Asymmetrical Speaker Time Delay and Loudness? (1 Viewer)

[LarryGS]

All the literature about setting speaker time delay and volume-level assumes the listener is sitting in or near one location. But what if that's nowhere near the case? What if you have people sitting in various places, ranging from the extreme side of the room, to the center, and to the rear? Or even on the floor directly in front of the screen? Since any adjustment for one listening position inherently weakens other listening positions, is it best to just throw in the towel and set all speakers to an identical distance and volume level?

Theaters must face this same problem. How do they deal with it?

Thanks!

 

[Stephen Weller]

In the words of a famous person, "You can please some of the people some of the time..."

If your listening area is large enough, you may be able to create a large enough sweet spot. In practice, it's probably not possible. Even in large theaters, *everyone* doesn't get to sit in the sweet spot.

 

[Bob McElfresh]

Yes, this is a problem.

First, forget about time-delay's. These dont have much effect.

The volume - is a big issue, usually for the rear speakers where people on the ends of a couch are 'wearing' one speaker right by their ear.

The secret is to try and get some large distance between the ears of the extreme-edge listeners and the speakers. This may mean moving speakers back, installing them up on a wall or even turing speakers to fire AWAY from the listeners to create a 'bounce' off of a nearby side or back wall.

Now the volume difference from the central to the edge should not be as great.

 

[LarryGS]

Thanks for the replies! You've pretty much confirmed what my gut was telling me... In this scenario, the job isn't so much to optimize a given spot as to make most of the room as passable as possible.

I've aimed the surrounds parallel to the wall, pointing rearward, and it hasn't hurt the surround channels much. Maybe even improved the room's ambience slightly. I've also reduced the volume level of the surrounds slightly to make sure the front and (especially) center speakers are clear throughout the room. I've left all timings alone.

Again, thanks!

 

[Bob McElfresh]

Well... you SHOULD use a SPL meter (Radio Shack $40) and test-tones to re-adjust the levels if you move or change the angle of your rear speakers. You DO want the rear-effects to be loud/soft as the sound engineer intended.

The SPL meter will also allow you to put numbers to the volume at the central location, and at various spots in the room.

The goal is to find a speaker placement/alignment that minimizes the difference from the central to the edge listening position. But this is with the rear-speaker volume set to match the front speakers for the same sound.

 

[LarryGS]

Thanks, Bob. I do have the RS SPL meter and that's what prompted this thread. "minimizes the difference from the central to the edge listening position" is exactly the problem I ran into when trying to set up an asymmetric room with wierd seating.

In weighing all the considerations, shouldn't a clear rendition of the front channels (especially center) be the primary concern for everyone in the room? To me the strategy would then be to find a surround level that enhances everyone's experience while interfering as little as possible with center dialog for rear or side seated listeners.

I assume sound engineers and Dolby/DTS encoders take this into account when mixing the various channels since theaters face this same problem. Or do they in fact design the sound for a sweet spot?

 

[Bob McElfresh]

Sure. You can usually crank the center up a bit. But the center is not typically a problem. It's the L/R rear speakers. Someone sitting on the ends of a couch is sitting 2-3 feet from one speaker, but 8-9 feet the other rear and about 10 ft to the center. The too-close speaker can drown out the others.

Off the top of my head, I would do this:

- Adjust the rear speakers to give more distance, fire indirect, etc.

- Use the SPL meter at the center seat to level-adjust all the speakers. (HINT: Write down the values on the reciever for the center and left/right speakers. This is the +/- db values on the display).

- Move the SPL meter to one of the extreme seats and measure the volume from the nearest rear speaker. If you used 75 db to adjust everything, the near speaker may give ... 95 db for example.

- Take the difference (95 - 75 = 20 db) and cut it in half to give you 10 db.

- Put the SPL meter in the central position and with the front speakers producing 75 db, make both rear speakers produce 75-10 = 65 db. (I'm assuming both rear speakers have nearby seats.)

Now pop in a favorite, effects-heavy DVD and listen. Listen from the central position, and the problem position. Make a judgement call: "Is this a good volume where the central position can still enjoy the rear effects, but someone at the edges is not being blasted by the speaker?"

Adjust both rear speakers up/down a few points as a fine-tune.

Keep in mind: this IS a compromise. The central seats will not hear the rears as loud, but the edge seats wont suffer as much. This will give a over-all better experience for everybody.

Let us know what you try and what works/does not work. You are not the only person with either a problem room or problem seating. Your feedback will inspire others.

 

[LarryGS]

Well, I don't know about the "inspire others" part, but here's where I am now:

Using the sound pressure meter, I calibrated all the most common listening positions, thinking I would just use the average correction settings. One immediate problem is that I was getting HUGELY different readings from two different sets of test tones. The built-in test tone on my Pioneer VSX-812 is more of a "hiss" (white noise?) sound, while the test tones on my calibration DVD is more of a "roar" (pink noise?) sound. The corrections required to balance the sound levels from any given listening position are drastically different for each set of tones, probably due to varying speaker size. (I have a matched pair of large Advent front speakers, a matched pair of 8" 3-way surround speakers, and a 2-way Sony center that sounds weak at the high end.)

I ended up taking the average of all readings from all listening positions, using BOTH sets of test tones. The net result, despite some wildly divergent readings (close to specific speakers) was almost exactly where my ears initially predicted. Using the white noise alone resulted in a too-soft center and too-loud surrounds, while using the pink noise alone gave the reverse.

Finally, as you correctly pointed out, I had to compromise a bit to overcome a couple of particularly bad listening positions. By boosting the center .5 db, and pointing the surrounds toward the rear, away from nearby listeners, I think I have a reasonably good balance from most positions. It's a compromise, to be sure, but you can hear all channels, with clear center speech and reasonable surround effects and ambiance, from all locations.

I'd be interested to hear what others think about the white noise vs. pink noise issue. Thanks!

 

[Stephen Weller]

I have a vintage spectrum analyzer with a built in pink noise generator. What I have "seen" coming out of my Yamaha is neither pink nor white noise. It's some narrow band of tones centered around 1KHz. Which is annoyingly convenient, since that's what everything seems to be spec'd at. :frowning:

As such, I consider the built-in tones only useful as a go-no go test. Or to verify that all channels are working.

In your case, test disk is the best reference. But I'm wondering if you shouldn't be more concerned with peak than average. Undecided because the majority of surround information creates ambiance (as opposed to special effects). In terms of ambiance, it should be fairly easy to give everyone a "piece of the pie."

 

[LarryGS]

Yes, you're pretty much describing what I'm hearing.

And yes, giving everyone a piece of the pie is what I think I've ended up with.

Thanks!