As a mixing engineer, you learn to use prime numbers in delays, to scatter and smooth out comb filtering effects. For example, 3 and 5ms delays, at the same time create golden ratios and sound very pleasing to the ear and much more smoothed out! Try a delay with 3ms left, and 5ms right! I've used this for simple stereo haus effects for years. Hope this helps.
@@lorenznew2208 You’re forgetting the source that is parallel at 0ms. Also you have to think about feedback with delays. So know you have 5/3 and becomes closer to a Golden Ratio. Don’t knock it till you’ve tried this.
I used a similar technique on a gig yesterday. For a dance performance, the client requested 4 side fills on stage Two monitors placed Upstage L&R and Downstage L&R. I was able to eliminate a great deal of phasing problems by simply throwing a 3-4 millisecond delay on just my upstage monitors. Worked great.
This is world changing. You sell yourself short sir! I once tried to just delay one of a pair of speakers to deal with this phase issue, but I lost the bass. Running a high pass and only delaying over 1K is so ingenius I am floored
That is a very promising approach. Now it would be interesting to see to what extent a similar result can be achieved with all-pass filters. This would simplify processing considerably.
К сожалению, у меня сейчас нет под руками ни процессора ни измерительного оборудования, но благодаря вам, я сейчас стал просто на голову круче, обязательно применю эти знания в работе. Спасибо вам огромное !!!!!
Cool! There is a slight flanger esque effect increasing in pitch as you move towards the left. Probably nothing that would cause an issue in a live situation though.
It's a subjective thing, but we thought it was less disturbing than the unprocessed flange sound crossing through the phantom center. It allows the artist to move side to side a bit in the area and not constantly be hearing the particular flange of the phantom center. Unless of course, they like that sound and know how to regulate their exposure to it, in which case we'll have a talk with them about which mode they would like to use, just like how we might talk with them about whether they want one or two wedges, etc.
How about two different pink noise sources going to each monitor? Dave Ratt has good videos on decorrelating sources to fix comb filtering effects. With a vocal mic in this situation this would not be possible but for something else it would be useful.
musician Bumblefoot talked a whole bunch about dealing with Phase while home-recording his solo albums Normal and Abnormal, hopefully the videos still exist but pretty cool stuff, you can hear in the final result how "heavy" the guitars sound without much gain being actually added to it, more like a "natural heavyness"
Is this something you discovered purely through trial and error? Or is there a deeper set of principles or information you're basing these changes on? I believe the explanation for the effect is simply that you're interrupting the classical comb filtering pattern by essentially ensuring that the constructive/destructive interference occurs more randomly across the spectrum and hence seems less noticeable. I wonder how this affects a singer's own perception of their voice? Since the singer hears another primary signal through their own head which resonates with the sound coming back from the monitors. This is why you don't want to invert the polarity of the monitors otherwise you may dampen a singers own perception of their voice. I want to know the underlying science here to see if there is a way to calculate a starting point rather than trial and error.
Interesting but how can you do this in a live situation with usual hardware ? Kinda hard to split signal, delaying one and send to the monitor with actual processing habilities with brand-locked presets in the processor and without fancy console routing…
Dumb questions.... 1) In this video, are the wedges daisy chained? 2) When you change the phase, are you referring to inverting the signal 180 degrees ??
2.- No, that would be a polarity reversal, he delays one signal so when both signals sums there is a phase shift that depends on frequency as can be shown in the graph.
dang! the info of your vids are sooooo educational.... btw, those speakers are used for monitors, meaning faced towards the performer? If so, would they still be needed if the performer uses an In Ear Monitor type of earbud/earpiece/headphone for monitoring purposes? Thanks in advance. God Bless and Have a Good Day Bro!
Right but where have you seen this done? It exists in various forms for sure, everyone does it a little differently of course. We’ve always wanted a processor that randomizes the phase of two signals without too much audible artifacts, but unfortunately have not discovered it yet if it exists!
@@devinlsheets_alphasound I don't pay close attention to how we process the signal for line arrays, but if I recall the standard technique is a different EQ on each channel, that together sums to zero, so the average signal isn't affected but the phase is. And at any given listening position, there's not going to be any dead spots because any given frequency will come predominantly from one channel. E.g. a comb filter on the left channel, and the opposite of that filter on the right. As I think about your solution more, putting a delay behind a high pass filter... isn't this the same thing as moving the high driver/horn physically closer to the audience (assuming everyone is listening from roughly the same direction), or putting a delay on only the horn?
Was there trial and error in determining the 2ms delay? Is there any correlation in the delay and the distance between the two monitors? 2ms is about 2.25 feet which is close to the distance I see between the two. Just an observation and question. Follow up, how do you thing this would work in a "wall" of monitors on stage (at ruffly the same spacing) in a line but having each one alternate between each normal and processes signal?
By definition, an allpass filter is any filter that changes phase but not magnite of the signal. So a 2ms delay is defintely a (very simple) allpass filter. What he has setup in the video is (ignoring that the little wiggle at 2k is only EQed out approximately, so the magnitude is changed slighlty) in fact an allpass filter. One could probably get something similar going using IIR allpass filters, the only difference being that you only get one 180 or 360 degree phase rotation with a single IIR allpass as opposed to the very very high number of phase rotations that the delay creates
Wow when showing the example I noticed the same phase issues with my home subs. I'm here trying to figure out how to deal with the 80% bass cancelation I'm getting. Not sure how to address this issue with my Denon 540. No matter how I place the 2 klipsch Synergy 10s I end up loosing bass with both vs 1. I've tried changing phases and it gets stronger locality but less everywhere else.
Not saying the distances were exactly the same, but also, the sun changes. The demonstrations happened about half an hour apart from each other because I did many takes of each. I did try to keep the distances about the same but certainly didn’t use any objective measurement tool, because I found that it didn’t matter too much if it was off by a few inches… the feedback performance was the same.
Devin, absolutely brilliant. question: I am setting up a playback system for my studio with one - dual 18" subs (20-100Hz) and two - rh 12" with conical horn tops (100-20kHz) per side. I think they are 60x40 horns (as opposed to 90x40), so limited dispersion. with the two full range cabs on each side, I am getting lots of lobing in the mids and highs (per side - just talking one side here). I am experimenting with pivoting the rh cabs toward each other a little (about 20 degrees maybe). I tried away from each other (about 20 degrees) and the lobing seemed worse. the horns are symmetrical and can be turned 90 degrees within the cabinet in order to change dispersion from wide to tall. I feel that pivoting the cabs toward each other about 20 degrees and turning the horns for taller dispersion somehow allows for less lobing and wider dispersion. this is just with pink noise. do you have any opinion? rh suggests these cabs can be set up in an array but I can't find any documentation (they are trapezoidal with lots of fly points). I am fine with a little lobing but would prefer to minimize it. I use a mic sometimes. so far the feedback is minimal, but is indicative of some lobing in the mids and highs. thanks! (sorry about the essay! hard to describe in few words).
So i have two flown speakers side by side maybe 1m … so one speaker will be unprocessd and one speaker will have 3,4ms delay on hi fq? And low fw will be untouched? Right?
Interesting. But unlike our TWO ears that hear phasing from the different distances our ears are positioned at, a microphone is hearing from ONE point. The microphone hears things differently phase wise. We may not like the phasing we perceive, but how does the microphone interpret phasing? Try blocking one ear as much as you can, but don't just turn your head 90 degrees to the speakers (Dual Wedges) and hear the difference with the same test with one ear because in this way you will be hearing what the microphone hears and remember, it's source to listen from isn't the speakers BUT what it's replicating. STILL... I get what you're trying to achieve, and getting the pattern as smooth as possible is a good achievement and there's no argument when pointing the microphone to the wedge getting it not to squeal. I have done this 60 degree with wedges for years hearing the separation via pink noise regarding phase cancelation and hot spots. I can actually hear more potential problems using the one ear method and learning what the microphone will highlight and thus, what to avoid taking into consideration the wedges dispersion specs too. Great clip all the same. A+ ;)
Really impressive! I could see this working in an theater. It would be cool to actually mix an actual ATMOS show. Roger Waters is doing another big tour, I’ve seen him three times but I’d pay double to hear it in ATMOS.
sepertinya ini sangat bermanfaat sekali,terutama pemula seperti saya,sayang saya tidak mengerti bahasa anda... apakah anda mempunyai laman website,agar saya bisa menterjemahkan dalam bahasa kami ?
As a mixing engineer, you learn to use prime numbers in delays, to scatter and smooth out comb filtering effects. For example, 3 and 5ms delays, at the same time create golden ratios and sound very pleasing to the ear and much more smoothed out! Try a delay with 3ms left, and 5ms right! I've used this for simple stereo haus effects for years. Hope this helps.
Do you have an audio example I could here?
This sounds amazing
So this is 2ms difference between your speakers...Just as he showed in the video. What's the point of having 3ms on the first speaker?
@@lorenznew2208 You’re forgetting the source that is parallel at 0ms. Also you have to think about feedback with delays. So know you have 5/3 and becomes closer to a Golden Ratio. Don’t knock it till you’ve tried this.
I used a similar technique on a gig yesterday. For a dance performance, the client requested 4 side fills on stage Two monitors placed Upstage L&R and Downstage L&R. I was able to eliminate a great deal of phasing problems by simply throwing a 3-4 millisecond delay on just my upstage monitors. Worked great.
This is world changing. You sell yourself short sir! I once tried to just delay one of a pair of speakers to deal with this phase issue, but I lost the bass. Running a high pass and only delaying over 1K is so ingenius I am floored
This has happened to me before with large bands on very small stages playing at moderate to loud levels. Thanks for the tip!
Great trick! Sounds much fuller when moving around. Thanks for sharing!!
That is a very promising approach. Now it would be interesting to see to what extent a similar result can be achieved with all-pass filters. This would simplify processing considerably.
К сожалению, у меня сейчас нет под руками ни процессора ни измерительного оборудования, но благодаря вам, я сейчас стал просто на голову круче, обязательно применю эти знания в работе. Спасибо вам огромное !!!!!
you can kinda get that same sound effect if you set static to your phone speaker, set it to full crank, and start approaching a wall.
Cool! There is a slight flanger esque effect increasing in pitch as you move towards the left. Probably nothing that would cause an issue in a live situation though.
It's a subjective thing, but we thought it was less disturbing than the unprocessed flange sound crossing through the phantom center. It allows the artist to move side to side a bit in the area and not constantly be hearing the particular flange of the phantom center. Unless of course, they like that sound and know how to regulate their exposure to it, in which case we'll have a talk with them about which mode they would like to use, just like how we might talk with them about whether they want one or two wedges, etc.
this is super interesting! How would this work if the speakers were facing the same direction and the "main" lobes were parallel?
Probably would be a similar result, you should try it and let us know!
Wow. This channel is cool as hell. Glad I found it
How about two different pink noise sources going to each monitor? Dave Ratt has good videos on decorrelating sources to fix comb filtering effects. With a vocal mic in this situation this would not be possible but for something else it would be useful.
You my friend is like me, always sunburned in the face! Nice video explaining wave sounds propagating
Cool concept and great results!
musician Bumblefoot talked a whole bunch about dealing with Phase while home-recording his solo albums Normal and Abnormal, hopefully the videos still exist but pretty cool stuff, you can hear in the final result how "heavy" the guitars sound without much gain being actually added to it, more like a "natural heavyness"
here's the video th-cam.com/video/vcMmhCD6qdc/w-d-xo.html
First thing most singers will do is turn those monitors toward them …
1:35 - 250 Hz, quite the phallus.
I knew someone had to mention hahaha
Have you used this in a live situation yet? If so, what was the response of the performer?
Nice work!
Is this something you discovered purely through trial and error? Or is there a deeper set of principles or information you're basing these changes on? I believe the explanation for the effect is simply that you're interrupting the classical comb filtering pattern by essentially ensuring that the constructive/destructive interference occurs more randomly across the spectrum and hence seems less noticeable. I wonder how this affects a singer's own perception of their voice? Since the singer hears another primary signal through their own head which resonates with the sound coming back from the monitors. This is why you don't want to invert the polarity of the monitors otherwise you may dampen a singers own perception of their voice. I want to know the underlying science here to see if there is a way to calculate a starting point rather than trial and error.
yes!
That was quite interesting. Thank you for sharing.
Interesting but how can you do this in a live situation with usual hardware ? Kinda hard to split signal, delaying one and send to the monitor with actual processing habilities with brand-locked presets in the processor and without fancy console routing…
Dumb questions.... 1) In this video, are the wedges daisy chained?
2) When you change the phase, are you referring to inverting the signal 180 degrees ??
2.- No, that would be a polarity reversal, he delays one signal so when both signals sums there is a phase shift that depends on frequency as can be shown in the graph.
dang! the info of your vids are sooooo educational.... btw, those speakers are used for monitors, meaning faced towards the performer? If so, would they still be needed if the performer uses an In Ear Monitor type of earbud/earpiece/headphone for monitoring purposes? Thanks in advance. God Bless and Have a Good Day Bro!
5:03 what is this software you used? thanks
How is this different from stereo decorrelation? I think they use this technique to avoid lobing on mains speakers too, especially line arrays.
Right but where have you seen this done? It exists in various forms for sure, everyone does it a little differently of course. We’ve always wanted a processor that randomizes the phase of two signals without too much audible artifacts, but unfortunately have not discovered it yet if it exists!
@@devinlsheets_alphasound I don't pay close attention to how we process the signal for line arrays, but if I recall the standard technique is a different EQ on each channel, that together sums to zero, so the average signal isn't affected but the phase is. And at any given listening position, there's not going to be any dead spots because any given frequency will come predominantly from one channel. E.g. a comb filter on the left channel, and the opposite of that filter on the right.
As I think about your solution more, putting a delay behind a high pass filter... isn't this the same thing as moving the high driver/horn physically closer to the audience (assuming everyone is listening from roughly the same direction), or putting a delay on only the horn?
I remember to do something like that once in my house trying to sing in a microphone in front of my monitors 🤣🤣 it worked 👌👌👌👌👌
Was there trial and error in determining the 2ms delay? Is there any correlation in the delay and the distance between the two monitors? 2ms is about 2.25 feet which is close to the distance I see between the two. Just an observation and question. Follow up, how do you thing this would work in a "wall" of monitors on stage (at ruffly the same spacing) in a line but having each one alternate between each normal and processes signal?
Brilliant
Good information thanks
how does this differ from an all-pass filter?
Same question here...
By definition, an allpass filter is any filter that changes phase but not magnite of the signal. So a 2ms delay is defintely a (very simple) allpass filter. What he has setup in the video is (ignoring that the little wiggle at 2k is only EQed out approximately, so the magnitude is changed slighlty) in fact an allpass filter. One could probably get something similar going using IIR allpass filters, the only difference being that you only get one 180 or 360 degree phase rotation with a single IIR allpass as opposed to the very very high number of phase rotations that the delay creates
Am I wrong, or is this just creating an all-pass filter?
great trick... 👍 ill say very useful in small places.
Was interested, but the unskipable adds was a deal breaker
I guess you don’t get to find out how to eliminate phase problems
@@devinlsheets_alphasound I know how to use a balance control. If you're out of phase the balance from left to right will let you know
in ears my friend….in ears !!!….amazing technique though, congrats!!
Dunno about you, I don’t want to hand in-ears out to the entire audience every time I have to work with live sound 🙄
Big help,..thank you sir
Wow when showing the example I noticed the same phase issues with my home subs. I'm here trying to figure out how to deal with the 80% bass cancelation I'm getting. Not sure how to address this issue with my Denon 540. No matter how I place the 2 klipsch Synergy 10s I end up loosing bass with both vs 1. I've tried changing phases and it gets stronger locality but less everywhere else.
try putting the two sub together with the woofers facing outward to the left and right instead of forward facing
What’s that?
Im not saying it doesnt work, but the distances were not the same. Check the shadows of the mic. all the best
Not saying the distances were exactly the same, but also, the sun changes. The demonstrations happened about half an hour apart from each other because I did many takes of each. I did try to keep the distances about the same but certainly didn’t use any objective measurement tool, because I found that it didn’t matter too much if it was off by a few inches… the feedback performance was the same.
@@devinlsheets_alphasound I think if you somehow measure the distance better, that woulb be nice.
that's great, actually
Interesting...😎👍👍✌✌
Devin, absolutely brilliant. question: I am setting up a playback system for my studio with one - dual 18" subs (20-100Hz) and two - rh 12" with conical horn tops (100-20kHz) per side. I think they are 60x40 horns (as opposed to 90x40), so limited dispersion. with the two full range cabs on each side, I am getting lots of lobing in the mids and highs (per side - just talking one side here). I am experimenting with pivoting the rh cabs toward each other a little (about 20 degrees maybe). I tried away from each other (about 20 degrees) and the lobing seemed worse. the horns are symmetrical and can be turned 90 degrees within the cabinet in order to change dispersion from wide to tall. I feel that pivoting the cabs toward each other about 20 degrees and turning the horns for taller dispersion somehow allows for less lobing and wider dispersion. this is just with pink noise. do you have any opinion? rh suggests these cabs can be set up in an array but I can't find any documentation (they are trapezoidal with lots of fly points). I am fine with a little lobing but would prefer to minimize it. I use a mic sometimes. so far the feedback is minimal, but is indicative of some lobing in the mids and highs. thanks! (sorry about the essay! hard to describe in few words).
Absolutely
Amazing 🔊🎤👍
So i have two flown speakers side by side maybe 1m … so one speaker will be unprocessd and one speaker will have 3,4ms delay on hi fq? And low fw will be untouched? Right?
that should be it as i understand it ;)
Alpha Sound, have you ever tried this on every other box within a line array hang?
We have not! Sounds intriguing
If a true line array is designed and deployed properly the boxes should couple and behave as one linear source and not have this problem.
Interesting. But unlike our TWO ears that hear phasing from the different distances our ears are positioned at, a microphone is hearing from ONE point. The microphone hears things differently phase wise. We may not like the phasing we perceive, but how does the microphone interpret phasing? Try blocking one ear as much as you can, but don't just turn your head 90 degrees to the speakers (Dual Wedges) and hear the difference with the same test with one ear because in this way you will be hearing what the microphone hears and remember, it's source to listen from isn't the speakers BUT what it's replicating. STILL... I get what you're trying to achieve, and getting the pattern as smooth as possible is a good achievement and there's no argument when pointing the microphone to the wedge getting it not to squeal. I have done this 60 degree with wedges for years hearing the separation via pink noise regarding phase cancelation and hot spots. I can actually hear more potential problems using the one ear method and learning what the microphone will highlight and thus, what to avoid taking into consideration the wedges dispersion specs too. Great clip all the same. A+ ;)
Would you suggest adding 2ms of delay to every other monitor in a venue like this one? : th-cam.com/video/MqK13L1SNmI/w-d-xo.html
Really impressive! I could see this working in an theater. It would be cool to actually mix an actual ATMOS show. Roger Waters is doing another big tour, I’ve seen him three times but I’d pay double to hear it in ATMOS.
th-cam.com/video/SoOIa0Vm6gQ/w-d-xo.html
Ok so you just made a diy all pass filter i guess...
Wild mic
sepertinya ini sangat bermanfaat sekali,terutama pemula seperti saya,sayang saya tidak mengerti bahasa anda...
apakah anda mempunyai laman website,agar saya bisa menterjemahkan dalam bahasa kami ?
250hz
Jajaja yo leí tricky phase nmms jajaj
You were father back Hmmmmm void test
Subtitle indo