Home Theater Forum and Systems banner

1 - 5 of 5 Posts

·
Registered
Joined
·
39 Posts
Discussion Starter #1
I am currently using Audyssey Pro for calibration my home theater, but I do not like its "black box" routine. I can definately hear, that its doing something to the sound, and compared to the "raw" system, its an improvement. But what I dont like about it is, that it makes dialogue etc sound a little diffuse. Hard to explain, but without Audyssey enabled, it is easier to pinpoint the sound from the speakers. Audyssey mentions a lot about the work in the time domain, but they do not mention exactly what is beeing done.

I have thought about using Smaart v7 with multiple microphones and then measure the average response of all mics using pink noise and setting up the system manually. I have a DBX Driverack 4800 which should do a good result. I would be able to adjust the system while watching the freq.response/phase in real time.

I would really appreciate some inputs and some discussions about the different ways of calibration the system. If someone has experience with the JBL HATS/ARCOS and how things are done, it would be very interesting too. They use the BSS Soundwebs which have the same filters/processing power as the Driverack 4800.
 

·
Registered
Joined
·
350 Posts
One can use Room Equalization Wizard (REW)(freeware) with a calibrated microphone and a compatible dsp crossover/equalizer (such as miniDSP) to equalize the bass. It is far more sophisticated than even Audyssey Pro.
I do have to say that my current system requires little equalization and I have REW and a Dayton calibrated test microphone.
 

·
Super Moderator
Joined
·
9,200 Posts

For starters, I’ve never been a fan of the vertical mic orientation that Audyssey and other auto-EQ systems use.

Using vertical orientation for a measurement mic, even with a calibration file, can be unpredictable and skew results. The “unpredictable” factor is your ceiling and walls – where they are in relation to your speakers and the mic itself, how absorptive they are (or not), how high your ceiling is, etc. For instance, you can naturally expect vastly different reflections arriving at the seating location from a cathedral ceiling vs. an 8-ft. high flat ceiling.

Measurement mics, even though they’re omni-directional, show directional characteristics at higher frequencies. As a result, off-axis upper-frequency response droops rather drastically. And that’s exactly what you have with vertical (90º) orientation: the mic element is severely off-axis from the primary signal source – the speaker. It’s not unusual for response to start sagging at 2 kHz.


measurement mic 90 off axis response.jpg
Common Off-Axis High Frequency Response of a Measurement Mic


Of course, when you use vertical orientation for the measurement mic, you want to use the appropriate calibration file, which compensates for the natural off-axis high-frequency loss. The calibration file allows vertical mic orientation to measure the same with as it would with the mic aimed more conventionally - horizontal and pointed directly at the speaker.

However, the same calibration that renders off-axis response flat also boosts on-axis response: If you use a vertical-orientation calibration file and aim the mic at the speaker, your measurement will show an erroneous rise in high frequency response. What this means for a vertical mic orientation is that a measurement can actually give preference to the reflected sound over the direct signal from the speaker.

For example, using the graph above as an example of the mic’s off-axis response: Let’s say your room has poor high frequency absorption, and the sound reflecting off the ceiling is attenuated only a couple dB at 8-12 kHz when it arrives - essentially on-axis - at the vertically-oriented measurement mic. The 90º calibration file has boosted on-axis response by 5 dB or more in the 8-12 kHz range. This means the measurement platform (REW, Audyssey, Smaart, etc.) will see the reflected signal as 3 dB hotter than the direct signal. The mic can’t tell a direct signal from reflected; it’s only going to “see” the one with the higher SPL. So an auto-EQ program will adjust response accordingly, ratcheting down response in that region.


Reflected signal to vertical mic.jpg


So, if your room / ceiling situation is such that you get strong reflected sound at the mic location, then naturally the reflections are going to have more influence on the frequency response measurement if you utilize a vertical mic orientation, than they would in a room where the reflected sound is better absorbed, diffused or directed away from the measurement mic.

I realize the above is a bit simplistic and based on a certain amount of assumption, but hopefully you can see how 90 º mic orientations can foul up an auto EQ or even a manual EQ process. For more on mic orientation, see here and here.

More specifically regarding to your question about measurement and equalization techniques, I assume you have ample EQ capabilities with your Driverack. The best approach is pretty simple: measure each main-channel speaker independently, and equalize each as needed. Alternately, you could take measurements of each speaker from more than one listening location EQ based on and average.

You can certainly take a combined-response reading, but as an FYI only. You certainly don’t want to do any kind of global EQ from a combined measurement. If there’s a peak in response somewhere it’s probably coming from a specific speaker. Global EQ would address the issue in the problem speaker, but it will also unnecessarily kill the same frequencies in the speakers that aren’t exhibiting the problem.

Naturally, subwoofer response is addressed separately from the main channels. If you have multiple subs, measure and EQ combined response. That seems to be the best and most effective approach. Many people find that separate EQing of multiple subs is an exercise in frustration: When you get response looking good in each individual sub, it’s often totally whacked when you take a combined reading.

Regards,
Wayne
 

·
Registered
Joined
·
39 Posts
Discussion Starter #5
Thanks for the responses! I've been a little busy with experimenting with the techniques at hand, and if you're interested, I will describe my discovery.

As I wrote, I have been using Audyssey Pro for a while, but felt like something was "missing" when it had done its work.. Punch and that "slam" and full sound, and the fact that I dont know exactly what Audyssey is doing, is messing up my mind!

I have always been a fan of the "keep it simple" rule, which for me, is applying the least amount of filters to sort out the issues the room might have. (My room has very good acoustics with an absorbing ceiling and absorbers placed around the room). I used REW for the measurements and filter generation.

I first set the X-over for the sub on my processor and measured and EQ'ed it by using the filters generated by REW.

The raw sub measurement and the filters generated by REW:


After applying the filters and re-measuring:

Very nice result in my opinion!

I then measured the LCR's (individually of course) with the sub still playing. This was to ensure a good integration to the sub. Raw measuremetn and filters:


Filters applied and re-measuring:


A comparison of the before and after measurement:


I am extremely happy with the graphs. They measure much better than what Audyssey did to my system.

And the listening impressions: Its amazing! All the things that I thought was wrong with what Audyssey did, is now completely gone, and all the things that were wrong prior to any calibration, is now gone too. I am really amazed by the power of REW and PEQ filters. (I used a maximum of 13 per channel).

Just wanted to share my results! :)
 
1 - 5 of 5 Posts
Top