I think I read somewhere, some time ago that some frequencies respond better to time correction (eg, treble and mid range) whereas others (eg, bass) tend to respond better to volume (read this as balance adjust, left vs right) when it comes to trying to set up a system with electronic time alignment/processing for good imaging and staging.

Anyone shed any light on this?
A link?
Your experience?

Cheers,
Winno.

Gee - i thought the balance cntrol was just considered to be the poor mans time alignment...

First of all, I could be wrong, so if anyone is in a position to correct me then please do so.

Now that I got that out of the way, I can respond to Winno's post.

What you're describing has something to do with the way our hearing works in relation to the frequency response. Specifically, it concerns two factors: IID (Interaural Intensity Difference) and ITD (Interaural Time Difference).

Basically, we perceive sound that come from a point in space either from the time it takes to reach us (ITD) or the intensity of the sound (IID). Sounds with frequencies that cover from sub-bass all the way to midrange can be considered ITD-type frequencies. While frequencies that fall in the treble region can be considered IID-type frequencies.

Case in point: If you have time correction feature on your HU, try increasing the "distance" of one of the midranges/midbasses while there's a solid vocal playing. Notice how the vocal shifts either closer to you or away from you. Do the same thing for your tweeters. Notice anything different? No, not exactly...(assuming that the balance is set to 0 or middle).

Now dial back your existing time correction setting, and play with your balance control only. Notice anything different? While the drivers on one side respond to an increase in loudness, pay attention to the tweeters.

Conclusion: playing with the balance control affect all drivers, but playing with time correction only affect drivers that are not tweeters.

My understanding of IID and ITD (and HRTF (Head Related Transfer Function) for that matter) are severely limited, so feel free to do more research on this subject. I too am interested, but I've yet to find time to do it what with all the assignments that are due in the next couple of days.

Cheers,

Bon

I've been looking for a particular link, it may have been what Winno was remembering, but I can't find it.

Anyhow, tomorrow I'll type up a basic post about the wonders of having 2 ears and different wavelengths of sound.

Oh boy, what a can of worms this has turned out to be.
I was surfing a little on and off at work today and ended up skimming medical articles and some pretty darn technical stuff to do with audio.

Thanks Bodapa for the info. You seem to be correct in what you say.
Alot of what I read was to do with pure tones but the audio stuff dealt in a small way with complex tones.
I'd thought that the lower octaves were intensity dependant but it appears the opposite with things changing between 1 and 4kHz, for simple tones at least.

I asked this because since reading the manual that came with my Clarion HX-D2 head unit, it's got me thinking about how the whole time alignment thing works.

Firstly, I will state that probably know less about IID and ITD than bodapa, so I may be wrong too. :wink:
This is a simplification, but should be easy for most people to understand.

Anyway, many animals on this earth have been given the wonderful gift of having a pair of ears. The advantage of having a pair of ears, is that it allows us to localize the direction that a particular sound is coming from (within practical limits).

Secondly, as we know, not all sounds are the same, they differ in amplitude and frequency.

Amplitude is basically how loud the sound is, measured with the logarithmic decibel scale.

Frequency is the amount of times the sound cycles every second. For example a 100hz (hertz) wave cycles 100 times per second.

There is an important relationship between the frequency and the wavelength of a particular sound.

Wavelength is basically the length between two corresponding parts (in the same phase) of consecutive cycles of a wave.

This is much easier to show with a diagram of a sine wave, but alas no, I don't have one right now. (search google if interested)...

The wavelength of a particular wave, is simply the speed of sound divided by the frequency.

Wavelength = Speed of sound / Frequency.

Ie, for 100hz,
Speed of sound = approximately 344.5 metres a second. (under average conditions, this value is accurate enough for this example)

Wavelength = 344.5 / 100
Wavelength = 3.445 m

So the Wavelength of 100hz is approximately, 3.4 metres.

As you may be suspecting, the wavelength has implications on how we can locate where a sound is coming from.

At very low frequencies, below 100hz for example the wavelengths are very long, so we have difficulty localizing the direction that the sound is coming from.

As the frequency gets higher, the wavelengths become shorter. Our ears will begin to notice that a sound that is from a source on our left will take longer to arrive at our right ear, compared to our left. Therefore we will be able to tell that the sound is coming from our left.

As the frequency gets even higher, the wavelengths start to become very short. Due to the distance between our ears in relation to the wavelength of the sound, our ears are no longer able to compare the time difference between the sound arriving at one ear, to the other. This starts to happen above somewhere around 1-2khz.

At these higher frequencies, because the sound cycles at a faster rate, it decays in volume more (compared to a lower frequency) over a particular distance. I'm sure we have all noticed that bass tends to carry a lot further than treble.
So at these higher frequencies, our ears will notice that the intensity of the sound is louder on one side, than the other. For example, if you had a tweeter playing on the right side of you, the ear on the right would percieve a slightly louder sound than the left, therefore you will notice that the sound is coming from the right.

Due to the location of our ears, we find it much easier to notice what direction the sound is coming from on a horizontal plane, rather than a vertical plane. Because of this, our ears are easier to fool in regards to stage height. Our ears are much more sensitive to higher frequencies in this case. This is why, with proper installation, tweeters in the a pillars (or at a similar) can give the the impression of a higher stage. I believe this is related to the HRTF as bodapa mentioned.

Of course, with all this in mind, one of the difficult parts is dealing with reflections, but that is another question altogether....

Didn't see Winno's post there. I guess this shows how long it took me to type this post up.......

No, no, thanks Bassaholic, it's been a great help and confirms what I read about azimuth not being so discernable whan it comes to stage height vs directionality across the stage.
I also learnt that the head acts like a low pass filter for the drivers fartherest from the opposite ear.

I was wondering what use I would have for the dual channel (left and right) equaliser in my head unit.

It's all quite fascinating!

Time is time. amplitude is amplitude. The correct protion of each is the go. SPL meter from Dick or jaycar is use to determine the volume of sound in the first initial setup. once you have complete your SPL aligiment then start the time alignment until you get the staging right.. If you are running bi-amp you may want to adjust the volume of each speaker individually.

Time might be time in your ears golf, but the time it takes the sound from the right speaker to reach your head and the time it takes the left speaker are 2 different times, hence path length is the key unless you could afford a McLaren F1 and sit in the centre or even move your drivers seat 2 inches to the left as you want to do in your car

I cant recall the exact freqs but below 900Hz is all volume dependent. ie the louder speaker sounds closer. Above that Midrange is time and volume dependent. ie arrival time and volume will determine which sounds closer. and High freqs its all time dependent. ie the closer speaker sounds closer.

Paul Graham explained this during gait training but I cant find my notes.

peace
Cyberpunky

Might have to get our training elsewhere now he's not doing it no more

Google found this

I came across a thesis that deals with HRTFs, IID and ITD. Fascinating subject! Just from skimming the thesis, I came to several conclusions:

1. The human body (from the top of the head to the tip of the toe) acts like a sound filter.
2. The binaural cues (ITD and IID) are important parameters to pinpoint sounds from the azimuthal plane (left right). But those two are not the only important cues.
3. ITD looks at the difference in arrival times of a sound's wavefront at the left and right ears.
4. IID looks at the difference in ampliture generated between the ears.
5. Sound is perceived to be closer to the ear at which the first wavefronts arrive. The larger the time arrival between the ears, the larger the ITD. This translates to larger lateral displacements.
6. And perceived lateral displacement is proportional to the phase difference of the received sound at the two ears.
7. But at 1.5 kHz, the wavelength is comparable to the diameter of the head, resulting in ambiguous ITD cues for azimuth. Not only that, but aliasing problems occur on 1.5 kHz and above, and phase differences no longer correspond to unique spatial locations.
8. Because of that, the ears begin looking for amplitude differences (IID) to localise sound in free space at 1.5 kHz and above.
9. So I was somewhat right in my first post on this topic! Yay! Normally I am shown to be wrong, so you'll excuse me if I'm jumping in joy here...
10. But since the duplex theory does not take into account how to localise sound on the vertical plane, then the theory needs to be augmented/perfected/polished/whatever.
11. Hence, HRTFs appear.
12. The thesis is too long to read, but I'll look at it in depth when I have the time.

So I think a friend of mine was right when he said that when tuning a system the presence of a body can affect the frequency response of a system. Assuming that the location of an RTA mic is made constant then the presence of 1 body would result in a different response compared to the presence of 2 bodies. And this doesn't even take into account the location (front, rear, driver, passenger side, both, all, etc.).

And another told me that 1/3 octave RTA is not sufficient enough, because when you see a flat or smooth curve on a 1/3 octave RTA you'd still find peaks and dips when using a 1/6, 1/12 or even 1/24 octave RTA.

And yet another told me that instead of having just 1 mic in aiding the tuning he would love to see a system where a mic array (or at least 2 mics) is used to get an average, complete response. If such an array is not available, the next best thing is to actually sweep the mic while taking a reading to get an average figure.

Question: has anyone gone to these kinds of extreme to just tune their systems?

Cheers,

Bon

Some links also mention the "Precedence Effect". Basically, two brief (very similar) sounds in succession will be heard as a single sound. This has particular implications in the case of reflections. If the reflected sound arrives within a short period of time of the first, the average location (between the source and the reflection) is heard. If the second sound is significantly louder (eg, over 15dB), then this will override the effect.

A more useful RTA system would take all of the variables into account. Two sensors is a must, taking time into account (ie by producing multiple response curves over small periods of time) is also neccesary (for taking reflections into account also), and a system to take the HRTF into account, the effect having the driver, passengers in the car etc.

In some ways it would be easier to just use your ears. :wink: