I would like to add a recommendation here. I set my trusty Radio Schmuck meter (analog version) on boxes or something similar on the chair in my prime listening position, and get the microphone as close as I can to the position where my head would be. I then get myself off to the side of the room somewhere where I'm not in the direct line of fire of any speaker to the meter, and accoustically I should be "out of the way", but where I can easily see the meter. I then run the test tones (with the remote, of course) and adjust from there. I have found this method to be more accurate than holding the meter while sitting in the chair. I have had a meter for quite a while, and have found it to be well worth the investment. The first time I used it I discovered that previously by ear I had all of my speakers balanced within 1 dB, except the rears were 1 dB louder than what the meter liked. I assumed this was because the meter is truely omnidirectional, and our ears are not. Our ears are pointed forward. So when I balanced by ear, I had set the rear speakers 1 dB louder to compensate for that fact. I decided than since I usually listen with my ears instead of the meter, I would leave the rear speakers set 1 dB louder. I have since run some tests with some friends of mine, and have discovered that indeed most of them cannot balance speakers by ear nearly as accurately as by using a meter. Why I can I am not sure, my hearing is just unususally good in some respects, but it's still nice to have the meter for verification. I actually checked one system where when the owners set it up by ear they had the center channel 8dB too low and the surrounds 10dB too high. Needless to say, after readjusting, they were amazed.

That Rat Schack meter is not as efficient at subwoofer frequencies, and therefore registers a lower meter reading than what the sub is actually putting out. SV Subs has a chart showing the offsets at
various frequencies, and proper sub calibration procedures:

http://www.svsubwoofers.com/faq.htm#meter

This is a mistake you hopefully only make once... (Bit me on my pre-950 setup)

Make double sure that none of your sources are active while you're performing calibration. If you were to mistakenly switch to one of them you'd go from relative quiet to full reference in a split second. This is guaranteed to wake you up, and could possibly damage your speakers...

Carlos

HTcrazed,
It sounded like you were asking...
I've only ever heard that "C" weighting and "Slow" modes be used for home theater speaker calibration.
The "Slow" so that you can read the response, or determine the average (as you sometimes have to do with a subwoofer). The "C" I'm not sure about, but it's the only setting I've seen be recommended.

also use an outside source for taking these measurments,

Avia,VE or Sound and Visions Tune Up,

the sound coming from your dvd player will be different then just using the tones in your preamp....

C and A weighting apply different values to different octaves. Can't remember for sure which is which, but one is based upon equal sound energy in each octave, the other is based upon the 'normal' human hearing curve. The slow response just keeps the meter from bouncing around too much, giving more of an average reading.

I did this once. I think I was lucky I didn't blow any drivers...

I read a THX document once that said you're supposed to point the mike straight up, and I've always done it that way since. Makes it easier because you don't have to change the position of the meter when measuring the surrounds, for example. Don't know if it makes a difference or not.



This is why when *anybody* says they can hear "etched highs", a "detailed midrange," and a "tight but robust low end," I don't put much stock in it. Human's hearing just isn't that great. And "auditory memory" is even worse. I want to see measurements darn it! Easier to compare components that way. Source components and pre/pros-receivers. Speakers are way different of course (interactions with the room, etc).


[This message has been edited by Kevin C Brown (edited August 02, 2002).]

Hmmm..
Kevin, I don't think a system's speaker levels would relate to your comments about etched highs or midrange.
That's usually releated to speaker's overall freq. response and associated equiptment sound quality -regardless of matching or unmatched surround levels.
I agree w/ the bass range though 'cuz then you can be talking about a sub's level being poorly leveled w/ the mains.

And I don't know Kevin... Human ears ain't that good? Specs are better?
I gotta disagree.

Most solid state amps all rate pretty much close to zero % dist. at normal list. levels, yet they do not all sound the same at this level to human ears. Are the lab tests wrong?

No, we just didn't also measure (probably many) other factors that our ears can easily hear when we switch one component for another.

Unlike our crappy human eyes that are easily duped into thinking we're seeing motion by still images switched as low as ~20 frames a sec. most (probably all of us) can tell very tiny changes in audio waves processed @ tens of thousands of times a second.

It may be possible to measure all of our equip. more accurately than our ears can hear, but these tests either don't all exist yet, and/or just aren't being done -hence the need for lots of human opinion reviews in add. to lab tests.

I really wish the human factor could be taken out of the equation, but it can't yet.
If it could then we'd be able to say the Outlaw 950 is just as good as pre/pro X and the long debates wouldn't be needed. We could just state specs and point to the winner.

This SLP meter thread's a great example.

The idea of perfectly setting all your speakers to the exactly same volume is a great idea, but 'real world' you see here all the probs that crop up in trying to do so...

The analog Rat Shack meter (prefered over the digital by everyone) is very inaccurate and needs correction after you (additionally inaccurately) read the meter.

Then do you point the meter up so it's more omnidirectional (though it's still not totally), or do you point it forward closer mimicing how are ears are 'aimed' forward?
Some say one way. Some say the other.
'A' or 'C' weighted?

Which brings us to the fact that identical main speakers placed behind you will sound different because of how your ears are shaped (muffling the high end in simplistic terms) so matching their volume isn't ever going to be totally accurate anyway.

Then you just calibrated w/ your pre/pro's test tones, BUT when you play a test DVD like V.E. or Avia, etc., you find it's calibrated differently, so you have to choose which to calibrate to 'cuz they both can't be 'right'.

And then you have the fact that after you do your very very very best you can to minimize all these inherent flaws in the calibration method, you have to contend with assorted DVD soundtracks that have diff. surround, sub levels anyway screwing up your effort far more than any previous variable.
And these DVD/CD's are what you calibrated you system for in the first place!
Almost a cruel joke it seems.

I can play one DVD and feel the rears are too loud. Play another and their too quiet. I'm sure others have found this to be the case also.
Some might say -'Well at least you know you're hearing it the way it was meant it to sound', but I'd bet that's not really the case.

And let's say 'system x' is calibrated 'perefectly' to 75 or 80db being -0.-most people still choose to lower the volume from this 'ref. level' altering the entire calibrated system anyway.
Do you ever tap the volume during a flick? I know I ain't the only one who can't put the damn remote down and just watch the freakin' movie! -heh

All these factors have stopped me from getting an SLP meter myself and calibrating my system to an assortment of these personally chosen variables.

Instead I listen to reference point tracks of several DVDs and CDs and adjust the surrounds and my dual subs so they sound well blended in on all these varied reference points.

Highly accurate? No, but not any less accurate that the current 'technical' method.

Just a thought I wanted to throw out there.
(i like to play w/ matches. let the flames begin -heh)

[This message has been edited by azryan (edited August 04, 2002).]

I don't think the value of an SPF meter is that human hearing isn't very good. If it weren't, how would we hear better results by using one?

I think it has most to do with the different speakers sizes and types used in a 5.1 plus setting and most of us not knowing what a multichannel setup should actually sound like.

For example, like most I'm using towers for mains, bookshelfs for surrounds and something larger than bookshelf for center. Though they're all B&W, they sound different when the tone's going through and its harder to compare fine gradiations in loudness to sounds of differing frequencies.

Also, even though I did hear my left main was louder than the right, I never would have dreamed of setting them to different levels (a leftover old wives tail from 2 channel days). And the default of the 950 makes the voice tracks easy to hear by having the center louder, but it turns out to noticeably degrade the overall soundstage unless you set it even to the other speakers (which I think resulted in a center speaker setting of -6 in my system).

I don't think the problem is with human hearing ability, more with our not being familiar enough yet with what a multichannel system is actually supposed to sound like when optimally setup. Which is why the SPF meter works so impressively.

Judgements about how revealing the mids, etched the highs, bloomy the bass, size of the soundstage, has nothing to do with comparing volume levels. And if we weren't capable of discriminating these sounds, non of us would ever need to consider the extra cost and fewer creature comforts of seperates in the first place.

It's ok you are satisfied with your calibration methods,azryan, but for the rest of us (vast majority) using the meter is a much better way to go. After all, most, if not all manufacturers recommend using a meter for best results. Why does the meter give the best results? Probably has something to do with that thing found between our ears. I think you need to establish a good base to work from and the meter will give you that. Once you do this- go ahead and tweak the settings based on the material you're listening to- IF you feel it's necessary. A properly calibrated system should reproduce the soundtrack as it was intended for you to hear it- with the appropriate sounds and volume levels played back in their assigned locations.