Originally Posted by anothermib
I am having a strange issue with Audyssey MultEq (6 measurement positions) on a Denon AVR 2309: The levels of all 3 front channels are consistently being set about 3db too low when compared with results obtained using a test DVD. Just using the Audyssey results clearly sounds off compared to the DVD results, which are balanced much better.
Any idea what may be going wrong here? I don't mind balancing levels manually and the measured speaker distances are more or less ok. However, I am wondering if any measurement issue may be impacting the frequency domain, which is harder to fix manually.
You're comparing apples with oranges. or at least comparing 2 different varieties of apple.
You're comparing results from 2 different test signals, actually different test signal types, and each of them is being measured with a different mic. There's almost certainly going to be different levels reported. I don't know what the specs on the Audyssey mic and whatever you used to measure the DVD tones are but the Audyssey mic is likely to operate within a +/- 2 or 3 dB range and whatever you used with the DVD tones is likely similar in which case you could end up with up to a 6dB difference between the readings from both mics if you were measuring exactly the same test tone at the same time, and the tones are different. The Audyssey tones are a very fast pulse while the DVD tones are a continuous white or pink noise. You were probably using a handheld meter of some kind to measure the DVD tones and meters have options for settings like fast or slow averaging and you'll get a difference if you change that setting while measuring the DVD tones. I'd say the difference in your results is well within the acceptable range given what you're measuring and what you're measuring it with.
As for the difference in sound between the results, if you're adjusting the trims set by Audyssey afterwards based on the results from your DVD measurements, then you're probably ending up with a slightly higher level output when you listen with the adjusted trims and it's been shown that if people compare the sound of the same thing and one of the comparison samples is slightly higher in level than the other but otherwise both samples are identical, people consistently prefer the louder sample. If you're not matching playback levels exactly for your comparison and the playback with your DVD based trims is slightly higher than Audyssey, then you will prefer the sound with the DVD trims. Try turning the master volume up a little when listening with the Audyssey settings and you may find yourself preferring the Audyssey setting. If you do then volume differences are likely playing a part,
And if you're comparing the sound with Audyssey on with Audyssey settings to Audyssey off with the DVD based settings, you've got another difference. If you aren't used to listening to sound with correction you may well prefer the uncorrected sound because of differences in the tonal character. When we compare sound and there's a difference in tonal character we often tend to prefer the sound we're familiar with but if we persevere with the other sound, listen with it for several days to a week or so and become familiar with it and then swap back to the other, we can find ourselves preferring the sound we were initially unfamiliar with so you may have to take the time to get used to Audyssey before deciding whether you prefer it or not. And in the end each of us has different listening preferences anyway. Not everyone prefers the sound with Audyssey and you may be one of the people who doesn't prefer it.
So, there's nothing surprising or unusual in you getting different results from your different setup procedures or in your reactions to the differences you hear when using different settings, and nothing is probably wrong. You're listening to 2 different things and if you hear a difference—which you do—then you are going to prefer one over the other. Which one you prefer is going to depend on whether one is slightly louder than the other, whether there are other differences and Audyssey does make changes in both the frequency and time domain, and then there's the issue of familiarity and personal preference.
My advice would be to just do the Audyssey setup and then listen with those results for a week or so to get used to it. If you still aren't happy with it then try adjusting the sub level by raising the sub trim settings a bit because many of us find Audyssey sets the subs a little lower than we prefer and many raise the sub setting by 3 to 5 dB. You can also experiment by changing between the Audyssey Flat and Audyssey Reference settings and with Dynamic EQ on or off. I would leave Dynamic Volume off. If you can't get a result you like that way, then try turning Audyssey off without adjusting the trim settings and just try listening with Audyssey off and adjusting the master volume to your preference. After trying those things, if you're still unhappy then perhaps you're one of the people who doesn't find Audyssey's results match their sound preferences in which case just set things manually using your DVD results and enjoy that.
The other thing to remember is that Audyssey bases it's results on a series of measurements at different positions. The number of measurements you make and the spacing of the different positions does make a difference. You can also experiment with fewer or more measurements and with more closely and more distantly placed measurement positions and that may well give you some very different results with Audyssey but I'd start with the things I suggested in the previous paragraph before considering embarking on a series of experiments varying the number of measurements you make and your microphone placement pattern because that can become a time consuming process. When I first tried using Audyssey it took me a number of different measurement runs before I found a measurement procedure which worked for me and it took me a while to adjust to the difference in sound with Audyssey. I ended up preferring results with Audyssey but it took me a few days to come to appreciate it.
Edit: Didn't see the responses above, all of which make good points, before posting the above. There's some other tips in them which are also good but I've left my original response untouched because it provides a bit of background on why you're seeing measurement differences, and because it also covers why you may have a preference for one result over the other and you may find something helpful in there.