L/R Sample Delays - Perception Change

[DesolationAngel]

So, I know there's probably some very sound (sic) reasoning behind this but...

Let's say I record a acoustic guitar track (in Logic) and add a stereo sample delay, little bit longer one side than the other, it 'widen's the perception of the track... but why does it change the apparent volume level? The VU meters show it's playing back at exactly the same level in both L/R but the delayed side SEEMS to be quieter. If I pan to balance, perceptually, then the meters show an imbalance.

Just wondering...

[KevWind]

Good question. I do not know the answer. I have some guesses.
It could be simply a function of the way the algorithm is written.
But I would think it may probably be more of a psychoacoustic effect.

I have noticed in general for example if you take any stereo track delayed or not and pan it in so both the L and R are panned to the same position, it sounds louder. And conversely the further you spread them the more noticeable the perceived volume drop.

Some questions
First it sounds like your putting the delay on the audio track itself, as opposed to having it be a parallel effect ?
Have you tried using the delay parallel instead (on its' own track that you send the audio track to?) and see if the same thing happens ?



And also is this track a mono or stereo track to start with ?

Lastly if you like the effect but prefer the sound "panned to balanced. Then in that case just let your ears make the final decision.

[DesolationAngel]

So, turns out it's the Haas Effect... which is based on the 'precedence effect'. That is to say, whichever sound hits your ears first takes precedence and allows us to 'localize' the sound source... psychoacoustics indeed.

[philjs]

Originally Posted by DesolationAngel

So, turns out it's the Haas Effect... which is based on the 'precedence effect'. That is to say, whichever sound hits your ears first takes precedence and allows us to 'localize' the sound source... psychoacoustics indeed.

If you'd like to play around with the Haas effect a bit, try this Apex Audio plugin...it's cheap and quite useful.

Phil

[DesolationAngel]

Originally Posted by philjs

If you'd like to play around with the Haas effect a bit, try this Apex Audio plugin...it's cheap and quite useful.

Cool, thanks, I'll check it out.

[Fran Guidry]

Originally Posted by DesolationAngel

So, turns out it's the Haas Effect... which is based on the 'precedence effect'. That is to say, whichever sound hits your ears first takes precedence and allows us to 'localize' the sound source... psychoacoustics indeed.

It's really really hard to remember/process/accept that we don't hear with our ears. We hear with a whole system of transducers, filters, processors, and interpreters strung out between our ears and our brain.

It's also extremely valuable to understand that most of what goes on happens outside of any conscious control or affect. So knowing about the Haas effect doesn't make it possible to ignore the result.

With these ideas in mind, it's not disrespectful to be skeptical of what someone reports that they hear, it's simply realistic. And we should be equally skeptical of what we "hear" ourselves.

Fran

[DesolationAngel]

Originally Posted by Fran Guidry

It's really really hard to remember/process/accept that we don't hear with our ears. We hear with a whole system of transducers, filters, processors, and interpreters strung out between our ears and our brain.

It's also extremely valuable to understand that most of what goes on happens outside of any conscious control or affect. So knowing about the Haas effect doesn't make it possible to ignore the result.

With these ideas in mind, it's not disrespectful to be skeptical of what someone reports that they hear, it's simply realistic. And we should be equally skeptical of what we "hear" ourselves.

Fran

Not quite sure if you're agreeing or disagreeing

My issue was always with hearing the balance go out of whack when using sample delay. The meters suggested that the balance was right. But I could A/B it and hear it, every time. Both sides. Off/on. Always a change in perceived balance. Going to be reading up on this effect some more, that's for sure because it sure is intriguing.

[KevWind]

Originally Posted by Fran Guidry

It's really really hard to remember/process/accept that we don't hear with our ears. We hear with a whole system of transducers, filters, processors, and interpreters strung out between our ears and our brain.

It's also extremely valuable to understand that most of what goes on happens outside of any conscious control or affect. So knowing about the Haas effect doesn't make it possible to ignore the result.

With these ideas in mind, it's not disrespectful to be skeptical of what someone reports that they hear, it's simply realistic. And we should be equally skeptical of what we "hear" ourselves.

Fran

While more or less true you forgot arguably the most important element
It is also not disrespectful to suggest that in fact we as individuals do not hear exactly the same. That what is true, for what you hear and don't hear, may not actually apply to what someone else hears, real or imagined. So I would suggest that what we ourselves hear or do not hear, may not be completely relevant or accurate for comparison as to what others may hear.

And none of above serves to change the fact that in mixing when it comes to the stereo field placement, by far the best determining factor is to close ones eyes and hear what is going on.

That said, and if I remember correctly the Haas effect is about perceived localization not perceived loudness. As far as I know the only mention of loudness in Haas being relevant, is that the second sound can be louder (up to a certain amount of db's) and still not change the initial perceived location determined by the first sound. So unless I have missed some further information (entirely possible) it would seem Haas may not be sufficient to explain any perceived loudness difference of the second sound.

BUT given Haas there no reason not to pan to suit ones individual perception of "balanced loudness" because up to a certain point the localization will not change from the initial location.

[Fran Guidry]

Originally Posted by DesolationAngel

Not quite sure if you're agreeing or disagreeing

My issue was always with hearing the balance go out of whack when using sample delay. The meters suggested that the balance was right. But I could A/B it and hear it, every time. Both sides. Off/on. Always a change in perceived balance. Going to be reading up on this effect some more, that's for sure because it sure is intriguing.

I didn't realize there was a point to disagree or agree with. It's not like Haas effect is an opinion (grin).

Fran

[Fran Guidry]

Originally Posted by KevWind

While more or less true you forgot arguably the most important element
It is also not disrespectful to suggest that in fact we as individuals do not hear exactly the same. That what is true, for what you hear and don't hear, may not actually apply to what someone else hears, real or imagined. So I would suggest that what we ourselves hear or do not hear, may not be completely relevant or accurate for comparison as to what others may hear.

And none of above serves to change the fact that in mixing when it comes to the stereo field placement, by far the best determining factor is to close ones eyes and hear what is going on.

That said, and if I remember correctly the Haas effect is about perceived localization not perceived loudness. As far as I know the only mention of loudness in Haas being relevant, is that the second sound can be louder (up to a certain amount of db's) and still not change the initial perceived location determined by the first sound. So unless I have missed some further information (entirely possible) it would seem Haas may not be sufficient to explain any perceived loudness difference of the second sound.

BUT given Haas there no reason not to pan to suit ones individual perception of "balanced loudness" because up to a certain point the localization will not change from the initial location.

If you think about it for a moment, a single element can only move in location by changing perceived loudness from one side to the other. Which brings up another key point: loudness is not an objective measurable, it's a subjective response. We have tools that can approximate human loudness perception, but it's not something that can be measured precisely like power or volume.

The point of my rambling is that hearing is more psychoacoustic than most of us recognize, even those of us who try to keep psychoacoustics in mind when listening.

Fran

[Bob Womack]

Here's the upshot: Your hearing can be steered by level and by time. The Haas effect demonstrates that, levels being equal, the nearest iteration will be prioritized by the brain. This can be used in a mix if you want the guitar to seem to exist across the front of the ensemble but want it to be perceived on the left, put the original, direct sound on the left and a delayed sound on the right. The brain will hear it as a mostly left-located sound but it will fill things out and keep the levels equal.

Another application: Back in the day I was mixing a piece with a mono guitar lead over an ensemble. At one part in the song I wanted the lead to spread across the entire sound stage without being too loud. I learned to pan the guitar to the center and add a slightly delayed copy panned outboard to each side. The center-stage original took precedence but the two delayed copies filled out the sound stage nicely.

Bob

[Eire]

Originally Posted by Bob Womack

Here's the upshot: Your hearing can be steered by level and by time. The Haas effect demonstrates that, levels being equal, the nearest iteration will be prioritized by the brain. This can be used in a mix if you want the guitar to seem to exist across the front of the ensemble but want it to be perceived on the left, put the original, direct sound on the left and a delayed sound on the right. The brain will hear it as a mostly left-located sound but it will fill things out and keep the levels equal.

Another application: Back in the day I was mixing a piece with a mono guitar lead over an ensemble. At one part in the song I wanted the lead to spread across the entire sound stage without being too loud. I learned to pan the guitar to the center and add a slightly delayed copy panned outboard to each side. The center-stage original took precedence but the two delayed copies filled out the sound stage nicely.

Bob

Always appreciate the practical application you bring to this site, Bob. Thanks! I've tinkered with this but never understood fully 'why' it did what it did. Much appreciated.

[KevWind]

Originally Posted by Bob Womack

Another application: Back in the day I was mixing a piece with a mono guitar lead over an ensemble. At one part in the song I wanted the lead to spread across the entire sound stage without being too loud. I learned to pan the guitar to the center and add a slightly delayed copy panned outboard to each side. The center-stage original took precedence but the two delayed copies filled out the sound stage nicely.

Bob

Actually I have seen this used and tried this technique for lead vocal as well.