Help in tuning 4100nex

[mattalter]

I have a new install 4100nex.  While it sounds ok, I feel I can get better sound out of it.  I have tried the auto tuning EQ with the mic.  While I get some results, not sure if there that great.

 

Any suggestions on what settings to take a look at beside the eq?

 

txs

M.

[ScottinPollock]

I have a new install 4100nex.  While it sounds ok, I feel I can get better sound out of it.  I have tried the auto tuning EQ with the mic.  While I get some results, not sure if there that great.

 

Any suggestions on what settings to take a look at beside the eq?

 

txs

M.

 

While I am a big fan of what Pioneer's MCACC did for my home theater system, It got just about everything wrong in the car (which is not all that surprising given the location of most in-car speakers).

 

Suggest turning the auto EQ and TA off. 

 

First set up the TA manually using a tape measure. Then start with the crossover at 100Hz and 18dB/oct. Play something with a good bass line and heavy kick and turn it up. Raise the subwoofer level so it is a bit on the heavy side. Toggle the phase button in the subwoofer settings (you'll have the correct phase when it sounds fuller rather than thinner).

 

Now it's time to set your speaker levels. I have all speakers set to 0 except for the RF which is -2. But I don't usually turn on the rears so you may need to balance the front and rear if you use them. Then adjust the sub so it is in balance (you may need to listen to a variety of program material to get this dialed in). 

 

Then listen to the various preset EQ settings to see which one you like best, and use that as a starting point for your custom EQ settings. Note that if you desire a lot of boost in the bass, consider upping the level on your subwoofer and dialing back on the mid-bass using the EQ settings.

 

After you get it semi-dialed in, you might find that adjusting crossover points and slopes will help you get smoother response. For me, I tend to want the crossover point as high as possible (with steeper slopes), as it puts more work on the more powerful subwoofer amp, freeing up head room on the head unit's amps.

[jhren]

If you are going to be setting up by ear, this link may prove useful...

 

http://www.audiocheck.net/audiofrequencysignalgenerator_sweep.php

 

Many calibration methods use pink noise, but without a spectrum analyzer, pink noise is of little use... and those methods use a microphone calibrated to a standard.  Ultimately, the true test is what you hear.  If your hearing is deficient at certain tones, no matter how flat a calibration says your system is, you will not perceive it as flat.

 

Using swept sine tones, you'll be able to distinguish what frequencies your drivers are capable of reproducing efficiently and with clarity (and not, of course), which will permit you to tailor your crossover frequencies... and perhaps use rolloff slopes to enhance flat reproduction.  For example, typical setup procedures will have you set sub and HF rolloff slopes to the same rate.  However, if your subs have a peak response just below the crossover frequency, you can use a gentler slope to attenuate that peak.

 

What you are shooting for is sweeping through the audio spectrum (20~20kHz) so all frequencies "seem" to be the same volume to you.  You'll likely experience some rolloff at the extremes... depends on your system and hearing, so don't overcompensate at the extremes.  Adjust as best you can with levels, crossover frequencies, and rolloff slopes before moving into frequency equalizer adjustments... in that order.  General rule of thumb is no more than +/- 3dB adjustment from adjacent bands.

[ScottinPollock]

What you are shooting for is sweeping through the audio spectrum (20~20kHz) so all frequencies "seem" to be the same volume to you.

 

You do realize that even individuals with perfect hearing will interpret loudness relative to frequency along something akin to a bell curve, and that this curve changes drastically depending on reference level. In a nutshell, it is impossible to do what you suggest and be anywhere in the ball park without an accurate SPL meter and calibrated capsule.

 

Combine that with the severe off axis response heard due to the placement of in-car speakers, and the numerous near field reflections and standing waves present in a small cabin, and you won't even be close to a bad guess. Even the square-wave based time align measurements the auto TA measurement returned were comical due to off axis measurement coupled with severe near field reflections. This means that not even time gated systems like UREI's Sonipulse would be of any value.

 

Unless you have rear facing front speakers, and front facing rear speakers, along with a reasonable acoustic environment, I doubt any measurement will get you better results than working with a handful of reference recordings that you know the sound of, and adjusting this by ear and a little common sense.

[jhren]

I'm not going to debate the issues with you.  I am technically adept with those issues, so I do understand the jargon.  However, the only place those terms have any use is in an acoustic chamber in a audio laboratory.  So unless you have a full array of laboratory grade analyzers, there is no point in even considering those issues.

 

If you understand C-weighted response, you would know that [upside-down] "bell curve" is actually what is desired for flat response... in sound energy.  We are not talking signal level here.  Yes the exact curve varies with reference level... but unless you have an accurate implementation of "loudness" control on the output signal, you have no means to compensate when volume is adjusted by user.  AFAIK, only the Pioneer audio-only head units have "loudness" control feature.  Without "loudness" control, the system needs to be set up at nominal everyday listening level... not at the system maximum output level.

 

And even though you have off-axis rendering, near-field reflections, standing waves, etc. in a vehicle cabin, they all combine together to what the human ear perceives as sound.  Just because direct wave is desired does not mean you can adjust for a direct wave flat response and expect to perceive the overall sound as flat.  Now if you want to go through the entire paradigm of sound treatment in a vehicle cabin to reduce indirect sound... have at it and please, by all means. report back how you achieved direct only response.

 

Anyway, I agree, the end test is listening to familiar recordings adjusting for best reproduction.  I suppose I should have mentioned that... but assumed readers would realize this is always the ultimate test.

[ScottinPollock]

the only place those terms have any use is in an acoustic chamber in a audio laboratory.

Not at all. Anechoic chamber measurements serve to take the real world out of the equation; to measure what the device is capable of, sans any effects introduced by boundaries. In fact, you'd find it pretty disturbing to listen to anything in one.

 

If you understand C-weighted response, you would know that [upside-down] "bell curve" is actually what is desired for flat response...

No. Flat response is flat response. The 'C' weighted measurements, or more accurately, the 'A' weighted measurements (the one based on Fletcher/Munson curve used in the 'loudness control' you mention below) were designed to make up for the difference in the perceived loudness of sine waves versus the full spectrum of noise the ear and brain evaluate.

 

unless you have an accurate implementation of "loudness" control on the output signal, you have no means to compensate when volume is adjusted by user.  AFAIK, only the Pioneer audio-only head units have "loudness" control feature.  Without "loudness" control, the system needs to be set up at nominal everyday listening level... not at the system maximum output level.

Again, flat response is flat response. While you could introduce a loudness contour for any set level, that would only mean all other levels would be wrong as I have yet to see a 'loudness' setting on any gear that actually track level and adjust the curve according to the ears' diminished sensitivity to deep bass and high highs at lower levels. More often than not these contours result in booming bass and piercing highs. If you like it that way, fine (I hear plenty of vehicles driving by with that kind of response curve), but it ain't accurate. And my hearing deficiencies may be different then the sampling done by Fletcher and Munson 50 years ago; does that mean I need to account for those as well? None of these curves are applied when listening to live music, so what is the point? Get it right first. If you then want more bass and treble at lower levels, use tone controls or EQ.

 

And even though you have off-axis rendering, near-field reflections, standing waves, etc. in a vehicle cabin, they all combine together to what the human ear perceives as sound.  Just because direct wave is desired does not mean you can adjust for a direct wave flat response and expect to perceive the overall sound as flat.  Now if you want to go through the entire paradigm of sound treatment in a vehicle cabin to reduce indirect sound... have at it and please, by all means. report back how you achieved direct only response.

Perhaps I should have been more concise in my original response to your suggestion to the OP that he listen to sine wave sweeps to adjust parameters based on perceived relative loudness. It won't work, and he'd be wasting his time.

 

Anyway, I agree, the end test is listening to familiar recordings adjusting for best reproduction.  I suppose I should have mentioned that... but assumed readers would realize this is always the ultimate test.

Well, we do agree on something. (c;

 

Edit: To be clear, flat, accurate response is not something that is really possible with a conventional in-car setup. There are just too many compromises with regard to driver placement, and the small cabin with its many obstructions. The manual TA setup is critical to clarity, but the EQ and driver levels are just something you will have to balance based on recordings you know. This process can take days or even weeks to dial in. But I will reiterate that the auto-EQ/TA provided with the unit is not something you want to start with.

[jhren]

That's nonsense. The purpose of anechoic chamber measurements is to take the real world out of the equation; to measure what the device is capable of, sans any effects introduced by boundaries. In fact, you'd find it pretty disturbing to listen to anything in one.

I didn’t say anechoic chamber… but I can see how you would make the leap to that given what I did say.

What I said amounts to, even if you have the means to discern “coloration†from direct, it will be for naught unless you have a means to control or otherwise influence the “colorationâ€â€¦ and I know it will require more than any listening test you can devise.

No. Flat response is flat response. The 'C' weighted measurements, or more accurately, the 'A' weighted measurements (the one based on Fletcher/Munson curve used in the 'loudness control' you mention below) were designed to make up for the difference in the perceived loudness of sine waves versus the full spectrum of noise the ear and brain evaluate.

You are correct on the weighting scale. My bad. I haven’t discussed weighting scales for a while now, and my last occurrence was regarding occupational noise levels and hearing protection. Many of the meters used to determine occupational noise level are C-weighted.

As far as curves are concerned, I was more so referring to the Robinson–Dadson model, as revised per ISO 226: 2003.

I have not heard or read anything that states the curve is different for sine waves than for noise (e.g. pink moise). Got a link to an online source which confirms that assertion?

Again, flat response is flat response. While you could introduce a loudness contour for any set level, that would only mean all other levels would be wrong as I have yet to see a 'loudness' setting on any gear that actually track level and adjust the curve according to the ears' diminished sensitivity to deep bass and high highs at lower levels. More often than not these contours result in booming bass and piercing highs. If you like it that way, fine (I hear plenty of vehicles driving by with that kind of response curve), but it ain't accurate. And my hearing deficiencies may be different then the sampling done by Fletcher and Munson 50 years ago; does that mean I need to account for those as well? None of these curves are applied when listening to live music, so what is the point? Get it right first. If you then want more bass and treble at lower levels, use tone controls or EQ.

The point is you/we are not listening to live music. Your paradigm is fine if recorded music were not EQ’d, compressed, ducked, etc. The processed electronic signal is far from flat as the original audio energy wave was. Believe me… I used to mix live sound and do multi-channel recordings for demo’s for several bands in my area. (I don’t do it anymore because my “day job†leaves me unavailable too often.)

The concept of loudness control represents the ideal. Implementation typically falls short, but some are better than others. Knowing how to set up a system for optimized contour is key in most scenarios, but the manufacturers that include a contouring feature never provide details for proper set up. But you are correct in that some to many implementations will just be outright intolerable to a refined ear.

And your assessment of the cars driving by may be inaccurate. The quality, if set up for such, will be inside the car. What you hear on the outside can be a world apart. But I do agree many are boomy… but I believe that mostly stems from people liking it that way… and not having any clue what accurate reproduction is all about.

Perhaps I should have been more concise in my original response to your suggestion to the OP that he listen to sine wave sweeps to adjust parameters based on perceived relative loudness. It won't work, and he'd be wasting his time.

To each his own… but that seems a bit contradictory with many system setup app’s using a combination of swept sine wave and pink noise RTA measurements.

FWIW, I have also been one-step short of using active acoustic correction on my home theater setup. The thought crossed my mind before I actually had it setup up without. Sounded so good without I dropped the thought. Perhaps one day I will follow through...

In the meantime, nothing you have stated changes my position…

[ScottinPollock]

I have not heard or read anything that states the curve is different for sine waves than for noise (e.g. pink moise). Got a link to an online source which confirms that assertion?

"Researches in loudness measurement". doi:10.1109/TAU.1966.1161864; but I can not find the entire document online unless you want to pay for it.

 

The point is you/we are not listening to live music. Your paradigm is fine if recorded music were not EQ’d, compressed, ducked, etc. The processed electronic signal is far from flat as the original audio energy wave was.

Regardless of what signal processing is used by the engineer, it is mastered to be heard as intended, without any weighting applied. I have recorded everything from commercial pop/r&b, to live, audiophile direct-to-disk and direct-to-digital, and the end result was always what the producer and I felt it should sound like given the monitoring gear available (which was usually pretty good).

 

that seems a bit contradictory with many system setup app’s using a combination of swept sine wave and pink noise RTA measurements.

Not contradictory at all since those systems are not using the human ear to discern levels as you suggest the OP should.

[jhren]

"Researches in loudness measurement". doi:10.1109/TAU.1966.1161864; but I can not find the entire document online unless you want to pay for it.

Not going to pay for it.  While it'd be nice to be in the know, it doesn't mean enough to me to shell out money in exchange for the knowledge.

 

 

 

Regardless of what signal processing is used by the engineer, it is mastered to be heard as intended, without any weighting applied. I have recorded everything from commercial pop/r&b, to live, audiophile direct-to-disk and direct-to-digital, and the end result was always what the producer and I felt it should sound like given the monitoring gear available (which was usually pretty good).

And that right there is half the problem with today's conventional recordings.  The true audiophile wants recordings to sound as if they are there ... live.  I don't want to listen to music how you want me to perceive it.  I want to listen to music how I want to perceive it.

 

 

 

Not contradictory at all since those systems are not using the human ear to discern levels as you suggest the OP should.

But at the same time, you want me to hear your recordings same as your ear discerns levels.

 

Ultimately, I offered a tuning suggestion.  The OP can use it or not.  If he does use it, and the end result sounds like crap, he's going to reject the method and resort to tuning by ear anyway.  If on the other hand, it gives him clues as to how to best tune his system, it will have accomplished my goal for making the suggestion.

 

"C'est la vie"

[ScottinPollock]

And that right there is half the problem with today's conventional recordings.  The true audiophile wants recordings to sound as if they are there ... live.  I don't want to listen to music how you want me to perceive it.  I want to listen to music how I want to perceive it.

You don't have that luxury. Yes, you can EQ the crap out of it, but that doesn't make it better; just colored to your particular palette, and in the case of radical change, adds a number of other issues that just make things worse. In the case of live recordings, you're not there, so you have to rely on the producers and engineers that are there to faithfully record the experience. Unfortunately, we all have great music that sounds as though it was recorded through a couple of tin cans and string. To me, engineering is just as much an art as the performance itself. Like you, I have a lot of great music that I don't often listen to because it sounds lousy. And dealing with instruments that have no live sound is a whole 'nother part of that art as you must visualize that performance as opposed to getting it sounding like 'live'. I'd like to think I have a good ear, and have received a grammy nomination that re-enforces that assessment.

 

But at the same time, you want me to hear your recordings same as your ear discerns levels.

C'mon! That argument is non-sequitur. Trying to EQ by auditory interpretation of a sine wave sweep has nothing to do with the mics I choose, the position they are placed, and the relative levels assigned to them in the mix, nor the setup of the monitoring system.

[jhren]

You don't have that luxury. Yes, you can EQ the crap out of it, but that doesn't make it better; just colored to your particular palette, and in the case of radical change, adds a number of other issues that just make things worse. In the case of live recordings, you're not there, so you have to rely on the producers and engineers that are there to faithfully record the experience. Unfortunately, we all have great music that sounds as though it was recorded through a couple of tin cans and string. To me, engineering is just as much an art as the performance itself. Like you, I have a lot of great music that I don't often listen to because it sounds lousy. And dealing with instruments that have no live sound is a whole 'nother part of that art as you must visualize that performance as opposed to getting it sounding like 'live'. I'd like to think I have a good ear, and have received a grammy nomination that re-enforces that assessment.

 

Congrats on the nomination.  I'd ask if that was a direct nomination or by involvement, but it really doesn't matter to me.  Kudos all the same.  Overall you appear to have a good disposition on the matter... but please do not try to pass off those publicity buzz phrases on me as it just weakens your position. E.g. "faithfully record the experience"...  I'm not saying you do not, but every recording and/or mix engineer, past, present, and future will make the same claim no matter how bad their "performance".

 

 

C'mon! That argument is non-sequitur. Trying to EQ by auditory interpretation of a sine wave sweep has nothing to do with the mics I choose, the position they are placed, and the relative levels assigned to them in the mix, nor the setup of the monitoring system.

Ahh... but it does follow.  It is just another tool in the toolbox.

 

You choose a mic and its placement for the resulting capture characteristics.  Believe it or not, that is coloring the sound to your particular taste.  If all mic's were created equal, there'd really be no difference in choosing one over another.

 

My ear cannot be in two places at once, but yet listening to a recording made with two or more mics mixed into one channel is putting my ear at two or more locations per se.  How is that not coloring the sound?

 

You assign levels in the mix so as to mimic an a la natural or pleasing mix... but in essence, you are coloring the music.

 

It all has to do with perception...

 

Don't get me wrong.  I understand the mix and recording process substantially more than the average reader here... and I'm not faulting that process over a stereo mic trying to capture a one off live performance worthy of the hall of fame.

 

Anyway, I'm willing to bet using swept sine wave one will find system "weak" points faster than any other method.  With your adamant stance, I have to wonder if you've ever tried it.

[ScottinPollock]

I have to wonder if you've ever tried it.

I've used an oscillator for making some speaker measurements and mapping impedance curves, but using an SPL meter, not my ears.

[jhren]

 

.... I'd like to think I have a good ear, and have received a grammy nomination that re-enforces that assessment. ...

I've used an oscillator for making some speaker measurements and mapping impedance curves, but using an SPL meter, not my ears.

 

I understand using a meter if you have to make a numeric analysis.  But other than a numeric analysis, does that meter tell you anything about how it sounds?  So your ear is not as good as you claim?

[ScottinPollock]

I understand using a meter if you have to make a numeric analysis.  But other than a numeric analysis, does that meter tell you anything about how it sounds?  So your ear is not as good as you claim?

The meter was used for lab measurements in the course of loudspeaker design. Again, I don't "listen" to sine waves to determine how any part of a system "sounds".

[jhren]

Okay...

 

So I believe we've each made our points.  Time to let things happen as they may.