Absolute Phase

[Mike Caffrey]

I'm curious about people's views on absolute phase within a mix.

A long time ago a friend explained the concept and I didn't expect to hear a difference, but when we flipped all of the drums simultaneously and kept the relative phase identical, there was definitely a difference. Not huge, but one way was definitely better.

A that point I used my kick as the reference and if the the kick needed to be opposite everything else, instead of flipping the kick, I'd flip everything else because I was sure my path from the kick all the way to the speakers was accurate.

Soon after someone pointed out that you never know if the listener has their set up right, yadda yadda, and said absolute phase really doesn't matter. So while I continued my practice, I thought that point made sense and was less concerned with absolute phase.

Then last night, I don't know why, but I flipped the bass which was recorded with a DI. There was nothing for it to have relative phase with. When I flipped it, it sounded pretty different and in this case definitely better.

I guess it's possibly that it's interacting with the kick differently, but I don't think that's the case. I'm pretty sure it's an absolute phase thing - although before this, I would have said I was sure that the path from the DI to the speakers was accurate - and I would have also predicted that I couldn't hear a difference with out the phase flip being relative to something.

So, I'm curious whether anyone here makes checking absolute phase a matter of course, whether you go beyond drums and bass, and if there's anything else that you do along these lines.

[JonCraig]

two sources, like the bass & kick you described can certainly have an impact on each other--especially when they're in a similar frequency band. after all... both signals are being summed (one way or another) into a single, stereo sound, correct? when i mix on a console (not too often, though) i sometimes try flipping polarity on things... seeing what works where. i don't think it's an "absolute phase" thing, 'cause the spaker should move equally both ways, right? if it wasn't, we'd be listening to distortion in our monitors all the time.

it's more about how signals interact with each other.

it's late for me... hope that made some sense.

--jon

[Ethan Winer]

Mike,

> I'm curious whether anyone here makes checking absolute phase a matter of course <

Here's the deal: Absolute polarity (not phase!) really is inaudible when the playback system is perfectly linear. And that's the catch. Most loudspeakers become less linear below a particular frequency, and that's when the driver responds differently to different excursion directions. Because this is mostly frequency dependant, you hear it mainly on low frequency instruments. But nonlinearity can affect higher frequencies too. I'm not a speaker expert so I can't tell you the root cause of the nonlinearity, or why it's different for outward versus inward. But I'm certain that this is why in some cases reversing the polarity can make an audible change.

--Ethan

[jbuntz]

Not to mention non-linearity of amplifiers.

[max cooper]

Usually if the absolute polarity of a playback system is reversed, the low end will have a sort of 'sucked in' sound.

[dirk]

Originally Posted by Ethan Winer

Mike,

> I'm curious whether anyone here makes checking absolute phase a matter of course <

Here's the deal: Absolute polarity (not phase!) really is inaudible when the playback system is perfectly linear.

--Ethan

Where can I buy a perfectly linear playback system?


Almost all sounds are "DC-shifted"
eg. you hit on drums; you bow a violin; you blow into a horn.
This causes "DC-shift".

And I think I can hear the difference easely. But I did not do ABX test.

[Berolzheimer]

Originally Posted by Ethan Winer

Mike,

Here's the deal: Absolute polarity (not phase!) really is inaudible when the playback system is perfectly linear.

--Ethan

I must respectfully disagree, Ethan. most waveforms in nature are asymetrical, and our ears/brains know the difference-and we're evolved to recognise sounds as they occur in nature. Listen to some dialog or vocal tracks buried in noise sometime, then reverse the polarity, and you'll probably find that it's more intelligible one way than the other.

[Nut]

Originally Posted by max cooper

Usually if the absolute polarity of a playback system is reversed, the low end will have a sort of 'sucked in' sound.

Only if you reverse the polarity of one of the speakers.

If you reverse the polarity of both speakers then everything coming out is still perfectly is phase with itself.

[Mike Caffrey]

Yeah, but that's releative phase. If you reverse the polaritiy of the speakers, the first excursion of the woofer responding to the attack of a kick drum with a mic placed in front of it will make the speaker "suck" away from you rather than push towards you.

There's definitely a difference in the tone between suck and push.


I'm going to have to re-read the initial responses which I'm better rested.


(I hate the term polarity, even if it is more accurate than phase - except just a couple of ocasions. My 1073s say "phase', so I'm going to continue to be wrong like they are. If you like the way it sounds... right?)

[Fletcher]

Originally Posted by Mike Caffrey

I guess it's possibly that it's interacting with the kick differently

It was absolutely interacting with the kik differently. You might have also noticed a difference in "solo", but that difference would not have have been nearly as pronounced as when you listened to it in reference to the kik.

The other question that springs to mind was whether the bass was laid at the same time as the drums or as an overdub. If the bass was laid as an overdub there is always a time difference from when the audio enters the unit and when it's commited to storage [I believe the minimum delay is like 1 sample... which isn't really enough to fukk with the music, but enough to create some relative phase related havoc which can sometimes be minimized with an "absolute polarity" reverse].

Interesting phenomenon... just wondering, do you have an "IBP" to play with, and if you do how did it change the bass sound in relation to the track? Just kinda curious.

[Bob Olhsson]

Originally Posted by Ethan Winer

se <

Here's the deal: Absolute polarity (not phase!) really is inaudible when the playback system is perfectly linear. And that's the catch...

According to Jim Johnston, Bell Labs research shows that there is a hearing mechanism that would respond differently to different polarity. The problem is that there are no transducers linear enough to test it. In his opinion it's a bigger stretch to assume you can't hear the difference than it is to assume that you can.

[Yannick]

This wav file clearly shows there is a difference.
Just don't play it too loud !

IMO the difference is in the monitoring system, because on our newer monitoring system, there is barely a difference, on the good old speakers it's BIG.

Non-linearity in positive/negative woofer excursion ?

[Elof]

Originally Posted by Yannick

This wav file clearly shows there is a difference.
Just don't play it too loud !

IMO the difference is in the monitoring system, because on our newer monitoring system, there is barely a difference, on the good old speakers it's BIG.

Non-linearity in positive/negative woofer excursion ?

I got very interested when reading this thread and have to try it myself.
I listened to the file in headphones and the latter half of it had a more "fat" sound. it wasn´t too obvious though.
IMO the first half was the "suck" and the latter was the "push", is that right?

[Ethan Winer]

Bob,

> In his opinion it's a bigger stretch to assume you can't hear the difference than it is to assume that you can. <

That makes sense, so let me clarify: It's not that I don't think absolute polarity can never be audible. Just that it's very subtle at best, and that speaker nonlinearity is the main reason you hear a difference.

There are a lot of things that can be heard but don't matter much. As an extreme example, changing the volume by half a dB is audible, but not damaging. And look at the very audible damage added by even the best analog tape recorders. In that case, a lot of people like the damage and are willing to pay handsomely for it.

Of all the things to fret about, absolute polarity is way WAY down the list, even if it is potentially audible.

--Ethan

[Biaseelfsu]

One can never go wrong, by simpily using their ears to obtain a sound in which they are looking for. We can get into the physics of phase and polarity, however, some of the greatest sounding albums have been recorded, utilizing little technicalities. Just a thought.