Sound Localization with the Haas Effect

Last week I wrote about the multi-dimensionality of a stereo mix. While that article focused on distinguishing localization from depth, I'll be specifically focusing on localization this week.

As a reminder from the last post, localization is often referred to as "pan," and it describes the placement of a sound in the left to right stereo spectrum. However, the pan position is not the only way to place a sound to the left or right of the listening position.

Another technique for achieving this effect involves the use of short delay times, sometimes referred to as the "Haas Effect." It is so-named after the person who is credited with discovering the technique, but I'll leave that bit of history for the reader to discover on his/her own.

The effect is achieved by duplicating a mono sound and panning each copy to the extreme left and right sides of the spectrum, respectively. Next, you add a delay to either the right or left copy so that one is triggered a little later than the other. The delay amount used is very short, between 1 and 35 milliseconds. You can also avoid duplicating the mono sound if you have a stereo delay on the track, in which case you can choose to delay only one side of stereo spectrum independently from the other. The combined effect is that the sound's perceived localization changes depending on the delay amount chosen.

Before I continue this technical discussion, you may be asking why anyone would want to venture beyond simple panning to achieve left to right spread of sounds. The reason is that panning isn't always ideal, and indeed there are times where none of the pan positions will render an acceptable distribution of instruments in the mix. This is the time when the Haas effect really comes in handy. Since the technique relies on creating small phase changes, we're doing more than just turning one side of the stereo spectrum up and the other one down. Of course, this also comes with the danger of creating phase-related problems when the piece of music is reduced to mono, so be sure to check your mix in mono as you work just to be sure that it still sounds acceptable.

Localizing a sound using this method can create all kinds of interesting effects depending on a few factors, the most obvious being the delay time that you've chosen on one side of the stereo spectrum. Beyond this, you can alter the volume of the delayed copy, but be sure that you don't increase it any more than 10 db louder than the dry copy because this can destroy the illusion of changing the sound's overall position. One last easy trick is to add a low-pass filter to the delayed side, carefully carving off some of the high frequencies. This will serve to downplay the delayed copy, biasing the perceived position toward the dry copy.

I've kept this discussion brief in order to whet your appetite for experimentation. Don't take my word for it, give this stuff a try and let us know what tips and tricks you come up with in the comments section!

Sound Localization with the Haas Effect by

Tags: , , , ,

14 Responses to “Sound Localization with the Haas Effect”

  1. James MNovember 18, 2009 at 11:16 am #

    Interesting stuff. I’ll have to give this a try. I was told once before about having two monos, one extreme left and extreme right, and then adding a little chorus to one of the sides. I’m not sure if this is still the same thing? (Or I was told/understood wrong!)

    So you’re saying it’s a case of actually delaying the output of one of the channels by those milliseconds, not actually using a delay/echo effect?

    • adminNovember 18, 2009 at 1:43 pm #

      Hi James, the technique you mentioned is certainly a variant of the one that I outlined in the post but not quite the same thing. A chorus takes multiple versions of a timbre, changing each one a little bit from the other and layering them to create a shimmering effect. Since there is phase manipulation going on here, it’s definitely a technique that you could use for localization. However, I think the Haas effect is a little more predictable since you’re not at the mercy of a chorus algorithm but rather an easily controllable delay amount.

      As for your second question, the two things you said are the same thing. You could simply add a few millisecond delay to one copy through your DAW, or you could put a stereo delay on a mono sound and change the delay time of only one side by a few milliseconds. The effect should be identical unless I’m forgetting something (and that’s been known to happen).

  2. lateralNovember 18, 2009 at 5:16 pm #

    Guys Thanks for posts, I have a question.

    If you are working with stereo wav saples, how can you test the sound as MONO?

    Simply pannig everything to one side?

    I ask this because I’ve heard that is good to check a Stereo Mix Using Mono. But I’m Not even sure how do that.

    Thanks

    • adminNovember 18, 2009 at 5:48 pm #

      @lateral The easiest way I can think of is to use some sort of plugin on the master output. In Ableton Live, for instance, there is a plugin called “Utility” that allows you to set all incoming audio to mono by reducing the “width” parameter all the way.

  3. lateralNovember 19, 2009 at 5:20 am #

    thank you.

    Panning everything one side may be a solution?

  4. adminNovember 19, 2009 at 1:38 pm #

    @lateral, you’re welcome! Re: panning everything to one side, I suppose this might work for the most part. However, there are different “panning laws” that determine how panning sounds in different DAW’s, not to mention issues of panning sounds that are stereo rather than mono. I would still go with the plugin solution as your best bet.

  5. lateralNovember 19, 2009 at 5:39 pm #

    @everybody:
    the pleasure is all mine

    keep making great passionate-posts!
    and thank you for advices

  6. Mark TNovember 20, 2009 at 3:30 pm #

    Hi Nick,

    Great post and interesting. Thanks for sharing.

    -m

  7. chuck SilvaNovember 21, 2009 at 1:51 pm #

    Hey Nick.

    Indeed these posts are great!

    I like your no nonsense approach.

    I’d like to see a video tutorial on mixing techniques/EQing especially getting a great clear low end that dosen’t crowd the kick/bass?

    Thank you!

  8. guitar speedDecember 11, 2009 at 1:19 pm #

    Nice. I was familiar with just the panning technique, but I had no clue that there was another way to do it. When I’m feeling ambitious I’ll reread this and see if I can’t get it to work. Thanks for your contribution.

  9. PieterApril 17, 2012 at 9:21 am #

    Hi Nick!

    I was wondering what kind of stereo effect is used on this kick:
    http://soundcloud.com/vvv/beautiful-life-vvv-remix

    I’ve been trying to make sounds like that, with a great sound of depth, but been unable to do so…

    Thanks for your time!

    • NickApril 17, 2012 at 2:01 pm #

      Hi Pieter,

      That’s a really nice, lush track. Just FYI: The traditional advice is to keep kicks and bass sounds as mono as possible (straight down the middle of the mix) for a number of reasons such as compatibility with cutting to vinyl and club sound systems. If neither of these are likely to be issues for you (they never are for me), go for it ­čÖé

      There are all kinds of techniques for achieving a wide kick sound, so here are 2 hints that you can mix and match until you get something that sounds nice:

      1) Use two identical kick sounds, both panned the same amount to the left and right, respectively. Lower the gain on both kicks until they meld into a nice overall sound that comes across as one gigantic, wide kick. For extra fun times, try subtly detuning them from one another, or try messing with their phase interactions by shifting one of the copies a few ms. forward or backward from the other kick on the timeline. Be careful with this technique, because even a ms. or 2 can make the difference between a really cool, wide sound that sits well in the mix and sonic mud.

      2) Since you are clearly going for a very atmospheric kick sound, don’t be afraid to experiment with reverb on one or both kicks. Start with the same settings on both reverb units, but then try differentiating their decay times from one another.

      There are plenty of other ways to go about this, and most traditional mixing engineers would tell you to steer well clear of these techniques. I’m not one of those guys. That being said, be sure to read up on phase cancellation so you’re aware of the downsides!

      • PieterApril 17, 2012 at 3:35 pm #

        Great advice, i ‘ll try to experiment with this as soon as i can. Very nice to see someone being open-minded and thinking out-of-the-box about traditional settings of certain musical elements. That’s what makes music for me so exciting.

        cheers,

        Pieter

      • PieterAugust 1, 2012 at 12:40 pm #

        Hi Nick!

        I tried the techniques you mentioned, and indeed that is some golden advice! It also made me think differently about sounds and the phase of sounds and how some fx make use of phase shifts. For example chorus and flanger are such techniques right?

        I was wondering, are you familiar with Vesco FX for example this free one:
        http://www.vescofx.com/vfxFreeHaas
        ?

        Do you think it can replace the “manual” Haas trick?

Leave a Reply