Originally Posted by giomania
I listened to some music, and was pleased with the result, but I am going to perform another Audyssey calibration, because it can't hurt to try. I will still have the calibration that resulted in the above result in the app, so there is no down-side other than using my (limited) time.
For the new calibration, I am looking for input on two changes:
1) I am thinking to make the microphone positions closer together this time. I have always kept them 18-24" apart, but am thinking to try around 12" apart.
2) I would like to try and measure with the subwoofers at the actual levels I use them at, because that probably affects the Audyssey result. It doesn't make sense to me to reduce the subwoofer levels to satisfy the subwoofer level matching screen requirements, when I just revert the settings post-calibration.
Background on my subwoofer adjustments
On my very first run, I ignored the subwoofer level matching screen, and the calibration worked fine; the subwoofer trim levels in the processor were set at -12.0 dB. I don't see this as an issue, as I know what I would manually adjust the final trim levels to anyway. Since this calibration run was made before I made my cables, I am not going to use it.
Ever since that first calibration run, I have been using the subwoofer level matching screen, but I have to turn the gain level settings on the subwoofers down significantly to satisfy that tool in the app. Then, after calibration, I set the subwoofer level gain settings back to the original positions and also reduce the trim levels in the processor; the end result is a definitely a net gain. Please note these gain settings on the subwoofers and the trim levels in the processor match the 2010 Audyssey Pro calibration performed with my Denon 4311. I realize the custom curve in the Pro calibration is going to be different, but the only thing that changed since that time was the processor
. In addition to these post-Audyssey subwoofer gain and processor level adjustments, I also add +5 dB boost to both Sub1 and Sub2 via the "Options" menu.
Thanks for any input/advice.
Originally Posted by BluesDaddy56
I don't see where anyone else responded. I think closer distances for the various measurements makes sense. I try to do mine approx. 12" apart. Regarding sub levels, if Audyssey sets the trim at -12, that's its maximum so you don't know if it should have been MORE than that or not. Set your sub gain so Audyssey trims between -11.5 and -7. That allows you to know it shouldn't be more and gives you plenty of headroom to play with setting them higher post calibration. I have two subs and Audyssey consistently sets one at -10.5 and the other at -7 or -8. I bump the 3db together post calibration.
Originally Posted by giomania
Thanks for the response. Regarding the subwoofer levels, I know exactly where both the subwoofer gain level and the processor trim levels were set in my system for years. Like I said, the only thing that changed was the processor. Also, Audyssey tells you exactly where they want the subwoofer levels via the subwoofer level matching screen that appears as the first step in calibration.
My question is if the Audyssey correction will change due to the subwoofer levels being set where I want them in the end, versus where Audyssey wants them at the start of the calibration. The issue I have is that Audyssey applied a several dB cut at around 38 Hz (as measured with REW), and there is no null at that frequency when Audyssey is off (as measured with REW). The curve editor did not work to remove the dip in the response, but maybe I don't know how to use it.
Originally Posted by giomania
Did you measure with REW to verify these results?
Such inaccurate results as adding a 4 dB boost at 20 Hz, and that affecting the 32 Hz range explains why I could not eliminate the 38 Hz dip in my sub frequency response by boosting at 38 Hz.
Sent from my iPhone using Tapatalk Pro
Originally Posted by Jon AA
Yes, I've been using REW and Omnimic to verify everything I do.
Keep in mind, if a dip is a room null/cancellation, adding boost with any software often won't fix the problem. Relocating the sub/adding another sub, etc might be required.
OK, I think I finally achieved good results with the curve editor, as now the response is within 1.6 dB (maximum) below 38 Hz.
It seems like I only put around a 4 dB boost at 45 Hz, but I put a couple of other points in there (wish I wrote them down), but they are not showing when I go back into the curve editor for the subwoofer. I uploaded it, checked it with REW, was pleased with the result above. Then I listened to some music, and was also pleased with the improvement. Then I came back to the computer to capture my graph and post here.
Someone mentioned that Audyssey will try and flatten the response, which I guess means that it takes the curve editor control points we adjust and interprets them into an alternate result. Based on the curve adjustments I made, and the follow-up measurements, it looks like that is what happened to me because that 38 Hz "dip" (which is not there with Audyssey off, so not a null) is still there, but the response below that was flattened. I am not sure what happened the last time I tried to edit the curve, and it didn't change at all. Maybe that was a glitch in the app?
For the Audyssey calibration today, I did not make any changes in the microphone pattern distance, as I realized my microphone pattern was fairly tight anyway, around 12" apart. I did change measurements 7 & 8; I raised those up around a foot above the MLP, almost directly above measurements 2 & 3. There was not much change in the result with the default parameters from Audyssey, so I guess that just means I have the measurement process fairly consistent.
Today I did not adjust the subwoofer gain level settings down
as I had in the past, as the levels on the subwoofer level matching screen were only slightly above the green target. Basically, each pair of subs (sub 1 and sub 2) were only one speaker icon above the green target. I figured that would be fine, and it turns out it was. The only issue is that the processor trim levels were set at -12.0 for the subwoofers, but as noted above, I know the subwoofer gain level settings are fine in my setup.
After the calibration, I re-adjusted my subwoofer distances as I had confirmed they were the best blend in previous posts. Then I boosted my subwoofers by 7.0 dB in the processor trim level settings menu.
I am not going to say it is perfect, but with music, it sounds much better in the bass region. We will see when I watch a movie tonight.
Thanks for everyone's input.