Originally Posted by Overrid3
I posted the message below over in the REW section, but wanted to ask over here, too. My situation is described in more detail below, but basically I'm wondering if it's OK to change the phase of my sub after the Audyssey calibration? It appears that the 180 setting is actually correct, but I've been running at zero all this time (including when I calibrated with Audy).
Question about subwoofer phase—
I'm running a 3.1 setup with Wharfdale towers and center, and an SVS sub. Ever since I've set it up, I've had the sub phase dial at zero. I've always had some issues with music not sounding full enough (I listen in 2.1). My AVR is a Marantz with Audyssey XT32, and with EQ enabled, the speakers lose some of their bass. I did some measurements, and sure enough, these speakers have a bass hump centered around 100 Hz (see first graph).
Now with the Audyssey reference curve enabled, that hump goes away, BUT worse yet, it appears there is a trough centered around 90 Hz. I think this explains the lack of bass I hear in that area in pure direct mode vs Audyssey EQ'd mode.
I played around with different crossovers for my sub, but nothing seemed to really help. Pushing it all the way up to 120 Hz actually fixed that trough, but I don't think I want to cross it that high. Finally, I remember the phase knob on my sub. I changed it from zero to 180, and it looks like the low bass is a lot flatter. Everything else appears to be unaffected. I guess this is the correct setting?
OK, I know nothing about REW but I can comment on the Audyssey side of things and make some other comments.
First, in theory you should not change the sub phase setting after Audyssey calibration. The reason for that is that Audyssey measures and corrects for phase as well as for frequency response. Changing the sub's phase setting after calibration changes the sub's in room response from what it has determined will be the correct setting and it will change what you hear at the listening position.
Now, I don't know what's going on with the REW measurements because, as I said, I know nothing about REW but I can make the following observations:
- You'll be making your REW measurements with a different mic and a different test signal. That means that what you're measuring with REW and what Audyssey measures with its test tones are completely different in some important ways. How that difference is going to affect things is impossible to predict but if you could get an accurate readout of Audyssey and REW measurements after correction you would expect to see some differences because of it. Unfortunately we can't get an Audyssey readout after correction because Audyssey doesn't take any measurements after it does its corrections so we have no way of doing such a comparison.
- Audyssey does a number of measurements in a spaced pattern. When it comes to bass frequencies the result of those measurements is extremely dependent on mic location. Bass response in a room can change considerably in the space of a few inches. I've been in a room auditioning speakers and been unable to hear a prominent bass note at the listening position but on standing up and taking one step forward the same note was clear and loud. That's the most extreme variation in bass levels in a room that I've heard and it goes to show the importance of the listening/measuring position when assessing bass performance. If you didn't have your mic in EXACTLY the same position for the REW measurement as the Audyssey mic was in for the first Audyssey measurement and if the mics don't have EXACTLY the same response, then Audyssey and REW are going to give you different measurements. With measurements involving separate measurements at different locations and their integration, as occurs during Audyssey calibration, you would need to repeat the measurements with REW with a mic with identical response and placed in absolutely identical positions and do exactly the same integration calculations or the results are going to be different. I don't t think there's any way of doing that so there's always going to be differences.
- Audyssey is intended to produce a result that most people like over an area larger than a single listening position. It cannot produce the same result over the whole of that listening area but it tries to optimise the result so that wherever you sit in that area the sound you hear will sound good. That almost certainly means that the sound at the primary listening position, the first measurement position, is not going to deliver the best result at all frequencies, it will be better at some frequencies than other positions are, and worse at some than other positions are, and every position is going to be slightly different. Whether you use Audyssey or some other room correction system, no system can deliver identical results everywhere within a seating area that accommodates several people. That's simply impossible.
- Just because a particular measurement shows a dip like your 90 Hz dip doesn't mean that you will hear that dip. Our brains are remarkably good at processing the signals they get from our ears to "correct" for the kind of differences in level that occur in rooms at low frequencies. You really can't judge how good or bad the sound will appear to be at a particular position by looking at a measurement chart from measurements taken at that position. You really have to sit in that position and listen if you want to know what things sound like because our ears and brain work very differently to the way a microphone and measuring equipment work.
All of that leads to to 3 things:
1- How does the Audyssey result actually sound to you when listening to music in the listening position? If it sounds good, then that's what it's supposed to do and you can just stop worrying about measurements, no matter how bad they look. Change things to make the measurements look better and the sound will change and there's no guarantee that you will find the change in sound to be an improvement, it could easily sound worse.
2- If things don't sound good to you after Audyssey calibration then you can either change some of the things that Audyssey lets you change or repeat the calibration with slightly different mic positions in order to get a result that you do like.
3- If what you want is a result that delivers the best that Audyssey is capable of at just one single position then you can do the calibration process of several measurements without moving the mic to different positions, That should produce a result that one would expect to sound and measure best for that single position but it will probably also produce a result that sounds worse for other listening positions. If you listen on your own most of the time then this is a reasonable option but if there are usually other people listening with you then it's not something I'd recommend.
The bottom line is that you really should not obsess over what measurements show, what counts is the actual listening experience you have. If you enjoy the listening experience then that's a good result, if you don't then that's a bad result. What the measurements look like is irrelevant if it sounds good to you and those listening with you.
And if things don't sound quite right to you at the moment and changing the subwoofer phase setting to 180 degrees makes things sound better to you than they do if you don't change the phase setting, then change the phase setting. Just don't change it just because the measurements look better because the measurements are not always a good predictor of how things are going to sound and what's important is how they sound to you, not how they measure.