Thinking is the hardest work there is, which is why so few people engage in it. I do find that top-end response characteristics make a difference to me.
Lead the way Paul! I have said for decades that one of the problems with the audio industry is being stuck in the 50's and 60's test and measurement. Back then, which I was, even generating a good clean sinewave was expensive. Much less variable frequency with accurate display. We did not have digital readouts. I challenge the claim that we had equipment that was flat enough in the '50's to accurately reproduce in order to measure the 20Hz-20Khz way back when it was first claimed. Much less time domain anomalies. Dr Matti Otala's TIM in the '70's was the first I saw that moved past simple steady state sinewave. But our auditory system just like music is not based on frequency. It is based on time domain, variations, transients... Phase shifts and delays are more audible than slight variations in frequency response. But not measured with our current T&M. We can get a better measurement of phase response accuracy by using squarewaves. I was taught in Waveform Analysis 101 that the first ten harmonics of the fundamental are as critical as the fundamental. Thus a ten times bandwidth is needed to reproduce a fundamental accurately. So to accurately reproduce to 20Khz, you need a bandwidth out to 200Khz. But yes a good phase based measurement is needed! One paper I read shows that the "phase" difference between ears, the ability of our auditory system to resolve phase, transient reception between ears is as short as 10-20 microseconds! The human auditory system can resolve detail as high as >100Khz. Just not a stead sinewave.
Well said!!! We "hear" much more than frequency. It's evolutionary. The hair on our arms, sounds from behind us, time, reverb, phasing, etc... We pick up a lot of info, a lot of ways when "listening". 👍👍
An extended high frequency response also can affect apparent transient ease. I was playing a solo acoustic guitar recording and I decided to connect a tweeter with more high frequency extension. The sound took on such an effortless quality on the transients, as he plucked the strings. The system just sounded like it was right on top of things with the more extended tweeter. Spectral power amps have been known to go up to 1 million hertz, according to their specs. Some in the forums said that you had to use a special constructed speaker cable with them, for them to honor the warranty on the amps.
Just use 18" dia. copper plumbing coax with nitrogen for the dielectric. Runs out to several hundreds of Mhz, handling power up into megawatts. Guaranteed no phase shift.🤣
If the transient response is even a tiny bit slow, the whole thing sounds like a reproduction; and a mechanical one at that. That's one of the reasons I use great cables, and on phono, a moving coil cartridge or some equally exotic type, in place of moving magnet. That slight delay in hitting the leading edge of notes, gives it away & tells the ear that it's only a reproduction.
@@ianbigsand7 Is that indicative of helium for moisture abatement as opposed to atmosphere for the dielectric? I know I said nitrogen because that's what they use to displace moisture in some other disciplines. I know you can have actual tonnage of that stuff in large combiner rooms feeding large community towers.
Paul is right... phasing is important but not as an independent function. What needs to be appreciated is that you are not feeding your amplifier 12khz, and 22khz and 30hz all at the same time. You are actually sending it a single voltage that changes over time. On an oscilloscope this would appear as a 30hz waveform with the 12 and 22 khz components riding on top of it... sort of a fuzzy 30hz waveform. That waveform is also how you are asking your speaker cones to move. All the rest of it ... the FFT, spectrum, frequency etc. are all mathematical abstractions taken from that single fuzzy waveform. While they have value in examining how things are behaving, they are not exactly what you are asking your system to do. The answers lie in the "Transfer Function"... All electronics introduce a delay, it takes a finite amount of time for the outputs to react to the inputs. This also limits how well they can respond to rapid changes, ideally they should be able to reproduce a change fully before the next change comes along. So, what happens is that we have our fuzzy 30hz waveform. The amplifier might be fast enough to perfectly track the basic 30hz part but the smaller bits riding on the waveform may fall victim to "propogation delay" and "slew rate" as above and not be fully reproduced. Whether this is crucial at frequencies we cannot hear is an ongoing debate. The fact is the inability to accurately reproduce the smaller more frequent "fuzzies" on our waveform does, in fact, change the waveform itself in very subtle ways. But as always, the question is whether that is audible or not...
Paul, when he said 44 kHz, that's such an odd number to throw out there that I think he was referring to the CD sample rate (44.1 kHz), or confused it with the CD frequency response (to 22 kHz). He was apparently talking about his components playing to higher frequencies than his speakers. That's what it sounded like to me anyway. I think maybe you misunderstood what he was asking because his question was pretty confused.
It's interesting that you discuss the audibility of phase shift in the highest octave while not discussing the significant phase shift in coil and capacitor based crossovers. An LR4 crossover, for example, walks through a full 360 degrees of phase shift in the source material in the octaves above and below the crossover frequency. Since a square wave is a combination of the fundamental and its odd harmonics correct in both amplitude AND phase, it is impossible for a speaker with such a crossover to accurately reproduce a square wave. The fundamental and its harmonics simply can't all align. An accurate digital FIR phase free filter, perhaps including time alignment and some phase error compensation for the drivers, is essentially the only way (other than using strictly 6 db/octave crossovers) to enable a speaker to accurately reproduce a square wave. This reflects on its ability to keep fundamentals and their harmonics time and phase aligned -- which as you indicated -- can affect the sound of the music.
Firstly, response concatenates, a -3dB at 20KHz pre-amp and a -3dB at 20KHz power amp’s -3dB point is 14.4KHz. (This is another longer subject BTW) The question is around that phase begins to shift a decade down, so for a 20KHz filter, phase starts to shift at 2KHz. Pauls explanation is grounded in good Eng. Lets play with a 7.5KHz tone (nearly double highest piano note) and its 15Khz first harmonic. Phase magnitude matters, so what is say, 10 degrees excess phase shift at 15Khz due to BW cut-off going to look like. Well the wavelength of the 15Khz harmonic is 2.28cm and a delay due to 10 Degrees phase shift is 2.28/360, * 10 = 0.63mm (24thou) Keeping in mm, sound is 343,260,000mm/ sec, so a 0.63mm delay represents a = 0.000000018s (1.8nS) delay in the 15KHz harmonic of 7.5KHz. You have to remember your ears don’t work at ~2nS = 500MHz. This said, my main amp has a -3dB limit of 200KHz, and the great Rupert Neve had 200KHz in his elite mixing desks. You always lose when you win, so extended BW brings issues of intermodulation distortion possibly increasing etc.
There are similar rules-of-thumb with wide-band voltmeters and oscilloscopes. Whatever the high-frequency -3dB cut-off is, if you want accurate amplitude measurements, stay a decade below the -3dB cut-off. If you want accurate phase measurements, stay TWO decades below the -3dB cut-off.
Given a -3dB spec of 50khz, given a single dominant pole roll off, the phase is -45 degrees. One octave below 50 kHz, it is -26degrees. At one decade below 50 kHz, the phase is -5.6 degrees.
I asked the same question a while back so I made sure that my dual subs weren't. What a fine improvement it was in all aspects. Since I experiment a lot, less so now since I have gotten a system sound so close to what I want, I have come up with a personal concept on speaker design and setup. I call it homogeneous sound. Everything in time, and in physical alignment, even your low frequency subs. Keep sound as close to point source as you can reasonably do. The fact that our hearing does not perceive any directional quality below around 70hz does not mean that the speaker's mechanical cone focus does not present a directional wave. I postulate that that wave must be homogenous with the rest of the primary speakers wave focus. This also helps explain why dual subs, linked closely with the primary speakers adds to the perception of the entire stereo image as well. Try it and see for yourself. Be sure and space the speakers from the side and front walls and other reflective surfaces for your test, even if impractical for daily use. After over 60 years of tinkering and experimenting it was quite a surprise for me. Of course, no 2 rooms are the same, unless built specifically same. And as always, set height, toe in/out etc. and listening positions according to your room acoustics and personal taste.
@@OldTooly I have about 50yrs of working with this addiction, and yes you're exactly right....the subs ALWAYS sound better in the same plane as the tops. Always get as close to point source as possible.
Paul's answer doesn't conflict with that fact. He only advocates you get that band perfectly reproduced, and he asserts that to do that you need a system that is capable beyond that band.
At 76 years of age, my poor old ears cut off completely above 8 kHz. I doubt that I could detect any phase issues from equipment that did not extend much above 20kHz.
Sounds like my 68 year old ears. Hearing test shows my ears start dropping off at 3.5 Khz, nothing above 8 Khz. I can't detect any difference when changing the treble control from min to max. My kids can but I can't.
If we have a natural 25Khz sinusoid and at 25.5Khz sinusoid, would we hear a beat frequency of 500 Hz? If the two sinusoids were within the audible range but have the same frequency difference we would hear a 500Hz tone in addition to the other two. So, if our music has components that extend into ultrasonics like, for example percussion, its seems to me ultrasonic performance would be important to achieve the best quality reproduction in addition to elimination of the phase distortion associated with a rolloff.
Good analogy. A Bugatti which can set the speed record for production cars is a pretty nice place to be even at the speed limit. Making stable at 200 makes it handle pretty well at 60.
Of course, its important to make the distinction in that speakers are physical systems, and electronics which needs more bandwidth to contend with having sufficient feedback at high frequencies and in digital you have to contend with the Nyquist frequency and the need for antialiasing filters. Moreover even in healthy young people the ear's response rolls off SHARPLY above about 12 kHz. See "Fletcher-Munson curves" if you need convincing.
@@WeeWeeJumbo Sure it does. He may be wrong about the 10k part, but if your amp is -3dB at 20KHz, and your speakers are -3dB at 20KHz, the result will be -6dB at 20KHz.
Hilarious...Paul "Flexing" in front of that monster system. Yes we are all Sooooo! jealous at your access to a system of that caliber everyday. We also hate you metaphorically. All in fun, we congratulate your success...I also thank you for my MKII DAC and S-300 Amp. I must say, I like your flexing sir...P.S. You need to pull those sub banks out. They are too far back. lol me telling the expert. Still think I'm right. Time arrival issues in my mind.
I always thought that my investment in high-end audio was so my pets can enjoy the music also since dog hearing is: 67 Hz-45,000 Hz, and cat hearing is: 48 Hz-85,000 Hz. Another thing to note is you need to be careful as dogs can hear sounds of a volume as low as 5-15 dB (note that a human whisper is 20-30 dB). With dog hearing being more sensitive, it makes sense that quieter noises (85+ dB) hurt canine ears, so again be sensitive with your volume levels. Fun fact: On the Beatles album Sergeant Pepper’s Lonely Hearts Club Band, at the end of the track “A Day in the Life,” there’s a high-frequency sound that only dogs can hear.
Paul, enjoyed your explanation on ultrasonics and phase shifting. What are those monster speakers you are standing between? Are the the Infinity speakers you’ve referenced several times? Are those just tweets in array out front and bass and mid drivers in back? Wow!
The average 35 year old male has 11 dB hearing loss at 8KHz and probably no hearing above 15KHz. Speaker and headphone manufacturers who claim response to ultrasonics are playing to the technically challenged customer. Be very careful testing your hearing by playing high frequency tones through your headphones. The test should only be done at normal volume. It is tempting to play a 15KHz tone or above and increase the volume until you think you can hear it. This is very dangerous and may cause hearing damage.
Typical Audiophile, worrying over trivia. Phase shift hits our ears in everyday activities, including at a live performance. Loudspeaker drivers, crossovers and room acoustics produce absolutely massive amounts of phase shift So what?
If you think "phase" matters, measure speakers in a room lol. Its not uncommon to have tens of thousands or even _hundreds of thousands_ of degrees of phase shift indicated in the measured response due to the propagation delay and reflections. Does it matter? Absolutely not. There are places where it does, but not here...
Well reflections are bad, right? Would you want to listen to your stereo in a glass room? Plus, we're not talking about phase shift between speakers, or due to propagation delays. If you add two 10KHz sinewaves that are exactly in phase, it's going to be louder than two 10KHz sinewaves that are not in phase (all else being equal).
@@Hyxtryx There are two problems. The first is that you not only cannot eliminate all the reflections, but doing so is highly undesirable as it severely penalizes the sound of the speakers. With principally direct sound, conventional loudspeakers wont sound correct at all. There are various psychoacoustic reasons for this that are beyond the scope of this discussion. The second is that two speakers playing 10 kHz in a room will sum in very complex ways. The room will have a complex pattern of nodes and antinodes due to interference (an effect called comb filtering). It won't simply be 2x louder at the listening position due to the complex path it takes thru the room. Remember: Well designed speakers have a diffuse sound field like an old-school soft-white lightbulb. Most typical hi-fi speakers radiate over 100-160 degree angle across the front baffle. Only a small portion of the sound gets there directly. The rest gets there by reflecting off of things in the room.
@@mysock351C I'm thinking of tones coming from a single speaker, just to keep things simple and to make a point. If you're not centered between 2 speakers, obviously there's going to be a phase difference hitting your ears. And room reflections, yes. That happens in live settings as well. There's so many variables, why not try to eliminate the ones that you can, to get a little bit closer to "perfection"? What about headphones? That all but eliminates reflections. Does that mean a headphone amp needs to be designed to a higher spec than a regular amp? And, I just thought of this now... A similar thing can be said about Total Harmonic Distortion. A little googling says that 0.5% is good for a tube amp, not that good for many solid state amps, but it is fantastic for a speaker. Does that mean that THD doesn't matter for an amp, because your speaker will always have way more? If your amp has a flat 50KHz bandwidth and low THD, that equates to piece of mind, that at least that part of the signal chain is not coloring the sound.
Great explanation Paul! This makes alot of sense
Thinking is the hardest work there is, which is why so few people engage in it.
I do find that top-end response characteristics make a difference to me.
Lead the way Paul! I have said for decades that one of the problems with the audio industry is being stuck in the 50's and 60's test and measurement. Back then, which I was, even generating a good clean sinewave was expensive. Much less variable frequency with accurate display. We did not have digital readouts. I challenge the claim that we had equipment that was flat enough in the '50's to accurately reproduce in order to measure the 20Hz-20Khz way back when it was first claimed. Much less time domain anomalies. Dr Matti Otala's TIM in the '70's was the first I saw that moved past simple steady state sinewave.
But our auditory system just like music is not based on frequency. It is based on time domain, variations, transients... Phase shifts and delays are more audible than slight variations in frequency response. But not measured with our current T&M.
We can get a better measurement of phase response accuracy by using squarewaves. I was taught in Waveform Analysis 101 that the first ten harmonics of the fundamental are as critical as the fundamental. Thus a ten times bandwidth is needed to reproduce a fundamental accurately. So to accurately reproduce to 20Khz, you need a bandwidth out to 200Khz.
But yes a good phase based measurement is needed! One paper I read shows that the "phase" difference between ears, the ability of our auditory system to resolve phase, transient reception between ears is as short as 10-20 microseconds! The human auditory system can resolve detail as high as >100Khz. Just not a stead sinewave.
Well said!!! We "hear" much more than frequency. It's evolutionary. The hair on our arms, sounds from behind us, time, reverb, phasing, etc... We pick up a lot of info, a lot of ways when "listening". 👍👍
I enjoy these Q&A. The comments/discussion are as interesting as the video.
An extended high frequency response also can affect apparent transient ease. I was playing a solo acoustic guitar recording and I decided to connect a tweeter with more high frequency extension. The sound took on such an effortless quality on the transients, as he plucked the strings. The system just sounded like it was right on top of things with the more extended tweeter. Spectral power amps have been known to go up to 1 million hertz, according to their specs. Some in the forums said that you had to use a special constructed speaker cable with them, for them to honor the warranty on the amps.
Just use 18" dia. copper plumbing coax with nitrogen for the dielectric. Runs out to several hundreds of Mhz, handling power up into megawatts. Guaranteed no phase shift.🤣
If the transient response is even a tiny bit slow, the whole thing sounds like a reproduction; and a mechanical one at that. That's one of the reasons I use great cables, and on phono, a moving coil cartridge or some equally exotic type, in place of moving magnet. That slight delay in hitting the leading edge of notes, gives it away & tells the ear that it's only a reproduction.
@InsideOfMyOwnMind Try some Heliax, nothing better for high power rf.
So the only difference between the first tweeter and the 2nd tweeter was the high frequency extension? You're instantly ruling out other factors?
@@ianbigsand7 Is that indicative of helium for moisture abatement as opposed to atmosphere for the dielectric? I know I said nitrogen because that's what they use to displace moisture in some other disciplines. I know you can have actual tonnage of that stuff in large combiner rooms feeding large community towers.
Speakers look absolutely magnificent in the background .
Love your Infinities !!!
I’ve Got the QLS-1’s !! I’ve Had them since 1978 !!
Although the topic is interesting, the IRS v's steal the show - beautiful line array and gorgeous they certainly are.
Paul is right... phasing is important but not as an independent function.
What needs to be appreciated is that you are not feeding your amplifier 12khz, and 22khz and 30hz all at the same time. You are actually sending it a single voltage that changes over time. On an oscilloscope this would appear as a 30hz waveform with the 12 and 22 khz components riding on top of it... sort of a fuzzy 30hz waveform.
That waveform is also how you are asking your speaker cones to move.
All the rest of it ... the FFT, spectrum, frequency etc. are all mathematical abstractions taken from that single fuzzy waveform. While they have value in examining how things are behaving, they are not exactly what you are asking your system to do.
The answers lie in the "Transfer Function"... All electronics introduce a delay, it takes a finite amount of time for the outputs to react to the inputs. This also limits how well they can respond to rapid changes, ideally they should be able to reproduce a change fully before the next change comes along.
So, what happens is that we have our fuzzy 30hz waveform. The amplifier might be fast enough to perfectly track the basic 30hz part but the smaller bits riding on the waveform may fall victim to "propogation delay" and "slew rate" as above and not be fully reproduced.
Whether this is crucial at frequencies we cannot hear is an ongoing debate. The fact is the inability to accurately reproduce the smaller more frequent "fuzzies" on our waveform does, in fact, change the waveform itself in very subtle ways.
But as always, the question is whether that is audible or not...
Paul, when he said 44 kHz, that's such an odd number to throw out there that I think he was referring to the CD sample rate (44.1 kHz), or confused it with the CD frequency response (to 22 kHz). He was apparently talking about his components playing to higher frequencies than his speakers. That's what it sounded like to me anyway. I think maybe you misunderstood what he was asking because his question was pretty confused.
Paul’s answer was pretty confused, too
The question wasn’t stated very clearly, which happens quite often, and neither was the answer, which also happens quite often!😵💫🤣
It's interesting that you discuss the audibility of phase shift in the highest octave while not discussing the significant phase shift in coil and capacitor based crossovers. An LR4 crossover, for example, walks through a full 360 degrees of phase shift in the source material in the octaves above and below the crossover frequency. Since a square wave is a combination of the fundamental and its odd harmonics correct in both amplitude AND phase, it is impossible for a speaker with such a crossover to accurately reproduce a square wave. The fundamental and its harmonics simply can't all align.
An accurate digital FIR phase free filter, perhaps including time alignment and some phase error compensation for the drivers, is essentially the only way (other than using strictly 6 db/octave crossovers) to enable a speaker to accurately reproduce a square wave. This reflects on its ability to keep fundamentals and their harmonics time and phase aligned -- which as you indicated -- can affect the sound of the music.
*** Paul how you feel legacy speakers dealing with the flagship Valor and microphone set up parameters..
Firstly, response concatenates, a -3dB at 20KHz pre-amp and a -3dB at 20KHz power amp’s -3dB point is 14.4KHz. (This is another longer subject BTW) The question is around that phase begins to shift a decade down, so for a 20KHz filter, phase starts to shift at 2KHz. Pauls explanation is grounded in good Eng. Lets play with a 7.5KHz tone (nearly double highest piano note) and its 15Khz first harmonic. Phase magnitude matters, so what is say, 10 degrees excess phase shift at 15Khz due to BW cut-off going to look like. Well the wavelength of the 15Khz harmonic is 2.28cm and a delay due to 10 Degrees phase shift is 2.28/360, * 10 = 0.63mm (24thou) Keeping in mm, sound is 343,260,000mm/ sec, so a 0.63mm delay represents a = 0.000000018s (1.8nS) delay in the 15KHz harmonic of 7.5KHz. You have to remember your ears don’t work at ~2nS = 500MHz. This said, my main amp has a -3dB limit of 200KHz, and the great Rupert Neve had 200KHz in his elite mixing desks. You always lose when you win, so extended BW brings issues of intermodulation distortion possibly increasing etc.
"That said..." I was awaiting a big conclusion. And you came and told about you amp and someone's desk and blabla :/
Audio Guru🙏
There are similar rules-of-thumb with wide-band voltmeters and oscilloscopes. Whatever the high-frequency -3dB cut-off is, if you want accurate amplitude measurements, stay a decade below the -3dB cut-off. If you want accurate phase measurements, stay TWO decades below the -3dB cut-off.
Given a -3dB spec of 50khz, given a single dominant pole roll off, the phase is -45 degrees. One octave below 50 kHz, it is -26degrees. At one decade below 50 kHz, the phase is -5.6 degrees.
Well…. At 65 I can only hear up to 11khz., but I love my stereo’s music production 😀
Headroom is one thing. But the phase of the signal doesn't shift simply from increasing frequency.
still i dont understand how we feel or sense those ultrasonic harmonics if our ears cant hear them
Speaking of phase shift / delay, why are the subs in a different time zone ?
I asked the same question a while back so I made sure that my dual subs weren't. What a fine improvement it was in all aspects. Since I experiment a lot, less so now since I have gotten a system sound so close to what I want, I have come up with a personal concept on speaker design and setup. I call it homogeneous sound. Everything in time, and in physical alignment, even your low frequency subs. Keep sound as close to point source as you can reasonably do. The fact that our hearing does not perceive any directional quality below around 70hz does not mean that the speaker's mechanical cone focus does not present a directional wave. I postulate that that wave must be homogenous with the rest of the primary speakers wave focus. This also helps explain why dual subs, linked closely with the primary speakers adds to the perception of the entire stereo image as well. Try it and see for yourself. Be sure and space the speakers from the side and front walls and other reflective surfaces for your test, even if impractical for daily use. After over 60 years of tinkering and experimenting it was quite a surprise for me. Of course, no 2 rooms are the same, unless built specifically same. And as always, set height, toe in/out etc. and listening positions according to your room acoustics and personal taste.
@@OldTooly I have about 50yrs of working with this addiction, and yes you're exactly right....the subs ALWAYS sound better in the same plane as the tops. Always get as close to point source as possible.
The phase is supposed to change as the frequency response falls. That's how minimum phase works. Minimum phase is natural.
Do musical instruments have any output at 50 khz? You say our ears are sensitive to phase shift but can that be proven?
Philips decided for us, with the making of the CD that 20Hz-20.000Hz is all you need ...
Paul's answer doesn't conflict with that fact. He only advocates you get that band perfectly reproduced, and he asserts that to do that you need a system that is capable beyond that band.
I always thought more headroom in a car was move the seat further back
At 76 years of age, my poor old ears cut off completely above 8 kHz. I doubt that I could detect any phase issues from equipment that did not extend much above 20kHz.
Sounds like my 68 year old ears. Hearing test shows my ears start dropping off at 3.5 Khz, nothing above 8 Khz. I can't detect any difference when changing the treble control from min to max. My kids can but I can't.
If we have a natural 25Khz sinusoid and at 25.5Khz sinusoid, would we hear a beat frequency of 500 Hz? If the two sinusoids were within the audible range but have the same frequency difference we would hear a 500Hz tone in addition to the other two. So, if our music has components that extend into ultrasonics like, for example percussion, its seems to me ultrasonic performance would be important to achieve the best quality reproduction in addition to elimination of the phase distortion associated with a rolloff.
I just can't take my 👀 off those speakers!! What he say , what he saaaay.
I know. My cars have to do at least 220 mph.
Good analogy. A Bugatti which can set the speed record for production cars is a pretty nice place to be even at the speed limit. Making stable at 200 makes it handle pretty well at 60.
@@user-od9iz9cv1w
th-cam.com/video/LO0PgyPWE3o/w-d-xo.html
th-cam.com/video/PrqYohBV58o/w-d-xo.html
Of course, its important to make the distinction in that speakers are physical systems, and electronics which needs more bandwidth to contend with having sufficient feedback at high frequencies and in digital you have to contend with the Nyquist frequency and the need for antialiasing filters. Moreover even in healthy young people the ear's response rolls off SHARPLY above about 12 kHz. See "Fletcher-Munson curves" if you need convincing.
People who ask questions like this should get out more!☹️
Lol I agree. I should get out more. 😂
Why? Is it not an interesting question?
@@anthonyconrad8183 I thought it was a great question and a thought provoking answer. Keep at it Anthony!
@@WeeWeeJumbo No!☹️
How about this one:
If every component was doing 20kHz at -3dB, the combined response would roll off at much lower frequencies, probably below 10k.
What are you talking about? That doesn’t make the least bit of sense, are you lost?
@@WeeWeeJumbo Sure it does. He may be wrong about the 10k part, but if your amp is -3dB at 20KHz, and your speakers are -3dB at 20KHz, the result will be -6dB at 20KHz.
Hilarious...Paul "Flexing" in front of that monster system. Yes we are all Sooooo! jealous at your access to a system of that caliber everyday. We also hate you metaphorically. All in fun, we congratulate your success...I also thank you for my MKII DAC and S-300 Amp. I must say, I like your flexing sir...P.S. You need to pull those sub banks out. They are too far back. lol me telling the expert. Still think I'm right. Time arrival issues in my mind.
Because the art of showing FR graphs with phase is a thing of the past to protect the innocent ..!
Lol
I always thought that my investment in high-end audio was so my pets can enjoy the music also since dog hearing is: 67 Hz-45,000 Hz, and cat hearing is: 48 Hz-85,000 Hz. Another thing to note is you need to be careful as dogs can hear sounds of a volume as low as 5-15 dB (note that a human whisper is 20-30 dB). With dog hearing being more sensitive, it makes sense that quieter noises (85+ dB) hurt canine ears, so again be sensitive with your volume levels. Fun fact: On the Beatles album Sergeant Pepper’s Lonely Hearts Club Band, at the end of the track “A Day in the Life,” there’s a high-frequency sound that only dogs can hear.
Paul, enjoyed your explanation on ultrasonics and phase shifting. What are those monster speakers you are standing between? Are the the Infinity speakers you’ve referenced several times? Are those just tweets in array out front and bass and mid drivers in back? Wow!
44kHz? He might be talking about sample rate wich I think has more to do with the digital conversion.
You hifi guys are crazy 😂
only harman/kardon has ultra wideband sound.
At their old glory days. Now there is no HK as a manufacturer. It is only a name that Samsung uses to play with.
The average 35 year old male has 11 dB hearing loss at 8KHz and probably no hearing above 15KHz. Speaker and headphone manufacturers who claim response to ultrasonics are playing to the technically challenged customer. Be very careful testing your hearing by playing high frequency tones through your headphones. The test should only be done at normal volume. It is tempting to play a 15KHz tone or above and increase the volume until you think you can hear it. This is very dangerous and may cause hearing damage.
Such an "audiophile" response
Typical Audiophile, worrying over trivia. Phase shift hits our ears in everyday activities, including at a live performance. Loudspeaker drivers, crossovers and room acoustics produce absolutely massive amounts of phase shift So what?
If you think "phase" matters, measure speakers in a room lol. Its not uncommon to have tens of thousands or even _hundreds of thousands_ of degrees of phase shift indicated in the measured response due to the propagation delay and reflections. Does it matter? Absolutely not. There are places where it does, but not here...
Well reflections are bad, right? Would you want to listen to your stereo in a glass room? Plus, we're not talking about phase shift between speakers, or due to propagation delays. If you add two 10KHz sinewaves that are exactly in phase, it's going to be louder than two 10KHz sinewaves that are not in phase (all else being equal).
@@Hyxtryx There are two problems. The first is that you not only cannot eliminate all the reflections, but doing so is highly undesirable as it severely penalizes the sound of the speakers. With principally direct sound, conventional loudspeakers wont sound correct at all. There are various psychoacoustic reasons for this that are beyond the scope of this discussion. The second is that two speakers playing 10 kHz in a room will sum in very complex ways. The room will have a complex pattern of nodes and antinodes due to interference (an effect called comb filtering). It won't simply be 2x louder at the listening position due to the complex path it takes thru the room. Remember: Well designed speakers have a diffuse sound field like an old-school soft-white lightbulb. Most typical hi-fi speakers radiate over 100-160 degree angle across the front baffle. Only a small portion of the sound gets there directly. The rest gets there by reflecting off of things in the room.
@@mysock351C I'm thinking of tones coming from a single speaker, just to keep things simple and to make a point. If you're not centered between 2 speakers, obviously there's going to be a phase difference hitting your ears. And room reflections, yes. That happens in live settings as well. There's so many variables, why not try to eliminate the ones that you can, to get a little bit closer to "perfection"? What about headphones? That all but eliminates reflections. Does that mean a headphone amp needs to be designed to a higher spec than a regular amp?
And, I just thought of this now... A similar thing can be said about Total Harmonic Distortion. A little googling says that 0.5% is good for a tube amp, not that good for many solid state amps, but it is fantastic for a speaker. Does that mean that THD doesn't matter for an amp, because your speaker will always have way more?
If your amp has a flat 50KHz bandwidth and low THD, that equates to piece of mind, that at least that part of the signal chain is not coloring the sound.