[greenh0rn]

In our quest for higher and higher SQ in our car audio, naturally we are interested in knowing how linear is the Frequency Response of our much-loved setup (HU, amps, speakers) at the final listening stage. One of the tried-and-tested ways to check the FR is to use an RTA. Now, I was searching through this forum and found precious little details on people actually having used an RTA in their car. Hence a few questions that popped up in my head, that I ask of all of you (gurus and SQ enthusiasts):

1. Have you used an RTA to check FR in your car (I know Navin used his old RTA in his ANHC and ibm_jennifer perhaps used one in his Safari)? If yes, what were your initial findings with the flat EQ? To put my question differently, how linear or off-linear did you find the FR of your car audio setup?

2. Please post brief details of your car, set up and a brief assessment of your RTA testing.

3. How many high-end audio installers in the country have an RTA, and more importantly, how many buyers demand to know the FR of their set up? If the number is low (which I guess it is), why is it so?

4. Finally, does anyone in Delhi/NCR has an RTA lying around (that he doesn't mind lending to fellow T-Bhpian(s) for testing )?

Cheers!

[panky12345]

Originally Posted by greenh0rn

4. Finally, does anyone in Delhi/NCR has an RTA lying around (that he doesn't mind lending to fellow T-Bhpian(s) for testing )?

Cheers!

The answer is ... Yes! You got it right.. Mr. LBM freedabad wale!

[greenh0rn]

LBM has it?? Sure?? Then I must plan another pilgrimage to Freedabad one of these coming weekends!

[greenhorn]

Imo Frequency response within the car varies so much spatially that I really dont think there is much of a point to it.
If i had to do it at gunpoint, I'd get an alpine HU and imprint it

[navin]

Originally Posted by greenh0rn

1. Have you used an RTA to check FR in your car (I know Navin used his old RTA in his ANHC and ibm_jennifer perhaps used one in his Safari)? If yes, what were your initial findings with the flat EQ? To put my question differently, how linear or off-linear did you find the FR of your car audio setup?

2. Please post brief details of your car, set up and a brief assessment of your RTA testing.

1. a response that is flat on the RTA might not be most pleasing to the ear.
2. all OEM car HUs have horribly un-flat response curves. The first tiem I saw the curve it reminded me of the curves I got from a Bose AM5 system. A lump here, a hump there, a valley between and a roll off of everything above 10kHz.

My object for using an RTA was to see the OEM system's response. If I was putting in a new HU and speakers etc.. I'd just tune by ear. After all it is what is pleseant to the ear that is most important.

[Bass&Trouble]

Originally Posted by greenh0rn

3. How many high-end audio installers in the country have an RTA, and more importantly, how many buyers demand to know the FR of their set up? If the number is low (which I guess it is), why is it so?

A few installers do. Very handy only when the system has the ability to be corrected based on the RTA findings. Otherwise there isn't much point.

[Invinsible]

RTA can help in determining the freq. response in car. However it also depends on the equipement one is using. There are various PC based software available. True advantage of RTA can come in use if Graphic or digital equaliser is available with atleast 30 band for every individual channel or driver being used or atleast a left & right individual EQ. This can help in boost or cut on a particular freq. If there are peaks or dip in particular freq/band to achieve a flatter response. Also, imp is to understand what a driver response is like. Once you achieve the flatter, linear response through out the freq band, one can futher tune to achieve the desired result.
Every car, every driver, mounting location will have varied result. Not many invest in sound processors or equalisers or going through this process as it's time consuming. And some land up convincing themselves that their system is not a big hi-end, which would make it worth to go through such a process or investment.
But as Navin said some may like it, some may not.

you would be better off correcting the frequency response to the equal loudness contours (fletcher munson curves) so you apply some kind of weighting to the curve instead of aiming for totally flat response on the RTA curves

explanation of equal loudness contours

Equal_Loudness_Contours

so if you use "true RTA" (freeware application program) as an example on your laptop with a good microphone there is a corrected "true flat" response curve provided based on equal loudness correction which is more useful than properly measured flat RTA curves

but its far easier to tune using your ear anyway, you need very good reference material though for example you need to listen to an unamplified musical instrument and compare this to a recording of the same as for example can be found on discs like this



this disc has for example loads of recordings of individual instruments and after listening to this you need to listen to that same instrument played live for example drums or an acoustic guitar - and you tune till the sound of the recording is similar to the real sound, if you measure the response curve at that point it should be close to an equal loudness contour instead of flat

though going to all of this trouble is only worth it in some form of competitive environment ie if you enter sound off competitions or want a truly accurate car system, if you just use the system to listen to music generally then it is far simpler to just tune to your preference after all you paid for the sound so you deserve to have it set according to your preference

[greenh0rn]

Originally Posted by navin

1. a response that is flat on the RTA might not be most pleasing to the ear.

Is it because our ears grow accustomed to non-linear response? Something like Indians growing up on butter chicken/tadka dal/sambhar rice and finding British food as bland?

I actually wanted to just gain an academic understanding of what type of FR will be yielded by a typical car setup comprising of carefully selected components. Your description of an OEM setup answers my question in a respect, but I'd definitely like to hear if LBM/others have used RTA and what was their experience/assessment.

Originally Posted by Bass&Trouble

Very handy only when the system has the ability to be corrected based on the RTA findings.

Yeah, I agree- a detailed parametric eq feature in HU or a dedicated processor is a must have before fiddling with an RTA.

Originally Posted by Invinsible

True advantage of RTA can come in use if Graphic or digital equaliser is available with atleast 30 band for every individual channel or driver being used or atleast a left & right individual EQ.

I was thinking to use RTA to test the FR from the precise listening position (headrest of driver's seat, passenger seat etc.). Totally agree with you and Navin that FR will probably look different by moving this test position a few inches up or down, left or right, forward or backward.

[greenhorn]

greenh0rn Sr. i think its because the world is inherently full of environments with varying frequency responses that we have grown accustomed to it. if we could easily detect a +/- 3 db peak or dip in frequency response, We probably couldnt walk across a room while listening to music without being bothered by a phaser like effect it would drive us - at least me mad. What would it be like if the colours and hue of things in the world changed as you walked by - like what happens in an LCD screen ?

Just like our ears naturally compensate and adjust for dynamic range, they probably do so for frequency response as well. This is based on my limited understanding of psycho acoustics. I could be wrong

I was thinking to use RTA to test the FR from the precise listening position (headrest of driver's seat, passenger seat etc.). Totally agree with you and Navin that FR will probably look different by moving this test position a few inches up or down, left or right, forward or backward.

the problem if you do this is that positioning the microphone at one point does not take into account the Interaural time differences and Interaural level differences that exist between each ear on either side of the head. Technically your head gets in the way of the right ear hearing the left speaker properly and vice versa - so if you wish to do this properly you have to account and adjust for this ... how to do this i have no idea

but it is one of the problems in why setting a flat frequency response according to an RTA still does not sound acoustically realistic even tonally speaking

[greenh0rn]

Originally Posted by naughty001

but its far easier to tune using your ear anyway, you need very good reference material though for example you need to listen to an unamplified musical instrument and compare this to a recording of the same as for example can be found on discs like this

I have this disc with me

Originally Posted by naughty001

though going to all of this trouble is only worth it in some form of competitive environment ie if you enter sound off competitions or want a truly accurate car system, if you just use the system to listen to music generally then it is far simpler to just tune to your preference after all you paid for the sound so you deserve to have it set according to your preference

I agree with you somewhat. See my idea is not so much to tune my audio so good that I only hear the exact sound that the artists produced at the time of recording (anyways my current equipment does not justify that level of expectation), but just an academic interest in seeing how good is my current set up in terms of producing a linear response when pink noise is played.

Originally Posted by naughty001

the problem if you do this is that positioning the microphone at one point does not take into account the Interaural time differences and Interaural level differences that exist between each ear on either side of the head. Technically your head gets in the way of the right ear hearing the left speaker properly and vice versa - so if you wish to do this properly you have to account and adjust for this ... how to do this i have no idea

No no I didn't intend to make it 'that' complicated- this whole RTA business is just for fun and experimentation in my scheme of things! But the trigger of course is a lingering feeling that some instruments don't sound the way they are meant to, on my system.

[greenh0rn]

Ok without further ado, gentleman, I introduce SynRTA.

New Page 1

Now, which microphone do you recommend I use with this lovely piece of software?

[greenh0rn]

I am an obsessive researcher Let's see what folks elsewhere on the Internet are saying:

Fixing the resonances and the holes and the peaks is fairly straightforward, but finding them isn't so easy.

For instance, I added my sub to my system on Sunday. I didn't measure the sub, I just plugged it in. Immediately, it was obvious that there was a hole in the response somewhere.

If this was 1999, I would've sat there in my car, obsessing about the hole in the response, tweaking crossover points on my electronic crossovers, listening, maybe adding a bit of EQ, listening some more, etc...

Since it's 2009, I didn't do that. I grabbed my microphone, measured the response, and looked at it. There was a big hole at 150hz. Then I lowered the gain on the sub amp, dialed in some EQ to fill in the hole, and measured again.

Boom, done, took ten minutes. I fixed the hole in the response in less time than it takes to get coffee before work. (I *did* tweak it a bit by ear after the fact, I don't live and die by what the computers says BTW. Measure first, listen second.)

Now obviously, you can EQ by ear, but it's IMPOSSIBLE to get it perfect. Mics have a sensitivity that exceeds our own ears. When you EQ by ear, you undershoot or overshoot all the time, and wind up with response variations on the order of 3db.

A 3db hump isn't massively offensive, but when you're chasing down perfection, you can't have errors of that order.

From this outstanding thread Improve Your Sound Quality for $80 - DIY Mobile Audio

[navin]

Originally Posted by greenh0rn

I am an obsessive researcher Let's see what folks elsewhere on the Internet are saying:...
Now obviously, you can EQ by ear, but it's IMPOSSIBLE to get it perfect. Mics have a sensitivity that exceeds our own ears. When you EQ by ear, you undershoot or overshoot all the time, and wind up with response variations on the order of 3db....
From this outstanding thread Improve Your Sound Quality for $80 - DIY Mobile Audio

1. EQing by ear will never be perfect but is a visual representation of perfection what you desire. I am a simpler man. I am quite happy with an aural approximation of perfection. When I used an RTA to EQ my home system I found that the system sounded too light and the bass was a bit too dry. By ear I must have got a 1db hump in the 40Hz region and a very gradual roll off above 15k (Given my are and exposure, I am sure my hearing above 15k barely exists anyway). This must also be a factor of the music I tend to prefer to listen to.

2. Those who are used to an RTA swear by it. Those who do not have access to one regularly still get by quite well without it. RTA can be a useful tool for sure but it is not essential. In some ways and under some circumstances it may only complicate things.