Delay Always Means Phase Shift But Phase Shift Does Not Always Mean Delay 

This video left me with more questions than answers

Sometimes I wish I could subscribe even harder.

"Full Positive Excursion" - that's a band name and a half, if anyones looking.

Again, spoken like a true audio ENGINEER... I wish more people would care about the science and function of things as much as they care about what they "feel" or "believe"....

I'm interested to hear more about how EQ effects phase shift, as well as MB compression. Please do more videos on this subject. And thanks for sharing these very useful insights.

Dan you are mad underrated, one of the only engineers out there doing real content explaining real fundamentals from a very technical yet understandable perspective. In RU-vid, there are a lot of content creators trying to teach, but no many of them owns their shit the way you do. Thank you very much for all the knowledge you help to spread, you are awesome!

With so many incorrect videos on the topic of phase and polarity, this video is the way to go for anyone who is not interested in understanding phase via complex numbers (although that is where phase comes from). It shows that when you focus on a single frequency, the phase can be the same across three operations, while the underlying operations are different. So you can get a phase shift of 180 degrees from 1) a polarity flip (180 degrees at ALL frequencies), 2) A time delay (180 degrees at a few discrete frequencies only), and 3) an allpass filter (180 degrees at some frequency). But they do differ in general, as transient behaviors are different, and their group delays are not the same, although their phase delays at that particular frequency are the same, since the phases are the same. For group delay, the operations need to be viewed across the entire frequency range, and so zooming in on a single frequency only tells us part of the picture.

Best techy-audio explanations on RU-vid! Thank you, Dan!

Nothing like starting your day with audio knowledge from Dan. Thanks man!

Totally revealing information. Thanks! But don't stop here...keep going! ☺️

Love your videos! Thank you for another excellent upload!

Great video. Looking forward to the next EQ one.

Can't wrap my head around to how to mix/blend real + sampled kick tracks when EQing 2 tracks independently before summing to a kick bus (and then add more processing) I made sure both tracks were visually time and phase aligned. Then I was forced to downmix the two tracks to save some CPU, and started going crazy. In the resulting waveform, both kicks had moved slightly in time, creating a strange and longer waveform. I had made sure that the recorded kick was time/phase aligned with the overheads (did the same with the rest of the drum tracks. In fact, I've become obsessed with this lately), and now everything was all over the place because of individual eq processing. I am 100% sure it is not a latency compensation issue. When mixing down/freezing tracks, my daw compensates for plugin latency automatically. So I decided to mixdown each track individually to see what was going on. Zero phase eq processing created a 3x longer kick on each track (because of different frequency phase missalignment, I guess), and linear phase created lots of pre-ringing and missing transients. The less destructive setting was linear phase in "minimum" setting, but still had issues. All of this considering I was using a sample kick sound that was nothing like the recorded one, except for adjusting the pitch to match the fundamental note. I got rid of HPFs and LSFs I was using to polish low end, and things started to improve a little. Then applied those filters back in the kick bus afterwards, and things kind of started working again. But now I'm not confident anymore about the kick staying phase aligned with the rest of the drum tracks (specially OHs) unless I bounce the bus output to audio and compare the timing to the overheads. And that's a bummer. If I've understood correctly how this works, despite of visually aligning the recorded audio tracks and its transients, heavy eq processing can missalign back the timing of frequencies separately, and the only way to make sure everything's ok, you have to use your ear, or bounce the individual/bus tracks to audio after processing to re-check the alignment before summing in a drum bus. My final thoughts: the best you record things and the later you process drum tracks after summing, the better. There's only so much correction and tweaking you can do before things start to get all oven the place and out of control. I really struggle to get things right when I receive badly recorded tracks for working on a mix. I feel like I spend 90% fixing the impossible before I can even enjoy the process.

Thank you! This was incredibly helpful.

You know it's getting real when the music stops in a Dan Worrall video :)

Well, this one of the very informative and educating channels, i am glad to have found it!

Still feel like a beginner, when I see content like this! Thank you, Master!

That was brilliant!

This is the most educational channel out there

This stuff is above my paygrade.

I agree with you actually more or less. But I feel like to make this argument you should really have tried to null against a slightly different filter: Mix a band-stop filtered version with a delayed and band-passed filtered version. This is what people are arguing the filter essentially does, so do it. This would give a clearer picture of what is happening. As it stands the primary reason your example doesn't null is that a patched delay phase shifts EVERY frequency, even the infinite ones cause by the jump discontinuity, whereas the all pass filter you showed only phase shifts a finite frequency window (shown in the graph by the size of the phase response bulge). Of course that isn't going to null.

Happy new year to you too!

Dan, can you please make a "nuts and bolts" video, the way only you can, on exactly how equalizers work and how they cause phase shift and resonance? I'm sure I'm not alone in my desire to understand on a deeper level how EQ works so that I can not only avoid the unintended consequences of equalization, but maybe even use it to my advantage the way that you seem to be able to do, and I just can't seem to find that much information on the subject. My basic understanding is that what is actually going on is that it is the mixing of delayed (phase shifted) signal with the original that causes changes in frequency response, but I don't understand how, from that basic principal, all the different filter shapes are possible, how a filter's bandwidth can be manipulated, or how resonance fits into the equation. I'm just now becoming acquainted with the idea that equalizers are tools to affect the envelope of a frequency as much as its volume, and I want to understand more. Thank you.

Thank you !!

Thank you!

Fascinating. Always learning something new, thank you Dan.

Brilliant!

I'm sorry, Dan, I have to disagree with you on this. The phase is described by the formula φ= ω៰t. Where the "t" is literally a delay. So phase ALWAYS means delay. And by the way, "invert phase" (so popular in DAW) does not make sense technically for this reason. If there is no delay, there is no phase. And in this case there is no delay. It is technically correct to call it "invert polarity". But from a mathematical point of view, there is no difference between A sin(ωt + 180°) = -A sin(ωt). Where the first is a phase reversal and the second is a polarity reversal. Because the signal is considered infinite in these calculations. It never starts and never ends. And what you say about the all-pass filter and the difference in sound - concerns the transition process and is not directly related to the phase at all. In fact, it's just a filter ringing, its response to an impulse. You even mark it yourself later, but somehow you don't relate it to the difference in the beginning. And although I'm sure you understand all this perfectly, the way you explained it can still mislead the audience.

I really need to watch all your videos

thanx for this

Happy new year again

These videos are so good

Dan: Happy New Year 2020: not so much

Thanks.

3:23 “this is why mixing delayed signals with the original results in comb filtering.” Very well explained. In my head I always imagined different slopes partially cancelling each other out but your phrasing is simplified while being more comprehensible. Thank you for explaining that. Now why is a short delay the basis for karplus-strong? What is it about a short delay that is physically similar to a vibrating string?

You've just explained why Analog/Modular Synths sound so much better than Digital...

You will get a better intuitive understanding of what is happening to the signal in time domain if you look at group delay graph instead of phase. In the typical loudspeaker crossover design software, group delay is one of the core graphs one can look at. Basically the delay is local to the range of frequencies affected by filters and not the whole signal. It's like stretching a rubber band locally between two anchor points. Talking about analog style minimum phase filters here, not phase linear FIR processing.

A great service. Used to be we learned the basics on this kind of stuff at a desk, then had it shown to us by the head engineer we studied under...long before we ever touched a tape machine. The kids today have ZERO background, so your stuff is PRICELESS...and us old guys get something too, as you go seriously in depth...intelligently.

Wow, super interesting! Can you make a video about getting higher peak values through phase shifts? 🤔

I wish Crane Song would hire you to make tutorials about Phoenix II, or to compare it with other saturation plugins. The thing is too subtle for me to evaluate if it is worth it to open Pro Tools just to use Phoenix II, or if I should just stick to Saturn on Reaper.

Phase shift is the filter when you want to kill false engineers

verry cool.

Great educational video again Dan. One quick question, how do you manage to get the extra options in the right click render menu? I can't render a selected area of a track like in your menu.

Hello, I was wondering when we turn a radio frequency, what makes the phase get tune in along with the resonant frequency.

thanks for this! phase shifting has always confused me a bit, especially because some effects have a continuous phase shift setting in degrees. That wouldn't make sense without that phase shift being specific to one frequency, isn't it? Otherwise it would only be possible to have 0 and 180 deg.

Tremendo, me dejó pensando

Top voice in the game, I want this dude to narrate my death

The best

Great video!! So what about the phase response of bell filters? Is that just a less pronounced version of this?

Support this channel!

You can do full inversion with no delay, but how does one shift phase a different angle without delay? And how is the angle determined?

SUBBED

Hi Dan :) I remember seen your videos on phase issues when cutting with non linear phase EQs, which were of great help as we regularly ran into problems while parallel processing. 🙏🏻 But now this: today we put a correlation meter in front and behind an EQ8 and used a simple low cut - no parallel processing or anything. A plain and simple lowcut. And surprise, the meter showed that the correlation dropped significantly. Then instead of EQ8 we used ProQ3 in linear phase mode. And still: the correlationmeter would display a significant drop in correlation 🤷 We have no clue other than we found glitch in the matrix... Any idea?

I still couldn't understand the phase shift that happens from an EQ because by logic, let's say we do the same phase shift at 100 Hz but on full song track; if the 100 Hz is on 180° and is 100% null, then what comes after it at both sides should reduce the volume of the frequencies according to the slopes. Well, it doesn't sound like this, or maybe the phase shift graph does not correlate with the amplitude? Because by this logic, this should make a wide surgical EQ cut which should be very noticeable cancellation, can you explain this please?

The Professor of Plugins

you deserve more views. but you're gonna have to start naming your videos something like " why the bass sucks in your mix!"

謝謝！

Do EQ cuts sound worse because the part of the signal that's phase shifted is "lowered" relative to the rest of the signal during a cut, whereas in a boost the part of the signal that's phase shifted is higher/louder so you can hear the phase distortion more?

thanks for video about this black magic 😂

thank you for this great video! what's the software you are using sir?

This may seem like a strange question but I have someone that wants me to recreate some old software for him and he claims that they panned these square waves from the left earphone to the right using phase shifting....is that possible? If so how does that work? I used a regular audio component to pan the sound from one ear to the other but according to him it doesn't give the same result...to me it sounds exactly the same....

2:50 isn't this only true if you're using very basic waveforms where each cycle is identical? if you move a kick drum ahead 5ms, you're not getting an inverted kick drum

Is there any way to avoid or controlling those peaks caused by phase shift? I really don’t wanna use just a rude clipper.

Dan, can you please help me figure out how to monitor the total phase curve, when using a minimum phase eq in a parallel track? i noticed you used an older VST Plugin Analyzer in a previous clip, is that the best tool for this? i have Plugin doctor at the moment, but i cant figure out how to view the total.

Will you share the vocoder track that's playing as the outro? 🙏

So I came into the video guessing I'd learn something about phase shift. All I know now is there are multiple types of phase shift caused by multiple things and I know none of them...fk...what does phase ringing mean? etc etc etc

Dan should be President one day. I would vote for him

I admire you god damn

Cutting EQ makes for a Louder mix? not that I want a "louder mix" - More depth perhaps ;-)

Dirac spikes are interesting to use for analysing filters, since they contain all frequencies (within the Nyquist limit, of course). I was amazed at how much ringing occurs in EQs, especially high-pass filters where it can seriously affect the sound of bass frequencies. Kicks and bass sounds with sharp impulses can be affected quite dramatically. Far too many people use sharp filter slopes in dance music nowadays, when all it does is cause ringing effects that aren't usually audible to most people without trained ears. Suppose it doesn't matter as much, but when you know what it sounds like, it really hurts to hear it.

Hi Dan, I've been researching the topic quite a bit and am still not perfectly clear about the origins. Most threads end up explaining how phase shifted signals cancel each other, but don't get to the point why the phase shift initially happens. I've heard people saying, it's due to latency issues. But when I interpret the phase shift curves in the eq correctly, it seems more like certain bands around the point of application get shifted. Can you maybe explain how that physically happens?

What's the name of the music in the beginning and in thе end of the video?

Thanks for another great explanation, watched an older one you made on this topic too, but I still don't understand why eq causes phase shift? I can see that it does, but why?

I'm waiting for the hour long video to outright explain what a digital filter and/or EQ does for real. All those side notes ("thats what truly happens anyway" sort of stuff) only makes us curious to study the engineering

Hi there thank you for the video. I got doubt with one of my productions I've just picked up a piano and listen it in SOLO without any processing it's a Full DRY signal and I noticed that in some octaves the phase polarity tend to goes to negative -1 and I don't hear any deleting informations in the stereo field, in fact when I play the full song my phase meter tells me that the song goes from 0 to positive +1 and it's ok, just in solo MODE when I hear just the melody part I got this doubt that I don't understand in the very long tail of the notes goes to negative, thanks if you will answer me :D

Please do more videos like this on EQ phase shift and how EQs work. How do I interpret the phase shift by just looking at the EQ. Lots of questions still. Please teach me more.

It seems like phase shifting can produce a crude addressable led strip.

🤯

great video as always dan ! you can pls answer me how to identify phase problems in mono with a correlation meter? I use span and more, and even if there are sounds that sound horrible together (phasing) and they are (mono) the span recognizes them as 1+ (good relations)

Polarity and phase are not the same thing. A sawtooth wave shows this clearly. The shapes of waves play a part in their tone.

When you think you finally got it... think again

"That sudden spike is a short burst of all frequencies" - Can someone explain why this is to me? ^^;

the unsung hero (sic!) of the video is, I believe , the music track in the background! Am I the only one to notice ? It grooves a lot ! Needless to say that the research you have done and explain to us is unique in all of your videos! Simply essentials and go to's at any time

Food for your brain. :)

I actually watched this DESPITE the Comic Sans thumbnail.

Wow

Where is that LCR fix video?

I couldn't understand ...

Figured I'd fall asleep to Dans' soothing voice, but the topic is too damn interesting..

How come phase shift makes my peaks levels higher!!! 😮 I don't need sleep I need answers!!!! 😱😱

Doesn't the processing of different frequencies take the same amount of time delay, no matter if you're boosting or cutting, implying that the opposite boost or cut would still cancel itself out seeing as the amount of delay would be the same in both cases? Meaning your comment about seeing into the future makes no sense?

Duuuuude hahaha xD

Very informative video Dan. It's also interesting (not mentioned here), different audio editors, although dealing with same material, process them all differently, because of their coded (programmed) structure... just like your dither video showed. Hence, basic ones will create more profound (and interesting) anomalies, to your ear... whereas, advanced ones, might automatically hide it.

I am still confused

he sounds almost exactly like mads mikkelsen its so damn uncanny

I understand the meanings of each word you said. But when put together, I have no clue what you are talking about.

Off topic question: Do you preffer filters or shelves to clean tracks? And Why?

And this is how the universe was created.

Oof the music ruins it...

I skipped ahead to this one.