I remember noticing, as a child (long before I'd heard of comb filters) that saying "sssshhh" while walking towards a wall produced an interesting sound like a passing jet plane. It occurred to me at the time, that I could use this to find the walls if I were ever in a room without any light. I used to practise by walking towards walls while making noise with my mouth... it was surprising to me at the time, how easy it was to stop right next to the wall before hitting it.
The professionalism and quality of his content tells me he is a subject matter expert. His youth, facial structure, and skin tone tells me I can purchase mushrooms off of him.
😂😂 let's listen to Pink Floyds dark side of the Moon in 8D😂😂 some songs are actually built into the floor the surrounds are play surround the room at different levels😅
OMG finally someone who knows what he is talking about. THANK you for making this most excellent and perfectly produced video. For years I have been fighting this rediculous idea of building center channel speakers with an array of drives positioned along a horizontal line under the TV. Everyone in the room other than the one person who gets to sit in the dead center will be experiencing comb filtering of the center channel material and that of course is mostly the human voice. Adding any type of distortion to the human voice is never a good thing. When I built my center channel speaker I susptended it vertically directly above the center of the TV aimed slightly down so it’s on axis with the average head height of the audience. This all but eliminates comb filtering. PLEASE do a video on this topic. You could really do it justice and perhaps enlighten the minds of people who think you just buy the speaker and place it under the TV. Thank you again for your great work.
Why not put one (or two) driver(s) above and one (or two) driver(s) below the screen? If drivers are not equidistant from heads, delay according to the average angle difference? Should create a phantom center right from the screen. If projection screens are made from cloth or perforated, they put speakers behind 'em.
I am a high school student who just got my first paid gig for sound since people knew me for working with it for our live musicals/plays. Your videos have been wildly helpful for me since there is no one available to further train me on our sound system. Thank you so much!
The accuracy of explanation in this video is just absurd. Excelent repetition of information, examples and overall quality. Thank you so much. Comb filtering occurs everywhere, it's so important to understand it. But it it also important not to be driven mad by it. This video covers it all.
Excellent subject, excellent explanations ! Just a small note : At 10:21 the delay is O.5 ms, not 5 ms. Otherwise the first cancellation would be at 100 Hz, not 1 KHz. The cancellation happens at half the period, which is 0.5 ms for a 1KHz signal having a 1 ms period. And at 13:12 the graph shows clearly a 0.0005 sec delay, which is 0.5 ms, not 5 ms.
Outstanding!!! This explains why the noise coming from a flying drone - recorded with a microphone on a tripod - sounds (and looks) much different than when it's recorded with a ground-plane microphone. Comb-filtering shows up because of the reflections coming off the ground, and using ground-plane microphones eliminates these reflections. It's very illuminating to see how the comb-filter changes with respect to the time delay! Such a remarkable demonstration. Thank you!!!
Intelligent delivery with no dumbing down. I never fail to learn something. We all get that back of our mind doubt about a problem we don't know is a problem yet. Nice to spend a evening here and stumble into that Eureka Moment.
Listening to music = 1 - enjoying the music, the beautiful sensations it gives you; Making music = 1 - enjoying the music, the beautiful sensations it gives you; 2 - enjoying the fact that you will offer the others the beautiful sensations that you love; 3 - enjoying all the complexity and all the beauty that goes into that beautiful final product.
YES okay this just helped me solve the weird phase sounds I get from my high schools lectern that uses dual mics when someone moves closer to one or another. THANK YOU
best explanation on youtube, thank you! one thing i wouldve liked to hear more about though is the pattern of frequencies that get notched/cancelled by comb filtering at the end of the video and specifically why that is the way it is
Great suggestion. It also doesn’t help that I misspoke when saying the amount of time shift. Here is a calculator that you might experiment with: www.mh-audio.nl/Acoustics/FFR.html
@@AudioUniversity Thanks so much! That site is a super helpful practical tool for figuring out how comb filtering might be affecting my actual recording situations! Still trying to understand the actual pattern and all a bit more as far as how it mathematically unfolds (something to do with the harmonic series from what Im reading), but your video and that website are a huge help along the way! 🤜🤛
First video I've seen on this channel and the first video I've seen explaining the illusive Comb Filter. You good Sir just earned yourself a subscriber. Thank you.
The verbal economy of these lectures is a paradigm. He concentrates exclusively on the subject matter and has banished autobiographical and self-indulgent topics, both of which are a great temptation in a medium that has no material supply cost. In comparison, there is another RUclips channel, home stereo related, always with promising titles but whose contents are full of the word "I" and whose presentation and content is not distant from this-- search, "grandpa simpson onion on my belt."
5 second delay causing the same pattern of comb filtering in the sweep and white noise (i.e. 1Khz) is crazy. I'm gonna need a lot more absorption panels in my home studio. Thanks!
I work in electronics a d what you are describing is a comb filter EFFECT, not a comb FILTER. A comb filter is used in signal processing to introduce destructive or constructive signal at the output by applying a delayed version of the original signal. So a filter is an actual variable delay line sometimes used in analog video processing foe time base correction. What you describe is the comb effect that can be caused naturally as u described or by introduction of a delay as you did by shifting the audio signal also called phase shifting. Phase shifting can amplify or cancel out a signal such as in noise canceling headphones There it hears ambient sound with a microphone, phase shift it 180° and insert that sound into the audio stream. Thus cancelling the ambient sound signal. :)
My other hobby is Amateur (Ham) Radio. In radio communications, many transceivers have been touted as having comb filters as if this is a good thing, a feature, and found only on high end radios. Now I will have to go figure out how or what exactly they are using com filtering for and why it is a benefit.
10:10 volume down for that section… thanks for this refresher… i was eager for the 3:1 rule part 7:04 … hehe and thanks foe the guide. always useful, and really dl’d for support
You can also get comb filtering from thunder, jet flybys etc. from reflections off of the ground / surroundings, but in those cases it actually sounds good!
Good speak! This is why when making recommendations on installed systems, my first option is a single speaker array that is dead centre. It shows in the acoustic simulation software. All too often center placement is not a feasible solution and so, we compromise by spreading the speakers to the left and right.
What a very interesting and informative series on the various aspects of sound reproduction. Much of it I had known years ago but age tends to restrict memory! Very well presented!
This is literally why in my synthesizing process I tune my delay to about 000.5 secs, just to get that resonance from the comb filter. Thank you for explaining why.
Sehr interessant und genau das was viele nicht wissen. Viele Lautsprecher, ein (unbenutzter) Equalizer und mächtig Lautstärke. Das allein zählt... es gibt halt viele Ignoranten.
Underrated channel..... So much useful content explained in a precise and proper way.. I am preparing for an exam and these are so useful... thanks man... cheers :)
Great description! As an RF engineer, I have always called 'comb filtering' as 'multipath fading.' No matter what you call it, the effect of the troughs is amazing large. Multipath fading causes a plethora of frequency nulls and crests throughout an auditorium and the locations of people affect the frequency response of a room. Years ago, I built a scalar audio network analyzer using a sinewave generator, a wideband flat amplifier, an oscilloscope, and a wideband microphone. It was controlled by a laptop driving a GPIB bus. The program sweeping frequencies included a calibration sequence to ensure the tones heard by the microphone were adjusted from 10Hz to 20kHz to be flat. Result: there were dozens of nulls in a typical house room and hundreds in an auditorium! What's more, the entire spectral response changed drastically as people were added to the room and moved around! Also, these spectral responses changed as the microphone was moved, as expected. Point is, you can never adjust equalization to the satisfaction of 90% of people in a room, and those 10% who are content with the spectral response will change as people move around in the auditorium. There is therefore no point in trying to be precise with adjusting your mixing board in an auditorium. The spectral responses of a large room are too numerous and varied to allow simple descriptions like this. The mixer adjustments only have meaning in small highly dampened rooms of a recording studio and in the subsequent mixing of recording tracks. The small room creates fewer nulls and the dampening sounds panels (or even bedsheets and laundry) reduce the multipath effect since the reflected signal is greatly attenuated as demonstrated in this video. Great demonstrations, Kyle!
Thanks for this explanation. Very useful info for those getting into audio recording. There's actually a lotta math--and history--behind this phenomenon... Christiaan Huygens, inventor of the grandfather clock, observed the (comb filtering) audio effect we now know as 'flanging' in 1693... Talk about 'back in the day'... That's even older than the Beatles, whose producer, George Martin, used tape-reel flanging!
3 to 1 is for full cycle full phase !is to combats the combfiltering .😅 If you remember this pls remember about separation filters between grups of sympfonic instruments!! The play same notations usually on near registers and could naturally occur the comb effect also on recordings !!!!😅
Thank you, now I understand the other side of Comb filtering. I use Comb filters on purpose for synthesis, like for physical modeling. This video taught me everything else I didn't know about Comb Filtering. much appreciated ...subscribed.
Excellent channel with pertinent data on the subject of sound engineering. Keep up the good work, your information is so relevant and important to managing sound.
Number 5: you use the comb filter of your Synth. Yeah, it is not always bad and can be used musically. But as long as you don’t add a comb on purpose you should get rid of it. Well explained.
This is great stuff. Thanks for you hard work. Although outside the scope of this video I would love to hear an explanation for the comb filtering that happens at outdoor venue when, the wind blows, the temperature changes and as day changes to night.
Thanks a lot for that amazing video!!!! Very interesting that explanation mixing white noise and the sweep. And the examples moving the reflector or the mic movement, really useful. Congrats!!! But I think that there is a mistake in the delay value, because is 0,5ms, not 5ms. If you put 5ms, the first cancellation appears at 100Hz. 5ms is the 200Hz period, not 2kHz.
Really cool video, excellent examples. I think that would be intetesting to make a video about Linear Phase EQ. Maybe talking about how EQ affects the phase of the input signal, or why it is important to use linear phase when equalizing.
hi im a amateur radio operator and i use xlr cables on my microphones and mixers and when i changed a lead on my mixer and was getting rfi in my cables when i transmitting, thats why i allways use ballanced leads.
An absolutely great educational video, its clarity and simplicity are among the best in the entire RUclips. Many thanks and bravo! A couple of questions: Comb filtering effect happens naturally, it is caused by a phase shift between two copies of the same signal. How will the frequency comb look like with three or more copies at different delay times? And are there any comb filters built to process signal and when are they used? Thank you!
Multiple comb filters occur in every room. It yields a bunch of combs on top of each other. To see a graph, look at an acoustic measurement of a room. Comb filtering is used in many many ways. One example in audio is flanger. However, as a few others have stated in the comments, comb filtering is also a tool in programming and computing!
Not to take anything away from your very professional and useful presentation, but I was also hoping to hear all about phase shifters, flangers, and comb filters used in synthesizers.
I don't know why I am watching this but it seems I suddenly understand filters in digital signal processing , which is a big part of what I am studying.
My listening environment is so bad I wonder if some aspects help me both the natural "First reflection points" are dead zones(out the dore one side and behind a raised fish tank the other. My speakers are 6' apart but that is already starting to shroud them. I am about 11' foot away on the perpendicular of the line between them, although I wish I could bring the sweet spot nearer me I fear I may never learn how to if it is even posible. I am comforted to see that the comb effects happen increasingly often as the tones get too high to hear. very interesting videos. All the best. Note I normally complain I can't hear sound differences on you tube "Sound tests" but yours illustrated your points brilliantly.
The sine sweep signals were often used in Atari video games as sound for different character movements or actions. For example, Pac man eating the pebbles or the ghosts. Very interesting.
My omnidirectional OHM G original style single Walsh driver speakers are very sensitive to this effect and need to be nearly 2 feet from the wall with drapery deadening to present a clear and accurate "sound stage" for the listener. The room and speaker placement (in 3 dimensions) of regular front fire speakers are less critical but still in play to accurately receive the intended sound that comes from the source media without room induced distortion.
Very straightforward thank you . My question is if there is a delay can one simply drag the track back into place? Iv heard people say yes and others say this can cause more problems. I’m not really sure what the answer is to that. I would think it’s fine but I’m not sure why someone would say it can be a bad idea . It seems recording a drum set would be impossible tho with all the mics and no 3 to 1 ratio being implemented
Good question, @Hey There! You’re right with the drum kit - you can only align the sound of one component of the kit in each mic because a shift here will cause a shift there, and so on. If the issue is a single instrument is delayed, such as the guitar amp example in this video, shifting the track could help. Just remember that there’s no guarantee that the indirect sound from reflections will be aligned on each track. If there is an acoustic comb filter, there isn’t much that can be done, unfortunately.
Could you please explain why it is better to use a cardioid microphone instead of a shotgun microphone for indoor recordings? However, in practice, shotguns such as the Sennheiser MKH 416 are always used for interviews and film recordings. Unfortunately, I haven't been able to get a clear answer to this question from anyone. Furthermore, it would be very interesting to have a detailed understanding of how the interference tube on a shotgun microphone works.
@@AudioUniversity I loved your analysis Kyle when comparing two identical signals that are out of phase with each other. So the dead spots where particular frequencies cancel each other out are mathematically predictable and you describe how to mitigate the effect. With vocal doubling the two signals are similar but with unpredictable differences. I know that vocal doubling is heralded as a way to enhance a vocal but it's clear from your analysis that the interaction can degrade the quality of the sound too. So I just wondered if there are ways to identify that and to mitigate against the effect. I'm totally naïve here but I could imagine cutting the second vocal track into suitable fragments and shifting each fragment back and forward in time w.r.t. the first vocal, listening for a sweet spot. That would take a lot of effort and it's a bit of a sledgehammer approach. Please educate me :-)
Wow, good thing I subscribed to you guys Audio University. Truly I’m learning a lot from you guys...thanks thanks thanks...nice presentation guys...it’s a A+ from me...
Great video thanks for posting it I just wanted to clarify the relationship between the delay time and the lowest cancelled frequency as you may have made a slip of the tongue. In your demonstrations (the first of which is at 9:15) you say that you’ve delayed the 2 sounds by 5 ms and that results in the lowest cancellation being at 1000Hz. I think you’ve actually delayed the 2 signals by 0.5 ms (the time display shows 0.0005 seconds so that’s 0.5 ms) , in which case the lowest cancellation would be 1000Hz as you say. However, a 5ms delay between two identical signals causes cancellations at 100Hz and all odd harmonics of 100Hz. This is because 5 ms equates to half the period of 100Hz and therefore a time shift of 5ms is equivalent to a phase shift of 180 degrees.
@@AudioUniversity A great video, but this error would have left me confused had I not come to read the comments! So my question with comb filtering involves hearing aids. One company has claimed their low latency of 0.5 ms fixes this problem of comb filtering caused by interference of world sound directly hitting the ear drum and the delayed signal through the hearing aid, but it looks like they've only kicked the problem up to higher frequencies that may not be an issue for many people but not for a musician. I notice it in my hearing aids with open ear tips as certain frequencies sound way off. Like the open G on my bass compared to the harmonic octave which will will sound like G#. With closed domes lowering the outside sound this seems to clear this up. I believe my hearing aids have a latency around 5 ms -> lowest cancellation at 100 Hz. Do you think this is indeed comb filtering I'm experiencing?
Great video. It's great to be able to see (and hear!) some examples in action. I'm wondering: In a technical book about acoustics I've read that the frequency of the first notch in the comb filter would occur where the period is twice the delay time, i.e. freq = 1/(2t), where t is the delay in seconds. In your example, for t = 5 msec, this expression would return freq = 100 Hz, but as you say in your video, your graph shows the first notch at about 1 kHz. Am I missing something?
Kyle: "Let me teach you how to avoid comb filtering." Me: "Let me add a comb filter to my synthesizer patch and be in awe." Great video Sir, insta-subbed.
One thing that I noticed, when using non-linear phase EQs, is that when I blend the wet signal with the dry signal, I can hear something similar to the comb filtering effect.
I remember noticing, as a child (long before I'd heard of comb filters) that saying "sssshhh" while walking towards a wall produced an interesting sound like a passing jet plane. It occurred to me at the time, that I could use this to find the walls if I were ever in a room without any light. I used to practise by walking towards walls while making noise with my mouth... it was surprising to me at the time, how easy it was to stop right next to the wall before hitting it.
In fact this is the strategy for blind people to recognize obstacles, especially with clicking tongue sounds.
Kid memories are the best because you’re exploring the world with fresh eyes
I thought i was the only one acting that weird 😂
You are a very smart person
I wonder if bat's echo location is actually more like this than the classic pip and stopwatch view.
The professionalism and quality of his content tells me he is a subject matter expert. His youth, facial structure, and skin tone tells me I can purchase mushrooms off of him.
Dead. I died laughing. You killed me. Murderer. 💜💜
😂
Lmfao
Outrageous 😂
😂😂 let's listen to Pink Floyds dark side of the Moon in 8D😂😂 some songs are actually built into the floor the surrounds are play surround the room at different levels😅
I don't work with audio at all, but the way this guy talks and illustrates is phenomenal, I'm interested
OMG finally someone who knows what he is talking about. THANK you for making this most excellent and perfectly produced video. For years I have been fighting this rediculous idea of building center channel speakers with an array of drives positioned along a horizontal line under the TV. Everyone in the room other than the one person who gets to sit in the dead center will be experiencing comb filtering of the center channel material and that of course is mostly the human voice. Adding any type of distortion to the human voice is never a good thing. When I built my center channel speaker I susptended it vertically directly above the center of the TV aimed slightly down so it’s on axis with the average head height of the audience. This all but eliminates comb filtering. PLEASE do a video on this topic. You could really do it justice and perhaps enlighten the minds of people who think you just buy the speaker and place it under the TV. Thank you again for your great work.
Good idea, E! A video on horizontal vs vertical driver orientation is a great suggestion. Thanks!
Why not put one (or two) driver(s) above and one (or two) driver(s) below the screen? If drivers are not equidistant from heads, delay according to the average angle difference? Should create a phantom center right from the screen.
If projection screens are made from cloth or perforated, they put speakers behind 'em.
I am a high school student who just got my first paid gig for sound since people knew me for working with it for our live musicals/plays. Your videos have been wildly helpful for me since there is no one available to further train me on our sound system. Thank you so much!
Glad to read this! I hope the gig went well!
i came here in search for a eurorack comb filter lol. now i have learned something about comb filters. i‘ll take it.
The accuracy of explanation in this video is just absurd. Excelent repetition of information, examples and overall quality. Thank you so much. Comb filtering occurs everywhere, it's so important to understand it. But it it also important not to be driven mad by it. This video covers it all.
Thank you!
Wow, I began as a free-spirited musician...now I'm turning into a neurotic audiophile.
The simplest tutorial ever , subscribed.!!
Excellent subject, excellent explanations ! Just a small note : At 10:21 the delay is O.5 ms, not 5 ms. Otherwise the first cancellation would be at 100 Hz, not 1 KHz. The cancellation happens at half the period, which is 0.5 ms for a 1KHz signal having a 1 ms period. And at 13:12 the graph shows clearly a 0.0005 sec delay, which is 0.5 ms, not 5 ms.
I was looking for this comment, because the math wasn't matching up.
I am learning so much! Thank you
I like bros energy, makes these tutorials way better. thanks man
This is an outstanding explanation.
Thank you, Texas Blues Alley! I'm glad you enjoyed it. Also, thanks for the shout out on FB. It means a lot to me!
God evening sir. I am from Odisha INDIA. Sir I appreciate you explanation. It's very useful and outstanding . Thank you sir
Thank you for watching, Rajkishor!
Very nice. I didn't understand it all, but that's the beauty of being able to watch it again and again until it sticks.
Glad you like it, @THOMMGB. Thanks for watching!
Outstanding!!! This explains why the noise coming from a flying drone - recorded with a microphone on a tripod - sounds (and looks) much different than when it's recorded with a ground-plane microphone. Comb-filtering shows up because of the reflections coming off the ground, and using ground-plane microphones eliminates these reflections.
It's very illuminating to see how the comb-filter changes with respect to the time delay! Such a remarkable demonstration. Thank you!!!
Best Channel on RUclips, your explanations are incredibly well organized, easy to understand and of course useful. Amazing job
Thank you!
This is the university I had to graduated with! Thank you so much for your great work!
Intelligent delivery with no dumbing down. I never fail to learn something. We all get that back of our mind doubt about a problem we don't know is a problem yet. Nice to spend a evening here and stumble into that Eureka Moment.
Listening to music =
1 - enjoying the music, the beautiful sensations it gives you;
Making music =
1 - enjoying the music, the beautiful sensations it gives you;
2 - enjoying the fact that you will offer the others the beautiful sensations that you love;
3 - enjoying all the complexity and all the beauty that goes into that beautiful final product.
YES okay this just helped me solve the weird phase sounds I get from my high schools lectern that uses dual mics when someone moves closer to one or another. THANK YOU
My teacher makes comb filtering in supercollider and I never understood how it works. Now I do.thanks
Wow! That sounds interesting! Thanks, David. Glad to help!
best explanation on youtube, thank you! one thing i wouldve liked to hear more about though is the pattern of frequencies that get notched/cancelled by comb filtering at the end of the video and specifically why that is the way it is
Great suggestion. It also doesn’t help that I misspoke when saying the amount of time shift. Here is a calculator that you might experiment with: www.mh-audio.nl/Acoustics/FFR.html
@@AudioUniversity Thanks so much! That site is a super helpful practical tool for figuring out how comb filtering might be affecting my actual recording situations! Still trying to understand the actual pattern and all a bit more as far as how it mathematically unfolds (something to do with the harmonic series from what Im reading), but your video and that website are a huge help along the way! 🤜🤛
First video I've seen on this channel and the first video I've seen explaining the illusive Comb Filter. You good Sir just earned yourself a subscriber. Thank you.
Glad to hear that! Thanks!
The verbal economy of these lectures is a paradigm. He concentrates exclusively on the subject matter and has banished autobiographical and self-indulgent topics, both of which are a great temptation in a medium that has no material supply cost. In comparison, there is another RUclips channel, home stereo related, always with promising titles but whose contents are full of the word "I" and whose presentation and content is not distant from this-- search, "grandpa simpson onion on my belt."
Excellent. I hope your viewers appreciate how much work goes into explaining foreign concepts in such a clear and concise manner.
Best explanation of this phenomenon I ever encountered. ♥️
Wow! Thanks, Joris!
5 second delay causing the same pattern of comb filtering in the sweep and white noise (i.e. 1Khz) is crazy. I'm gonna need a lot more absorption panels in my home studio. Thanks!
The best video i could find on this topic
Your videos are well explained
Thanks for providing quality videos for free
I searched for comb filter as I needed to know what it actually does (McombMB) now I know what it actually is, so much great explanations
I work in electronics a d what you are describing is a comb filter EFFECT, not a comb FILTER. A comb filter is used in signal processing to introduce destructive or constructive signal at the output by applying a delayed version of the original signal. So a filter is an actual variable delay line sometimes used in analog video processing foe time base correction.
What you describe is the comb effect that can be caused naturally as u described or by introduction of a delay as you did by shifting the audio signal also called phase shifting. Phase shifting can amplify or cancel out a signal such as in noise canceling headphones There it hears ambient sound with a microphone, phase shift it 180° and insert that sound into the audio stream. Thus cancelling the ambient sound signal. :)
Interesting distinction, Roy! Thanks for sharing your perspective on this and the extra info.
@@AudioUniversity your welcome. I only like to post positive comments that are helpful to others
excellent demos!! I never understood what comb filtering was. Great explanations and demonstration!! Thanks!!!!!!
My other hobby is Amateur (Ham) Radio. In radio communications, many transceivers have been touted as having comb filters as if this is a good thing, a feature, and found only on high end radios. Now I will have to go figure out how or what exactly they are using com filtering for and why it is a benefit.
You can also put some fome on your microphone to block reflected sound from walls or speakers.
10:10 volume down for that section…
thanks for this refresher… i was eager for the 3:1 rule part 7:04 … hehe
and thanks foe the guide. always useful, and really dl’d for support
Thank you. I have seen videos on comb filtering, but none that explained/demonstrated it so well :)
Thanks!
You can also get comb filtering from thunder, jet flybys etc. from reflections off of the ground / surroundings, but in those cases it actually sounds good!
Good speak! This is why when making recommendations on installed systems, my first option is a single speaker array that is dead centre. It shows in the acoustic simulation software. All too often center placement is not a feasible solution and so, we compromise by spreading the speakers to the left and right.
Have you checked out L’Acoustics L-ISA? I think you’d find it interesting!
What a very interesting and informative series on the various aspects of sound reproduction. Much of it I had known years ago but age tends to restrict memory! Very well presented!
Glad you enjoyed it!
This is literally why in my synthesizing process I tune my delay to about 000.5 secs, just to get that resonance from the comb filter. Thank you for explaining why.
Excellent ! Really clear explanations & practical demonstrations.
Sehr interessant und genau das was viele nicht wissen. Viele Lautsprecher, ein (unbenutzter) Equalizer und mächtig Lautstärke. Das allein zählt... es gibt halt viele Ignoranten.
You´re doing an amazing job, your videos are helping a lot of people right now. Thanks for your effort, im sure it'll pay off.
Thanks, Alejandro! I really appreciate your support.
Always know the sound but didn't know what caused it and what it is called. Thank you so much, this is such a great video.
Glad to help, mura!
Underrated channel..... So much useful content explained in a precise and proper way.. I am preparing for an exam and these are so useful... thanks man... cheers :)
Glad to hear that! Please share with your classmates if you think it would help them!
Great description! As an RF engineer, I have always called 'comb filtering' as 'multipath fading.'
No matter what you call it, the effect of the troughs is amazing large.
Multipath fading causes a plethora of frequency nulls and crests throughout an auditorium and the locations of people affect the frequency response of a room.
Years ago, I built a scalar audio network analyzer using a sinewave generator, a wideband flat amplifier, an oscilloscope, and a wideband microphone. It was controlled by a laptop driving a GPIB bus. The program sweeping frequencies included a calibration sequence to ensure the tones heard by the microphone were adjusted from 10Hz to 20kHz to be flat.
Result: there were dozens of nulls in a typical house room and hundreds in an auditorium! What's more, the entire spectral response changed drastically as people were added to the room and moved around! Also, these spectral responses changed as the microphone was moved, as expected.
Point is, you can never adjust equalization to the satisfaction of 90% of people in a room, and those 10% who are content with the spectral response will change as people move around in the auditorium.
There is therefore no point in trying to be precise with adjusting your mixing board in an auditorium. The spectral responses of a large room are too numerous and varied to allow simple descriptions like this. The mixer adjustments only have meaning in small highly dampened rooms of a recording studio and in the subsequent mixing of recording tracks. The small room creates fewer nulls and the dampening sounds panels (or even bedsheets and laundry) reduce the multipath effect since the reflected signal is greatly attenuated as demonstrated in this video.
Great demonstrations, Kyle!
Thanks for this explanation. Very useful info for those getting into audio recording. There's actually a lotta math--and history--behind this phenomenon... Christiaan Huygens, inventor of the grandfather clock, observed the (comb filtering) audio effect we now know as 'flanging' in 1693... Talk about 'back in the day'... That's even older than the Beatles, whose producer, George Martin, used tape-reel flanging!
Wow! That’s amazing! I had no idea the discovery of this phenomenon happened THAT long ago. Thanks for sharing, Carl!
honestly these are so good I feel like I should be paying for this
3 to 1 is for full cycle full phase !is to combats the combfiltering .😅 If you remember this pls remember about separation filters between grups of sympfonic instruments!! The play same notations usually on near registers and could naturally occur the comb effect also on recordings !!!!😅
super succinct, super informative, and with auditory/ visual demonstrations? and no spam?
THIS CHANNEL SMACKS
Thank you, now I understand the other side of Comb filtering. I use Comb filters on purpose for synthesis, like for physical modeling. This video taught me everything else I didn't know about Comb Filtering. much appreciated ...subscribed.
Glad to hear that, @Audio Artisan! Thanks for watching!
great explanation! gives me great input for my phd work. Thank you for your support!
Excellent channel with pertinent data on the subject of sound engineering. Keep up the good work, your information is so relevant and important to managing sound.
Thanks, William!
Number 5: you use the comb filter of your Synth. Yeah, it is not always bad and can be used musically. But as long as you don’t add a comb on purpose you should get rid of it. Well explained.
This is great stuff. Thanks for you hard work. Although outside the scope of this video I would love to hear an explanation for the comb filtering that happens at outdoor venue when, the wind blows, the temperature changes and as day changes to night.
Thanks a lot for that amazing video!!!! Very interesting that explanation mixing white noise and the sweep. And the examples moving the reflector or the mic movement, really useful. Congrats!!!
But I think that there is a mistake in the delay value, because is 0,5ms, not 5ms. If you put 5ms, the first cancellation appears at 100Hz. 5ms is the 200Hz period, not 2kHz.
very professional, much useful information
Just copy paste Texas Blues Alley. Outstanding explanation.
Thanks, Peter! Glad you enjoyed it.
This was clear and very helpful. Thank you for posting.
Thank you for watching!
Really cool video, excellent examples.
I think that would be intetesting to make a video about Linear Phase EQ. Maybe talking about how EQ affects the phase of the input signal, or why it is important to use linear phase when equalizing.
Thanks, Leonardo! That’s a great suggestion.
I second that
Thanks so much - you are so clear and concise and easy to listen to that it makes understanding easy!
hi im a amateur radio operator and i use xlr cables on my microphones and mixers and when i changed a lead on my mixer and was getting rfi in my cables when i transmitting, thats why i allways use ballanced leads.
Wow, what a professional and detailed explanation! Do you have detailed explanations about how choruses, phasers and other effects like this work?
Great suggestion!
Dude, Seriously dont stop I beg you
Keep it up👌👌👌
Thanks, Allan! I appreciate your support.
thank you alot, you explained it perfectly , greets from egypt
Best on YT. I wish I were this good when I teach my students tech stuff !
Thanks, rayrecordings!
To add to my last comment. This channel is absolutely effing amazing!!
An absolutely great educational video, its clarity and simplicity are among the best in the entire RUclips. Many thanks and bravo! A couple of questions: Comb filtering effect happens naturally, it is caused by a phase shift between two copies of the same signal. How will the frequency comb look like with three or more copies at different delay times? And are there any comb filters built to process signal and when are they used? Thank you!
Multiple comb filters occur in every room. It yields a bunch of combs on top of each other. To see a graph, look at an acoustic measurement of a room.
Comb filtering is used in many many ways. One example in audio is flanger. However, as a few others have stated in the comments, comb filtering is also a tool in programming and computing!
Not to take anything away from your very professional and useful presentation, but I was also hoping to hear all about phase shifters, flangers, and comb filters used in synthesizers.
what a goldmine of a channel :)
Thanks! Glad you’ve found it!
Loved this video, keep teaching more stuff, so informative. Thanks.
I’m very glad to hear that you’re enjoying these videos. Thanks!
nice work. I am sharing this resource with my volunteers!
You did an amazing job explaining in this video! Great examples and illustrations too. Thanks for creating such good content.
It’s worth it with comments like these, crawlkid. Thanks for watching!
I don't know why I am watching this but it seems I suddenly understand filters in digital signal processing , which is a big part of what I am studying.
Vey important concept explained in a very easy manner. Liked and subscribed
My listening environment is so bad I wonder if some aspects help me both the natural "First reflection points" are dead zones(out the dore one side and behind a raised fish tank the other.
My speakers are 6' apart but that is already starting to shroud them.
I am about 11' foot away on the perpendicular of the line between them, although I wish I could bring the sweet spot nearer me I fear I may never learn how to if it is even posible.
I am comforted to see that the comb effects happen increasingly often as the tones get too high to hear.
very interesting videos. All the best.
Note I normally complain I can't hear sound differences on you tube "Sound tests" but yours illustrated your points brilliantly.
I already know all this, just wanted to say that this is an excellent video on the topic!
Thanks, eraser! I appreciate you watching and leaving a comment!
This information is never going to be useful to me but I still enjoyed the video 😂
Glad you found it helpful, Jeff!
This is the best explanation video!
The sine sweep signals were often used in Atari video games as sound for different character movements or actions. For example, Pac man eating the pebbles or the ghosts. Very interesting.
My omnidirectional OHM G original style single Walsh driver speakers are very sensitive to this effect and need to be nearly 2 feet from the wall with drapery deadening to present a clear and accurate "sound stage" for the listener. The room and speaker placement (in 3 dimensions) of regular front fire speakers are less critical but still in play to accurately receive the intended sound that comes from the source media without room induced distortion.
Very straightforward thank you .
My question is if there is a delay can one simply drag the track back into place?
Iv heard people say yes and others say this can cause more problems. I’m not really sure what the answer is to that. I would think it’s fine but I’m not sure why someone would say it can be a bad idea .
It seems recording a drum set would be impossible tho with all the mics and no 3 to 1 ratio being implemented
Good question, @Hey There! You’re right with the drum kit - you can only align the sound of one component of the kit in each mic because a shift here will cause a shift there, and so on.
If the issue is a single instrument is delayed, such as the guitar amp example in this video, shifting the track could help. Just remember that there’s no guarantee that the indirect sound from reflections will be aligned on each track.
If there is an acoustic comb filter, there isn’t much that can be done, unfortunately.
Brilliantly explained Kyle thank you!
Could you please explain why it is better to use a cardioid microphone instead of a shotgun microphone for indoor recordings? However, in practice, shotguns such as the Sennheiser MKH 416 are always used for interviews and film recordings.
Unfortunately, I haven't been able to get a clear answer to this question from anyone.
Furthermore, it would be very interesting to have a detailed understanding of how the interference tube on a shotgun microphone works.
Great questions! Check out this video: ruclips.net/video/HagyNPzc-zs/видео.html
I'd be interested to hear your analysis on the practical impact of a recording when doing vocal doubling.
Good idea, Jim! Do you have any particular impacts in mind?
@@AudioUniversity I loved your analysis Kyle when comparing two identical signals that are out of phase with each other. So the dead spots where particular frequencies cancel each other out are mathematically predictable and you describe how to mitigate the effect. With vocal doubling the two signals are similar but with unpredictable differences. I know that vocal doubling is heralded as a way to enhance a vocal but it's clear from your analysis that the interaction can degrade the quality of the sound too. So I just wondered if there are ways to identify that and to mitigate against the effect. I'm totally naïve here but I could imagine cutting the second vocal track into suitable fragments and shifting each fragment back and forward in time w.r.t. the first vocal, listening for a sweet spot. That would take a lot of effort and it's a bit of a sledgehammer approach. Please educate me :-)
You're amazing, Kyle.
This is a very good knowledge. But I just wonder what it can cost if it's 20 or 14 or 45 or 8 ms delay.
Wow, good thing I subscribed to you guys Audio University. Truly I’m learning a lot from you guys...thanks thanks thanks...nice presentation guys...it’s a A+ from me...
I’m really glad to hear that, Paul! Thanks for your support!
Really amazing explanation!
Can you make a video about (the type of) distortion caused by audiocompression?
Brilliant explanation! Great work!
Great video thanks for posting it
I just wanted to clarify the relationship between the delay time and the lowest cancelled frequency as you may have made a slip of the tongue.
In your demonstrations (the first of which is at 9:15) you say that you’ve delayed the 2 sounds by 5 ms and that results in the lowest cancellation being at 1000Hz.
I think you’ve actually delayed the 2 signals by 0.5 ms (the time display shows 0.0005 seconds so that’s 0.5 ms) , in which case the lowest cancellation would be 1000Hz as you say.
However, a 5ms delay between two identical signals causes cancellations at 100Hz and all odd harmonics of 100Hz. This is because 5 ms equates to half the period of 100Hz and therefore a time shift of 5ms is equivalent to a phase shift of 180 degrees.
Thanks, George. I didn’t notice that mistake until now. I appreciate you clarifying that.
@@AudioUniversity A great video, but this error would have left me confused had I not come to read the comments! So my question with comb filtering involves hearing aids. One company has claimed their low latency of 0.5 ms fixes this problem of comb filtering caused by interference of world sound directly hitting the ear drum and the delayed signal through the hearing aid, but it looks like they've only kicked the problem up to higher frequencies that may not be an issue for many people but not for a musician. I notice it in my hearing aids with open ear tips as certain frequencies sound way off. Like the open G on my bass compared to the harmonic octave which will will sound like G#. With closed domes lowering the outside sound this seems to clear this up. I believe my hearing aids have a latency around 5 ms -> lowest cancellation at 100 Hz. Do you think this is indeed comb filtering I'm experiencing?
Great video. It's great to be able to see (and hear!) some examples in action. I'm wondering: In a technical book about acoustics I've read that the frequency of the first notch in the comb filter would occur where the period is twice the delay time, i.e. freq = 1/(2t), where t is the delay in seconds. In your example, for t = 5 msec, this expression would return freq = 100 Hz, but as you say in your video, your graph shows the first notch at about 1 kHz. Am I missing something?
You’re correct! I accidentally misspoke. That was a shift of 0.5 ms.
@@AudioUniversity Thanks for your reply, that clears things up!
Kyle: "Let me teach you how to avoid comb filtering."
Me: "Let me add a comb filter to my synthesizer patch and be in awe."
Great video Sir, insta-subbed.
Thanks, @Rayder Rich!
Great video and explanation, thank you!!!
I know comb filtering only form software synths like Massive, Serum or Vital xD
The explanation ist great btw
Thanks, @Leiocera!
One thing that I noticed, when using non-linear phase EQs, is that when I blend the wet signal with the dry signal, I can hear something similar to the comb filtering effect.
It’s possible! To be honest, I’m not sure. I think your logic makes sense though. Good thought!
While watching this video, an ad played of a guy in the car with comb filtering 😂
Fantastic explanation, thank you!
What an excellent explanation! Thanks man.
Thanks for watching!