other videos that make it clear Paul has very superficial engineering knowledge. The “audio fuses” and “elevated speaker cables” howlers come to mind. I know Paul’s videos are designed to sell his equipment but if he has had a hand in design, then I would not touch his products with a ten foot pole. Why he dose not prepare for these videos and run his script past a qualified engineer is a mystery. He is damaging his business.
Ah...You can't fool me. Bob Carver knows the answer to that question....There's no such thing as a "Linear Phase" but there is a "Phase Linear". Phase itself is a relative measurement, so the phrase about phase is meaningless. You can have a constant phase angle AND non-linear amplitude.
Put another way, two sines of different frequencies superimposed will produce a specific waveform when there is no differential phase shift between them. Alter the phase of one of them and that waveshape changes drastically amounting to distortion.
If Paul is claiming 99% true, then this is video is part of the 1%. This is why we need both art and science in classrooms. Being able to draw will make you a better engineer. Still.... gotta love Paul for trying.
Active speakers/monitors can have linear phase if the internal DSP is factory tuned. Many room correction software packages offer phase correction now to, like Dirac. This will provide a digital setup with linear phase.
First, some mistakes: 1:00: "Frequency" should be "period" or "periodic time" 2:06 Wrong: If the zero point would be the same and the rest stretched, we would have a lower frequency. I do agree with Paul that linear phase in relation to a speaker is a little weird to me. Let me explain why and what linear phase actually is. This is all in the context of "filtering" as that is what amplifiers, equalizers and speakers all do. First, let's start with zero-phase filtering. This is filtering where the frequency response (amplitude) is changed but there is no phase-change. This is impossible to do, as it requires a non-causal filter tat can predict the future signal: the impulse response is symmetrical in future and past. In a computer, however, we can do a trade-off, "looking ahead" enough to cover a big-enough part of the "future" impulse response. As computers can also not look into the future, the practical implication is that the whole signal gets delayed by a fixed amount of time for all samples. Delaying the whole signal with a fixed amount of time is linear phase. Translated into a frequency-phase diagram, this means that the phase _shift_ is linear with frequency. For a twice as high frequency to shift the same amount of time, you need to shift it with a twice as high phase, because the time-period of the higher frequency is twice as short in time. Done well, this can create an ideal crossover, but considerable and accurate computing power is necessary. It's actually only 1.5 times the computation needed by a normal IIR filter, but it requires extra buffering and programming talent that audio companies apparently lack. And in live situations, a delay or _latency_ is not always acceptable. For your home speakers I'd say it is perfectly possible to do perfect linear phase filtering and then amping each driver with its own amp. But that's another topic.
I use a PC based system + microphone that is calibrated that can actually measure the phase response of a speaker ... and also show group delay ... these two parameters are the most important ones of all regarding perfect imaging and 'snap' in the overall sound impression
The starting point of the sine wave is not the same when it is shifted. The entire sine wave is shifted along the x-axis (forward or back). You can also shift a sine wave in the y-axis direction although amplitude or volume is not the same as shifting along the x-axis. An example of a sine wave being shifted along the x-axis is given below: mathbitsnotebook.com/Algebra2/TrigGraphs/TGShift.html
The Meyer Sound Blue Horn System is a active full bandwidth monitoring system with a dedicated controller unit. That's not what most audiophiles dream of. The unique thing is, that the systems compemsates the non-linearity (like resonance, physical inertia) of the drivers and enclosures over a very wide frequency range for amplitude and (!) phase. What does that mean? It's obvious that a Tweeter is easier to move than a big woofer and the bigger mass if the woofer has much higher inertia than the litte tweeter. So if your source signal has a very low frequency for woofer and a high one for the tweeter, the high one sound will be reproduces before the low sound. The Meyer System claims it has solved that problem and the speakersystem over it's complete Bandwith reproduces it's input in phase. BTW: It's not a good idea to explain phase shift with a 180 geg shiftet sinewave. Most people don't get it and think a inverted signal is a 180 deg shifted signal. But that's wrong.
This video is wrong. Phase shift zero crossing points are not the same. It's not a distortion in the wave both waves starting at the same time like you're explaining, it's a time difference in the starting point of the resultant wave with respect to the source wave.
Paul, it is clear you are not an engineer. These half-baked white board “explanations” really makes one question the design quality of your products. Stick to your sales hyperbole and leave technical issues to qualified engineers.
Linear phase means linear phase shift and constant time delay. In contrast constant phase shift have linear time delay which is bad for sound because every frequency is not producing at the same time rather at different time that's what happens in LR4 filter where phase shift is constant 360 degree irrespective of frequency so system is gonna be in phase but not in time. For example high frequency gonna come out first and then some delay low frequency is gonna come out. that's not audible most of the time
Phase is about shift in the time domain and it can be frequency dependent. I’m not sure why Paul likes to talk about frequency response. He sort of obfuscates the topic. A filter can be with linear phase also btw. When the phase is perfect you can preserve the original signal curve, e.g. a square wave will remain appearing like a square wave when measured on the output.
@@HareDeLune Paul does a great job in explaining audio topics for the masses in an entertaining way and he occasionally makes some incorrect statements, but that’s alright and is very human, as we all say dumb things from time to time when we improvise. Had Paul been teaching stuff he just studied from a book, we could as well just read that book. Great teachers are those who can make us engaged by acting improvising and entertaining.
"When the phase is perfect you can preserve the original signal curve, e.g. a square wave will remain appearing like a square wave when measured on the output." To me, that sounds more like a job for high slew rate to achieve things like reduction of rise and fall time.
@@QoraxAudio When a system doesn’t have linear phase then different frequencies have different phase and a square wave is a composition of sine waves with different frequencies. This article describes it well: www.edn.com/for-signal-distortion-phase-matters/
The original Ohm F was time and phase aligned ,A company called HHR Exotics makes a much improved version called the TLS-1. Since one driver makes all frequencies there is no need of a crossover and it's phase distortion.
I really like your entertaining short videos, it often looks like you are trying your best to simplify for the benefit of less technically able viewers. Alas this time it was painful to watch and either you know nothing (not very likely) or you are coming unstuck struggling to explain a slightly more complex concept and making a real hash of it. Better luck next time.
Really helped understand this topic. What is the sound impact of a phase issue at the crossover point? 180 is obvious. What about 45 degrees out of phase? Am I correct in understanding that a first order crossover does not have a phase issue?
Proportionally to angle of shift from zero between two sources signal is wrongly dimmed and finally at 180 it gets null. Any crossover create problems - even simple capacity in series or parallel create filter because is connected to speaker coil induction Filtering with serial induction does not hurt so much but it is not popular in market at all . Why I cant say. I made such solution about 35 years ago and it in my perception it had it's advantages but that had to be corrected with EQ which is against market trends .
Yes but the out of band phase shift above and below the crossover point in say the midrange will be shifted in phase when compared with the arrival from the woofer or tweeter. Real linear phase crossovers can really only be achieved with FIR filtering. That what it took for me to get my phase "linear".
Even if that claim were true, it would be meaningless in a stereo two-speaker setup, because there will always be phase variations depending on the physical location of the speakers and the listener
Up to room conditions there is a mass of phase shifts equalizing each other and creating "room acoustic" . But when damage comes out of speaker set it is single, pure and no compensation is available.
I believe his question relates to the baffle and driver's and how there situated to be in phase with each other. Is this way some people like full range drivers such as ZU or Omega?
The diagram in the first half of this vid is not factual. Shifting the phase of a sine wave will just move it in tact in the x axis. To be fair a complex wave form will be un recognisable with individual FIR frequency manipulation.
"This question comes from *x* in *x* = faraway place, which somehow is supposed to lend credibility to the incorrect, meandering and misleading information.
Phase is basically a time differential between different frequency wavelengths. Certain frequencies travel faster through air than others and that's why you see some speaker designs whereby the tweeter is recessed back relative to the woofers. I'm sure crossovers are involved in the timing of frequencies to each other. It's a principle of physics used a lot in electrical engineering and mathematically represented by curves on a graph 'phase shifted' from each other. The degrees is a carryover term from an 'angular' representation of the same phenomenon and the cyclic nature of waves. Basically many different ways to look at the same principle.
I never bought into this concept of speaker design. Yes, different frequencies travel at different rates but making up for that by offsetting the drivers? Assuming that helps, to gain the advantage means the listening position would need to be precisely fixed as well. And who’s to say the flute player wasn’t in front of the bass. To take this concept to its logical conclusion you’d need a road map for every recording and adjust the drivers accordingly if you want to recreate the performance accurately.
Sounds with different frequencies travel with the same speed. The energy carried by sound wave goes up with the frequency. That is one of the reasons you need more power on low frequency reproduction ( another is Fletcher-Munson sensitivity curve)
I was in Naples there were a lot of cats, but Athens, Greece they are freakin' everywhere. Getting down right in the street. You just walk around them and look away.
Paul, please remove this video and do it again. This time your explanation about phase is completely wrong.
Agree! The Zero crossing points do NOT connect as phase shifts.
Agree! I really like Paul’s videos but this was off.
I agree as well. Start, middle cross and end are all off by the same margin. How much will change at different frequencies.
Would also help if he could actually draw
I have a lot of respect for Paul but I was sitting watching that squinty sine wave thinking " that really doesn't sound (or look) right.."
Yeah, linear phase and linear frequency response are not exactly the same topic and it’s better to keep them separate. Also, filters can be without phase distortion (using DSP).
All zero crossing points (including the starting point) change upon phase shift. In fact, the difference of the zero crossing times between the original sine wave and the phase shifted one IS the phase shift (if you measure the time in radians). There is, therefore, no tilting or skewing of the waveform to the left or right happening.
Came to write this. But couldn't have explained it better myself.
How does the 0 point move?
This begs the question , if he does not understand basics like phase , what does he understand ? I am truly shocked !
I was going to say this too. The whole waveform moves - so all the zero crossing points and all the peaks move. The way Paul drew it, the frequency changed.
@@gerritgovaerts8443 I believe you're judging too hard here. In front of the whiteboard, one can sometimes get lost a bit. :-) On several occasions i found myself writing "less correct than desirable" things on the whiteboard for students, too late realizing that a "more simple explanation" turned into an incorrect one. As the famous (Einstein?) quote says: "Everything should be made as simple as possible, but not simpler." I think, Paul was making a legit attempt to explain phase without using almost any of the relevant math, such as e^(i phi(t)) = cos(phi(t)) + i sin (phi(t)) and introducing phi'(t)=phi(t)+phi_0. Maybe two good old overhead slides with sines, overlaying and shifting them in t-direction would have been better, though.
Linear phase is a property where the phase response is a linear function of frequency. The result is that all frequency components of the input signal are shifted in time (usually delayed) by the same constant amount (the slope of the linear function). To go from phase to time domain requires a first derivative. So linear phase equals a constant delay. Paul - I'm a big fan but your explanation was not among your best.
Perfect explanation!!
Man , this is so wrong , that zero point DOES shift ! If that is your level of understanding , I'm afraid it's way below what is expected from even an average acoustic/electric engineer. This is really basic stuff and I mean , it can't get even more basic
So, stop shooting blanks and explain the damn thing the right way, so we can all understand it!!
Sorry Paul, but phase shift does not 'tilt' the sinus function; that would involve non-linear effects unrelated to non-linear phase. A linear phase response means that all inputs frequencies are delayed the same. That is it! For a flat amplitude response, a linear phase response is a non-minimum phase response for that particular amplitude (the amplitude aim for most loudspeakers), unless the phase is zero which a special case of linear phase and that would be the minimum phase response for a flat amplitude. A Linkwitz Riley cross-over is generally (in total, the LF part and the HF part added together) an allpass (flat amplitude response) filter with NON-linear phase, and thus input frequencies are NOT delayed the same. As a consequence you cannot recreate a square wave for example (should you want to). But a LINEAR phase all-pass filter would not have this time-distortion. It requires DSP to make though (unless you go first-order and unrealistic drivers). So in short; linear phase means all signals are delayed the same, but there is a delay, which could be an issue when watching a video along with the audio.
Great!
You've told Paul something he likely has known for decades.
Would you mind explaining it to the rest of us, in a simple way that even the least educated of us can understand?
@@HareDeLune Draw a sine wave with some frequency. Your vertical axis is amplitude, horizontal is time. Consider this your input. Now draw another sine wave, same frequency but shifted to the right compared to the input. Consider this your output signal. The output is lagging (although it is 'in front' of) the input, in that it peaks at later times than the input. There is a phase difference between the two signals, and while a phase difference does not necessarily come from a time delay, that is what we are considering here. Read out the time difference between the two signals on the horizontal axis. Now make two new sine waves but this time at a lower frequency, so the the wavelength is longer than before. Shift them so that they have the same TIME delay as the first sine waves. What you will see is that while the delay is the same, the PHASE difference is lower for the lower frequency pair than the higher frequency pair. So for a constant time delay, you will need different phase differences at different frequencies. Plot this phase as a function of frequency, and you will get a linear curve. Linear phase; constant time delay. That means there is no temporal distortion of any signal made up of any combination of sine waves, as any signal is only time-delayed, but retains its shape. Can you hear if a filter response is linear phase or not? I would say yes, as I could hear a difference for a pair of Kii3's (blind test, Bruno Putzey did the switching between the two states), but the track was also suited for pointing out the difference. Much work would suggest you cannot hear any difference for most music. See more in my blog post on the topic: www.acculution.com/single-post/2017/08/23/009-time-alignment-in-loudspeaker
@@Rene_Christensen
I really didn't believe you would actually respond with anything constructive.
Thank you for 'putting your money where your mouth is', so to speak.
I would also like to point out that you are the only person to do so thus far.
Reading your reply while half asleep at 2:30am, I was very distracted by your consistant mis-spelling of 'sine wave' as "sin wave", and kept thinking you must be plotting a hellish signal. 😆
Trying to follow your explanation in my head at this hour, you lost me at "...plot this phase as a function of frequency..."
Perhaps I could have followed along if you were using a whiteboard in your own video, or perhaps you got a bit more technical at that point.
Too sleepy to tell.
At any rate, respect to you for coming up with something I could mostly follow, even when half asleep, lol.
@@HareDeLune I was planning on making a video on this topic anyway so stay tuned (acculution channel here on youtube). I will try and get the sin/sine right ;-)
I am sure you appreciate straight feedback, so here I go. Your video series are remarkable and great, thank you. This one sucked.
This video is so totally wrong. I can point to a bunch of other videos that make it clear Paul has very superficial engineering knowledge. The “audio fuses” and “elevated speaker cables” howlers come to mind. I know Paul’s videos are designed to sell his equipment but if he has had a hand in design, then I would not touch his products with a ten foot pole. Why he dose not prepare for these videos and run his script past a qualified engineer is a mystery. He is damaging his business.
Ah...You can't fool me. Bob Carver knows the answer to that question....There's no such thing as a "Linear Phase" but there is a "Phase Linear". Phase itself is a relative measurement, so the phrase about phase is meaningless. You can have a constant phase angle AND non-linear amplitude.
Well designed Active speakers have excellent phase response across the crossover.
I liked it when the wave moved back in time
it only happens when you play Huey Lewis & The News!
Thats for satanic music to reverse back the lyrics.
@@DrLoveQc
Sanatic music?
Ya mean, like the Carlos Santana secret chord progression?
@@HareDeLune haha oops
Put another way, two sines of different frequencies superimposed will produce a specific waveform when there is no differential phase shift between them. Alter the phase of one of them and that waveshape changes drastically amounting to distortion.
If Paul is claiming 99% true, then this is video is part of the 1%. This is why we need both art and science in classrooms. Being able to draw will make you a better engineer. Still.... gotta love Paul for trying.
Active speakers/monitors can have linear phase if the internal DSP is factory tuned. Many room correction software packages offer phase correction now to, like Dirac. This will provide a digital setup with linear phase.
Thanks for another great drawing lesson ;-)
Hi from 🇧🇷 . Great lesson, Paul! 👨🏾🎓Phase is always the most complicated when we talk about crossover issues. 🏄🏿♂️
Paul, you should leave these questions for someone with speaker engineering knowledge.
Wrong... What you drew is distortion, not phase. Phase shifts the start and end points in time, not the shape of the wave.
Exactly why so many people posted salty comments on this video 😅
First, some mistakes:
1:00: "Frequency" should be "period" or "periodic time"
2:06 Wrong: If the zero point would be the same and the rest stretched, we would have a lower frequency.
I do agree with Paul that linear phase in relation to a speaker is a little weird to me. Let me explain why and what linear phase actually is. This is all in the context of "filtering" as that is what amplifiers, equalizers and speakers all do.
First, let's start with zero-phase filtering. This is filtering where the frequency response (amplitude) is changed but there is no phase-change. This is impossible to do, as it requires a non-causal filter tat can predict the future signal: the impulse response is symmetrical in future and past. In a computer, however, we can do a trade-off, "looking ahead" enough to cover a big-enough part of the "future" impulse response. As computers can also not look into the future, the practical implication is that the whole signal gets delayed by a fixed amount of time for all samples.
Delaying the whole signal with a fixed amount of time is linear phase. Translated into a frequency-phase diagram, this means that the phase _shift_ is linear with frequency. For a twice as high frequency to shift the same amount of time, you need to shift it with a twice as high phase, because the time-period of the higher frequency is twice as short in time. Done well, this can create an ideal crossover, but considerable and accurate computing power is necessary. It's actually only 1.5 times the computation needed by a normal IIR filter, but it requires extra buffering and programming talent that audio companies apparently lack. And in live situations, a delay or _latency_ is not always acceptable. For your home speakers I'd say it is perfectly possible to do perfect linear phase filtering and then amping each driver with its own amp. But that's another topic.
As an engineer that works in system optimization, this video doesn’t really explain phase or polarity well and why linear phase is important.
Perhaps you could enlighten us with a brutally simple explanation that even the least technically-minded of us could comprehend?
I use a PC based system + microphone that is calibrated that can actually measure the phase response of a speaker ... and also show group delay ... these two parameters are the most important ones of all regarding perfect imaging and 'snap' in the overall sound impression
The starting point of the sine wave is not the same when it is shifted. The entire sine wave is shifted along the x-axis (forward or back). You can also shift a sine wave in the y-axis direction although amplitude or volume is not the same as shifting along the x-axis.
An example of a sine wave being shifted along the x-axis is given below:
mathbitsnotebook.com/Algebra2/TrigGraphs/TGShift.html
Thanks !!! I'm phased out ........
Come on! They're funny to watch 😂
@@robkoppens9966
Eh, he'll grow out of it.
I'm sure it's just a...
...passing phase. 😏
@@HareDeLune Ooow bam! Still dropping heavy truthbombs! 🤣
The Meyer Sound Blue Horn System is a active full bandwidth monitoring system with a dedicated controller unit.
That's not what most audiophiles dream of.
The unique thing is, that the systems compemsates the non-linearity (like resonance, physical inertia) of the drivers and enclosures over a very wide frequency range for amplitude and (!) phase.
What does that mean?
It's obvious that a Tweeter is easier to move than a big woofer and the bigger mass if the woofer has much higher inertia than the litte tweeter.
So if your source signal has a very low frequency for woofer and a high one for the tweeter, the high one sound will be reproduces before the low sound.
The Meyer System claims it has solved that problem and the speakersystem over it's complete Bandwith reproduces it's input in phase.
BTW: It's not a good idea to explain phase shift with a 180 geg shiftet sinewave. Most people don't get it and think a inverted signal is a 180 deg shifted signal. But that's wrong.
Danley Sound Labs has been doing this for decades, starting with the SH50. Meyer is trying to fight back it seems.
This video is wrong. Phase shift zero crossing points are not the same. It's not a distortion in the wave both waves starting at the same time like you're explaining, it's a time difference in the starting point of the resultant wave with respect to the source wave.
Professor Paul is the best!
Just a pity this video is garbage.
Dr. Lewin is better😜
That answers some questions.
No, that questions some of his answers!
@@jjcale2288
Very witty, Wilde!
Paul, it is clear you are not an engineer. These half-baked white board “explanations” really makes one question the design quality of your products. Stick to your sales hyperbole and leave technical issues to qualified engineers.
Amen!
Linear phase means linear phase shift and constant time delay. In contrast constant phase shift have linear time delay which is bad for sound because every frequency is not producing at the same time rather at different time that's what happens in LR4 filter where phase shift is constant 360 degree irrespective of frequency so system is gonna be in phase but not in time. For example high frequency gonna come out first and then some delay low frequency is gonna come out. that's not audible most of the time
Phase is about shift in the time domain and it can be frequency dependent. I’m not sure why Paul likes to talk about frequency response. He sort of obfuscates the topic. A filter can be with linear phase also btw. When the phase is perfect you can preserve the original signal curve, e.g. a square wave will remain appearing like a square wave when measured on the output.
Your negative comment is more understandable than ninety percent of the other negative comments here.
Congratulations?
@@HareDeLune Paul does a great job in explaining audio topics for the masses in an entertaining way and he occasionally makes some incorrect statements, but that’s alright and is very human, as we all say dumb things from time to time when we improvise. Had Paul been teaching stuff he just studied from a book, we could as well just read that book. Great teachers are those who can make us engaged by acting improvising and entertaining.
@@ThinkingBetter
Absolutely agree! : )
"When the phase is perfect you can preserve the original signal curve, e.g. a square wave will remain appearing like a square wave when measured on the output."
To me, that sounds more like a job for high slew rate to achieve things like reduction of rise and fall time.
@@QoraxAudio When a system doesn’t have linear phase then different frequencies have different phase and a square wave is a composition of sine waves with different frequencies. This article describes it well: www.edn.com/for-signal-distortion-phase-matters/
The original Ohm F was time and phase aligned ,A company called HHR Exotics makes a much improved version called the TLS-1. Since one driver makes all frequencies there is no need of a crossover and it's phase distortion.
I really like your entertaining short videos, it often looks like you are trying your best to simplify for the benefit of less technically able viewers. Alas this time it was painful to watch and either you know nothing (not very likely) or you are coming unstuck struggling to explain a slightly more complex concept and making a real hash of it.
Better luck next time.
Question the claims of crossover with no phase shifts? Not if they are using FIR linear phase crossovers.
Really helped understand this topic. What is the sound impact of a phase issue at the crossover point? 180 is obvious. What about 45 degrees out of phase?
Am I correct in understanding that a first order crossover does not have a phase issue?
This was all incorrect. Have a look at my answer further up for an explanation.
Proportionally to angle of shift from zero between two sources signal is wrongly dimmed and finally at 180 it gets null. Any crossover create problems - even simple capacity in series or parallel create filter because is connected to speaker coil induction Filtering with serial induction does not hurt so much but it is not popular in market at all . Why I cant say. I made such solution about 35 years ago and it in my perception it had it's advantages but that had to be corrected with EQ which is against market trends .
Could that be physically corrected? say moving a woofer backward in relation to the midrange?
Yes but the out of band phase shift above and below the crossover point in say the midrange will be shifted in phase when compared with the arrival from the woofer or tweeter. Real linear phase crossovers can really only be achieved with FIR filtering. That what it took for me to get my phase "linear".
Even if that claim were true, it would be meaningless in a stereo two-speaker setup, because there will always be phase variations depending on the physical location of the speakers and the listener
That' a wanted Phase-Shift, aka Delay. That's one base of the ability to hear multidirectional.
Up to room conditions there is a mass of phase shifts equalizing each other and creating "room acoustic" . But when damage comes out of speaker set it is single, pure and no compensation is available.
I still dont get it
at crossover points there can be group delay, aka the phase changes compared to the other frequencies. Ported also has group delay
Have a look at my answer up top
Don't worry , Paul doesn't either.
@@Rene_Christensen
Your answer is not simple and easy to understand.
Try again.
@@HareDeLune says more about you than about him
I believe his question relates to the baffle and driver's and how there situated to be in phase with each other. Is this way some people like full range drivers such as ZU or Omega?
Full range speaker also easy produces phase shifted harmonics but usually full range is not able to totally lower to zero any served frequency.
The diagram in the first half of this vid is not factual. Shifting the phase of a sine wave will just move it in tact in the x axis.
To be fair a complex wave form will be un recognisable with individual FIR frequency manipulation.
Helped give me a major headache
Yes I can see that from your profile picture.
Cats are looking for rats which Nero mistakenly left behind in Rome.
2:30 I wonder how his phase modulator looks like 🤪
Skip to the last 10 seconds of the video!
Paul needs to listen to some Dunlavy SC-IVs or SC-Vs!
"This question comes from *x* in *x* = faraway place, which somehow is supposed to lend credibility to the incorrect, meandering and misleading information.
Yeah...that's true but are you having the effect of synergia?
Nice..what about Phase Linear?
Yeh. I have a 1978 Phase Linear pre amp. What about that?? 🤔🙄.
@@happyraccoon4791 what's that like is that the make
Is that the bob carver one? Ive heard really nice things about those.
@@happyraccoon4791 well it is named that for a reason?
@@baspravisrbin I don't know. But I have one🐱
Phase is basically a time differential between different frequency wavelengths. Certain frequencies travel faster through air than others and that's why you see some speaker designs whereby the tweeter is recessed back relative to the woofers. I'm sure crossovers are involved in the timing of frequencies to each other. It's a principle of physics used a lot in electrical engineering and mathematically represented by curves on a graph 'phase shifted' from each other. The degrees is a carryover term from an 'angular' representation of the same phenomenon and the cyclic nature of waves. Basically many different ways to look at the same principle.
I never bought into this concept of speaker design. Yes, different frequencies travel at different rates but making up for that by offsetting the drivers? Assuming that helps, to gain the advantage means the listening position would need to be precisely fixed as well. And who’s to say the flute player wasn’t in front of the bass. To take this concept to its logical conclusion you’d need a road map for every recording and adjust the drivers accordingly if you want to recreate the performance accurately.
I thought that was called "time alignment"?
Sounds with different frequencies travel with the same speed. The energy carried by sound wave goes up with the frequency. That is one of the reasons you need more power on low frequency reproduction ( another is Fletcher-Munson sensitivity curve)
Hare deLune Right
@@jjcale2288 I may be wrong but i thought that was only in a perfect vacuum that they all travel the same
I was in Naples there were a lot of cats, but Athens, Greece they are freakin' everywhere. Getting down right in the street. You just walk around them and look away.
Cats are useful animals and great pets.
You can bet your bottom dollar, where there's cats there's no mice to spread desease 🙂
@@m.9243 We have 2. We rescued them off the streets, both abandoned kittens. We also have a rescue dog from the shelter.
@@ericnortan9012
WELL DONE! These lucky animals will reward you with their love and devotion for years. Thanks for the reply Eric.
@@m.9243 🤓👍
Wishing all in phase 👍👍👍
Phase non-linearity = analog domain jitter? 🤔
No.
No.
It's a function of the charging and discharging of the capacitors. Capacitors shift phase. A positive example is a run capacitor in an electric motor.
@@elfd18s inductors also shift phase, in the opposite direction as capacitors. Unfortunately the phase shift is frequency dependend.
Wow! This man has to know how wrong he is, which makes this pretty sad. )-:
Great definition / explanation. Thanks!
@@ebarbie5016 omg, yes they are. Now everyone knows. This is embarrassing. Have a blessed day 💋
Yes John Mayer invented the true linear bluehorn speaker, watch his lecture taking the room out of the loudspeaker.
ruclips.net/video/OaeDWsDNIlk/видео.html
They copied it from Danley Sound Labs, who are doing this lineair phase for many years now. The DSL SH50 is where thsi started.
And I just thought he was just a singer and guitar player with a big ego! 😁
Did anyone hear these Blue horns? They seem directed towards studios more than us.
@@ettepet9308 well you don’t get a patent for stealing do you?
This is so wrong it should just be deleted.