AVS Forum banner

41 - 59 of 59 Posts

·
Premium Member
Joined
·
12,671 Posts
Use a dedicated center speaker with multiple drivers (a real 3 ways).
If it has the shape of a sound bar, place it just at the bottom of the screen and tilt it 60 degrees up.
That way the tweeter is at the center, one mid range on each side of the tweeter and one bass driver at each end of the sound bar.
I even have one bass reflex exit at each lateral extremity of the sound bar.
You cannot obtain the same spacial configuration with a tower that is vertical.
The biggest fallacy in your logic is this: "You cannot obtain the same spacial configuration with a tower that is vertical." That is simply incorrect. A horizontally symmetrical speaker, (such as MTM, W(TM)W, or W(MTM)W alignments), will, by the physics of driver interactions, exhibit horizontal off-axis driver interference. There is a massive amount of scientific data to prove this effect. There are ways to minimize this effect, some better than others. An MTM will be the worst, while 3-ways with center mounted mid/tweeter arrangeements will be better. Nonetheless, there is NO WAY to eliminate the driver interactions in the horizontal plane except by using a vertical alignment of the drivers. Three vertically aligned speakers, all placed at the same height horizontally, is the unequivocally best arrangement for a solid, cohesive integrated front soundstage. That is why so many commercial theaters, as well as the most optimally designed Home Theaters use this configuration.

The other fallacy is this: "If it has the shape of a sound bar, place it just at the bottom of the screen and tilt it 60 degrees up." Placing a CC below the screen can never work as well as placing it behind an Acoustically Transparent screen. While it's true that an AT screen will have some acoustic deficit, the newer woven screens trivialize that problem and never require the amount of EQ that can potentially damage tweeters. Here's a graph of the attenuation caused by the Seymour AV woven screens:


Virtually all speakers can handle a dB or 3 of HF boost to compensate for the screen. That is no more than what happens with a click or two of the treble control, or the flip of the treble switch on the speaker itself.

Also angling it up at 60 degrees may or may not be the most optimal angle.. The key is to angle it enough so that it aims at the listener's ears.

While using a horizontal CC below the screen can be made to "work", with results that many users find acceptable, it is not, as you seem to think, the most optimal or best arrangement. That would be reserved for the 3 vertically aligned speakers, all at ear height.

Craig
 

·
Premium Member
Joined
·
12,671 Posts
[Clip]Horizontal speakers typically produce comb effects and that's why they're bad. [/Clip]
Best post I've seen on this subject in quite a while. Thank you!

Craig
 

·
Registered
Joined
·
124 Posts
Well all depend of the room that is used as a theater.

Malco Theater in Menphis Tennessee: all the equipment are using horn speakers from Klipsch.
There is no tower speaker and all is professional equipment similar to concert hall ones.

From this extreme solution there are many lower range intermediary configurations that need to be validated.
I am very dubious about a tower speaker using a dome tweeter (or worst ribbon) placed behind the screen.
All depend of the targeted SPL level.
It is well known that at high SPL the first driver to burn is the tweeter.
Anyway each Home Theater setting is custom, and you may use a screen material that has a very low sound attenuation in the high frequencies.
 

·
Registered
Joined
·
3,527 Posts
The diagrams are just examples of typical homes. They have nothing to do with "ideal" or "best" types of speakers or anything remotely like it. Dolby is pushing "bouncy" speakers for god's sake. They lost a lot of credibility with that, IMO just like THX lost theirs when they started licensing computer speakers and cables. There are some on here that act like Dolby's guidelines are the last word on setting up Atmos and that's not true either. They're a starting point for typical one couch setups. They do not generally take into consideration things like multi-row home theaters (except their largest diagram) or unusual room issues. There's a whole world of audio experience that isn't nullified just because someone at Dolby (or someone Dolby hired outside of Dolby) drew some diagrams.

In other words, things like arrays still work in the home even if Dolby doesn't talk about them. Their angles in most of the diagrams don't account for having 6 overheads either and some complained when they showed front/rear heights at 20 degree angle minimum in those newer drawings. WHY!? Well, it's because the truth is it's about old-fashioned 1960s stereo phantom imaging angles. If you bridge two stereo pairs with a third in the middle, the phantom imaging is much stronger and you don't need the overhead pairs as close together. Ideally, a full Atmos theater would have height speakers above the screen and across the ceiling to the back wall. That's ideal at home too if you have the room for it, but few people have the space or money to do it. People on here tend to prefer "tops" over "heights" because they say the direct overhead sounds are weak with heights (in decent sized rooms; small rooms wouldn't have an issue). That's because of the angles in-between too. Dolby doesn't discuss "why" in a diagram. They just show an example setup. People think they have to have the couch 2/3 back into the room. No, you don't. If you divide the room into equal spacing quadrants for the speakers, you can place the seats anywhere you like. "Side surround" can be in front of you or to the side of you or behind you. They're arrayed at a real theater. You can array them at home in larger setups too with an active mixer. That's something a high-end installer would likely know all about, but it's not shown in Dolby diagrams. I array my front wides, side surrounds and surround #1 speakers together. It works great at home just as it does at a real theater because the larger your home theater becomes, the more it's like an actual cinema theater.

Horizontal speakers typically produce comb effects and that's why they're bad. A few better designs do not, but the reason they are horizontally mounted in the first place has to do with early home theater using large CRT TVs or rear projectors and cabinets where the ONLY place to put a center speaker was generally on top of the TV. With flat screens and projectors, this is simply not the case any longer in many setups. If you want a consistent sound stage across the front, the speakers should ideally be IDENTICAL. This means the imaging won't shift or change in character or suffer lobing effects or anything else across the front sound stage.

You'll find a lot of people complaining about Dolby removing the "center spread" feature of DSU that they liked to use for music. Well, why did they like to use it for music? It's because their center channels don't match the mains and the music sounds superior using the mains more than the center channel (a simple workaround solution is to just turn off the center channel for music playback regardless of how many channels the music uses and let the mains create a phantom center which is better than a "center spread" center since it doesn't use that inferior center speaker at all. With three identical speakers set up precisely there is no audible difference between "center spread" and regular DSU for the MLP and center spread weakens the anchoring effect of the center speaker for off-axis seats by using mostly the L/R mains. That is why three identical speakers work best. They don't have to be three towers. They can be three satellite speakers with bass sent to the subwoofer. Ideally, it would be best to have identical speakers in every location in the home theater for the best possible sound match, but typically having similar or the same drivers works well enough for most people.)
When you talk about arraying speakers, do you mean matrix mixing? It seems like matrix mixing would create challenging time alignment issues, since you wouldn't want to electrically mix signals with different delays. Do you address this buy shifting time alignment to an outboard processor? Does that work well in practice? How do you do the time alignment?
 

·
Registered
Joined
·
2,929 Posts
When you talk about arraying speakers, do you mean matrix mixing? It seems like matrix mixing would create challenging time alignment issues, since you wouldn't want to electrically mix signals with different delays. Do you address this buy shifting time alignment to an outboard processor? Does that work well in practice? How do you do the time alignment?
Repeating time delays don't really become noticeable until you're about 25 feet (7m) apart of two or more sources sources playing the same thing. The arrayed speakers in the home theater are within 15 feet (4.5m) of each other (entire room is only 24' long) so any effect is heard like a short reverb reflection (in an otherwise dead room), if noticeable at all. In other words, no delay is actually needed in that size room. The guide charts for PA alignment don't even typically start until 5m (16.4').

I suppose if you have a really large room, you'd have to add some delay. You'd probably want individual room correction at that point too. Mini-DSP units can perform DIRAC correction for about $449 a pair or $900 for up to 4 pairs of speakers for the 8-channel unit and should handle the delay too if you use one set of measurement positions for all units. Otherwise, an added unit might be needed, but you can buy active mixing boards with delay built right into them as well. It's about 0.973ms per foot for a precise alignment, but varies by altitude/pressure a bit.
 

·
Registered
Joined
·
3,527 Posts
Repeating time delays don't really become noticeable until you're about 25 feet (7m) apart of two or more sources sources playing the same thing. The arrayed speakers in the home theater are within 15 feet (4.5m) of each other (entire room is only 24' long) so any effect is heard like a short reverb reflection (in an otherwise dead room), if noticeable at all. In other words, no delay is actually needed in that size room. The guide charts for PA alignment don't even typically start until 5m (16.4').

I suppose if you have a really large room, you'd have to add some delay. You'd probably want individual room correction at that point too. Mini-DSP units can perform DIRAC correction for about $449 a pair or $900 for up to 4 pairs of speakers for the 8-channel unit and should handle the delay too if you use one set of measurement positions for all units. Otherwise, an added unit might be needed, but you can buy active mixing boards with delay built right into them as well. It's about 0.973ms per foot for a precise alignment, but varies by altitude/pressure a bit.
I'm looking at getting a processor with Dirac (probably HTP-1), but my room layout (seats, doorways) really wants an extra pair of surrounds between the sides and rears. The goal is to fill in a big hole in the sound field, especially for the second row. The simplest approach would be to duplicate the side or rear surround channels, but mixing sides and rears would be problematic unless the delays are matched (electrical mixing vs open air mixing.) Mixing requires an outboard box for both the mixing and (optionally) per speaker delays. So, it's really the mixing, not the delays that require an extra box. I could just select a box that can do both. Or, I could skip the box altogether and just duplicate side or rears.

The room is 30'L x 17'W.
 

·
Premium Member
Joined
·
12,671 Posts
e
I am very dubious about a tower speaker using a dome tweeter (or worst ribbon) placed behind the screen.
All depend of the targeted SPL level.
Your dubiousness likely indicates a lack of experience with AT screens and speakers. The first time someone suggested to me that I consider an AT screen, my initial response was: "But Jeff, that is the very definition of "throwing a veil over the speakers." He invited me over to listen to his system with M&K THX speakers behind a Stewart perforated screen. It took about 2 minutes of the T-Rex scene in Jurassic Park to change my mind. Having the CC behind the screen was such a HUGE improvement in the sonic imaging and the cohesiveness and integration of the front soundstgae that I could not wait to change out my non-AT screen for a true AT screen. Moving the CC from below the screen to behind the screen was one of the most beneficial things I have ever done in my theater. I haven't looked back since.

Of course, it is strongly recommended to use capable speakers behind an AT screen. I would never recommend using low sensitivity, low power handling speakers behind an AT screen. (I wouldn't recommend them for ANY application, but a Home Theater with an AT screen would be the worst use-case.) In addition, it's is not necessary to use tower speakers. In fact, using LCR-type speakers behind an AT screen makes the most sense.

It is well known that at high SPL the first driver to burn is the tweeter.
The reason most tweeters blow is because the amplifier driving the speakers is driven into clipping. It is important to adequately power the speakers to ensure the amps don't clip. Using low impedance, low sensitivity speakers would not be recommended as they are more power hungry and more likely to drive the amps into clipping. Also, if one uses LCR-type speakers with Bass Management re-directing the bass to the subwoofers, significant amplifier headroom becomes available to drive the speakers and the issue of amplifier clipping is reduced.

Bottom line, I would not recommend putting tower speakers behind an AT screen as the first choice, but if that was the only option, I would recommend using Bass Management and crossovers of the bass to subwoofers. This would eliminate any benefit of the towers, (i.e., deeper LF extension), and make the towers far less beneficial.

Craig
 

·
Registered
Joined
·
2,929 Posts
I'm looking at getting a processor with Dirac (probably HTP-1), but my room layout (seats, doorways) really wants an extra pair of surrounds between the sides and rears. The goal is to fill in a big hole in the sound field, especially for the second row. The simplest approach would be to duplicate the side or rear surround channels, but mixing sides and rears would be problematic unless the delays are matched (electrical mixing vs open air mixing.)
Again, I'm not certain why you think a delay is really needed over such a short distance. Side surround to Sides#1 is what, about 6 feet in 30' room (assuming true front wides are also used at 25%, sides at 50%, ss#1 at 75% and rears at 100% and mains at 0% give or take). You're talking about a 6ms delay at that point. You'll never notice it. I don't notice it one bit with three sets. I mix fronts with sides for a matrixed front wide and sides with rears for a matrixed ss#1. Both are based off the nearest speaker delays, so no timing is off more than 6ms each for 12ms at most from front to rear. I can't hear anything at all. The mixer also gives 3dB separation (matrix) for the channels so it operates both as an array for sides and matrixed in-between channels. You can adjust the mix level amounts relative to base channel amounts to compensate for the precedence effect of any timing difference and easily put the center point anywhere you want it for a given row while keeping the average channel level the same for all. I put the sides to phantom image along side each row despite the speakers actually being between rows. Everything pans smoothly around the 24' room.
 

·
Registered
Joined
·
2,876 Posts
A consideration in a symmetrical front sound stage without a stage elevation is the floor reflection. With bookshelf e.g. MTM speakers above floor level, they have a strong floor bounce that creates a bass boundary reflection notch and creates a sense of spatial localization. With towers e.g. 3-way, whether MTM or not, the bass woofer interrupts that floor bounce with mutual coupling to its floor reflection, and eliminates that bass notch as well as muddling the sense of localization.

Mixing a horizontal center with side towers mixes the two types of image perception and further muddies the sound in addition to the horizontal combing and lobing.

Another consideration is that the horizontal combing and lobing does more than just create frequency response aberration and muddy the sound for off-axis listeners. It also creates a sense of horizontal localization for each individual driver because horizontal localization perception is finely keyed off the binaural delay between ears in the head-related transfer function. This sense of horizontal localization is perceptible even when listening on-axis for any frequencies that are not mutually coupled between the drivers i.e. the 1/4 wavelength is shorter than the distance between drivers.

A third consideration is that with budget speakers that have second-order crossovers and phase inversion between drivers, mixing a two-way horizontal center speaker with a three-way set of l/r towers creates a phase mismatch between drivers that are producing the same frequencies because the crossover points vary between the 2-way and 3-way speakers. This phase mismatch is most evident to listeners seated between a l/r speaker and the center speaker and sounds very much like a classic phase miswire on a stereo system. The bass isn't nulled but the midrange and treble sound off from each other in two separate bands.

These are all excellent reasons to keep the front sound stage consistent for the best immersive image.
 

·
Registered
Joined
·
3,675 Posts
Your dubiousness likely indicates a lack of experience with AT screens and speakers. The first time someone suggested to me that I consider an AT screen, my initial response was: "But Jeff, that is the very definition of "throwing a veil over the speakers." He invited me over to listen to his system with M&K THX speakers behind a Stewart perforated screen. It took about 2 minutes of the T-Rex scene in Jurassic Park to change my mind. Having the CC behind the screen was such a HUGE improvement in the sonic imaging and the cohesiveness and integration of the front soundstgae that I could not wait to change out my non-AT screen for a true AT screen. Moving the CC from below the screen to behind the screen was one of the most beneficial things I have ever done in my theater. I haven't looked back since.

Of course, it is strongly recommended to use capable speakers behind an AT screen. I would never recommend using low sensitivity, low power handling speakers behind an AT screen. (I wouldn't recommend them for ANY application, but a Home Theater with an AT screen would be the worst use-case.) In addition, it's is not necessary to use tower speakers. In fact, using LCR-type speakers behind an AT screen makes the most sense.


The reason most tweeters blow is because the amplifier driving the speakers is driven into clipping. It is important to adequately power the speakers to ensure the amps don't clip. Using low impedance, low sensitivity speakers would not be recommended as they are more power hungry and more likely to drive the amps into clipping. Also, if one uses LCR-type speakers with Bass Management re-directing the bass to the subwoofers, significant amplifier headroom becomes available to drive the speakers and the issue of amplifier clipping is reduced.

Bottom line, I would not recommend putting tower speakers behind an AT screen as the first choice, but if that was the only option, I would recommend using Bass Management and crossovers of the bass to subwoofers. This would eliminate any benefit of the towers, (i.e., deeper LF extension), and make the towers far less beneficial.

Craig
I couldn't say it better.
 

·
Registered
Joined
·
2,929 Posts
One can avoid having to get an acoustically transparent screen (or like in my case, I can't fit the speakers behind the screen due to the bookcases on the sides blocking the location) by implementing a "dialog lift" effect using either a center height speaker (ala Auro-3D uses that can be combined with the function using an active mixer) and/or sending some of the L/C/R channels to front height speakers along with front height (with level controls for each) to raise the soundstage using a vertical phantom image. It works surprisingly well here and even helps avoid needing to raise a center channel above ear level for the back rows (or use risers there) as the front heights act as a phantom center channel for the rear center chair (only chair in the back row). It even works reasonably well without a center height speaker for the left and right chairs off-axis because the bed level center anchors the phantom image part way such that the dialog is off-center no more than 25% even without a center height speaker (diagonal lift). With three height speakers up top, it creates a perfect phantom image for all off-axis seats.

Believe it or not, it also creates a partial frequency vertical array below about 150Hz, limiting ceiling/floor interactions at the bass frequencies. By using arrayed front wides, you can extend that effect to the side walls as well. Given most room reflections are mostly undesirable under only about 200Hz, it's not a bad little room control effect. Combined with tapestries for early reflections and drapes on one side and the back of the room, the room response is greatly improved between 80Hz-150Hz (within 1dB) and nearly perfectly flat (beyond having my LFE levels higher than flat). Having two sets of slightly delayed extra reflections with wide speakers gives "stereo" mode an almost front bipole/dipole type ambient soundstage effect as well (reminds me more of my Carver AL-III dipole ribbons upstairs) with a near 180 degree soundstage (depending on the material) in stereo mode without using DSU/NeuralX/Auro-3D upmixing and despite the front wides being over three feet in front of the front row. I actually enjoy stereo music on my home theater now almost as much as on my Carver ribbon speakers, which was never the case when I had a mere 6.1 setup in the same room with all three speakers just under the screen and no wides.

REW Graph for combined bass response (sub crossed at 80Hz and believe it or not "Main + LFE + Large L/C/R" actually decreased the dip at 50Hz to half the previous value, something the so-called experts say is always bad to do, use your large speakers with the subwoofer instead of only the subwoofer despite the fact that in general more bass sources tend to smooth room modes. The trick is to add the mains back in after Audyssey but you have to have the subwoofer a few dB above flat or you might get peaks as well. By keeping the subwoofer level above the main level, which frankly sounds better anyway if you like bass, it prevents the "double bass" muddy effect by limiting the peak increases to within 1dB here.

Here's a graph of my system with Audyssey Room Correction Turned OFF for both measurements just comparing 2-channel stereo playback versus using arrayed "dialog lift" (add heights) and arrayed front wides (since no side channel signal, just an array in 'stereo' mode). As you can see, 6-channel arrayed stereo tames the room between 80-150Hz and really helping quite a bit down as low as 60Hz in the room. That's without Audyssey engaged even and even appears to have some benefits at higher frequencies (probably due to directional waves being more likely to average out any peaks for a single

measurement). Bright red is 6-channel stereo and dark red is 2-channel stereo.

3048796


Audyssey Turned ON to Reference on Both 6-ch and 2-ch Stereo measurements (Green is 6-speaker Stereo and Purple is 2-speaker stereo):

3048801
 

·
Registered
Joined
·
3,527 Posts
Again, I'm not certain why you think a delay is really needed over such a short distance. Side surround to Sides#1 is what, about 6 feet in 30' room (assuming true front wides are also used at 25%, sides at 50%, ss#1 at 75% and rears at 100% and mains at 0% give or take). You're talking about a 6ms delay at that point. You'll never notice it. I don't notice it one bit with three sets. I mix fronts with sides for a matrixed front wide and sides with rears for a matrixed ss#1. Both are based off the nearest speaker delays, so no timing is off more than 6ms each for 12ms at most from front to rear. I can't hear anything at all. The mixer also gives 3dB separation (matrix) for the channels so it operates both as an array for sides and matrixed in-between channels. You can adjust the mix level amounts relative to base channel amounts to compensate for the precedence effect of any timing difference and easily put the center point anywhere you want it for a given row while keeping the average channel level the same for all. I put the sides to phantom image along side each row despite the speakers actually being between rows. Everything pans smoothly around the 24' room.
With stereo, although I can't directly hear smaller differences in time alignment, I can still perceive the effect it has on imaging. You may notice it less with line source ribbon speakers and matrixed wides, two factors that will reduce the effect of focused imaging. I have point source LRs, no matrixed wides, and very strong, focused imaging. It sounds like you enjoy a big, enveloping sound - nothing wrong with that. Nonetheless, even when time alignment isn't directly recognizable, I assume that phantom imaging for surround speakers work similar to LR imaging.

But anyway, to answer your question, that's why I raise the issue. I'm assuming that time alignment for all speakers will proved the best, most focused phantom imaging between speakers. Granted, I'm sure this is less perceptible behind you and when speakers are spaced more closely. I'm not asking how to cut corners, though. I'm asking what is the best.
 

·
Registered
Joined
·
2,929 Posts
With stereo, although I can't directly hear smaller differences in time alignment, I can still perceive the effect it has on imaging. You may notice it less with line source ribbon speakers and matrixed wides, two factors that will reduce the effect of focused imaging. I have point source LRs, no matrixed wides, and very strong, focused imaging. It sounds like you enjoy a big, enveloping sound - nothing wrong with that. Nonetheless, even when time alignment isn't directly recognizable, I assume that phantom imaging for surround speakers work similar to LR imaging.

But anyway, to answer your question, that's why I raise the issue. I'm assuming that time alignment for all speakers will proved the best, most focused phantom imaging between speakers. Granted, I'm sure this is less perceptible behind you and when speakers are spaced more closely. I'm not asking how to cut corners, though. I'm asking what is the best.
My ribbons are in my upstairs music room. I use PSB T-45s for the mains in the home theater and B15s for the front heights, front wides and side surrounds (same drivers as T-45, only in monitor form). I've got X1T towers on the rear wall (fit in the same space; updated version of same drivers) and CS500 (the monitor version in a ceiling mount) on rear heights with S50 (binaural side wall version of T-45/B15 drivers) for top middle/surround height. They're pretty focused.

Arrays do not affect "focus" to my knowledge. They move the phantom image location. They're detected as arrays when the delay difference is under about 25ms. After that point, you start to hear a delayed echo. Most PA type systems don't add delay unless the farthest speaker is over 25 feet away (7.62m), although possibly as low as 16 feet (1m). It's altered also by relative level versus the precedence effect (the latter is tied to timing as the sounds tend to come from the closer speaker in time unless the levels/balance are shifted). So I alter the balance to put the phantom center where I want it (using the front wide mixer (which is Mains + side surround) for the side surround "location" for the first row and the mixer for the surround#1 speaker (side surround + rear speaker) for the 2nd row. It's "location" is easily adjusted with the balance for the mixer. I hear no difference in focus and I don't notice any reverb either. From what I've read the distances are simply too small (under 1m) to be noticed as any kind of delay effect other possibly slight reverb or a change in the center point due to the precedence effect. The "side surround" location is along side the front row when sitting there and just in front of and along side the 2nd row when sitting in that row. I can move it anywhere from the front wide speaker for the first row to the side surround location by altering the mixer controls and from the side surround to the rear surrounds with the other mixer. I think all delay controls would noticeably/audibly do is move the center point of the phantom image relative to the 2nd row center chair front-to-back slightly relative to the balance.
 

·
Registered
Joined
·
3,527 Posts
My ribbons are in my upstairs music room. I use PSB T-45s for the mains in the home theater and B15s for the front heights, front wides and side surrounds (same drivers as T-45, only in monitor form). I've got X1T towers on the rear wall (fit in the same space; updated version of same drivers) and CS500 (the monitor version in a ceiling mount) on rear heights with S50 (binaural side wall version of T-45/B15 drivers) for top middle/surround height. They're pretty focused.

Arrays do not affect "focus" to my knowledge. They move the phantom image location. They're detected as arrays when the delay difference is under about 25ms. After that point, you start to hear a delayed echo. Most PA type systems don't add delay unless the farthest speaker is over 25 feet away (7.62m), although possibly as low as 16 feet (1m). It's altered also by relative level versus the precedence effect (the latter is tied to timing as the sounds tend to come from the closer speaker in time unless the levels/balance are shifted). So I alter the balance to put the phantom center where I want it (using the front wide mixer (which is Mains + side surround) for the side surround "location" for the first row and the mixer for the surround#1 speaker (side surround + rear speaker) for the 2nd row. It's "location" is easily adjusted with the balance for the mixer. I hear no difference in focus and I don't notice any reverb either. From what I've read the distances are simply too small (under 1m) to be noticed as any kind of delay effect other possibly slight reverb or a change in the center point due to the precedence effect. The "side surround" location is along side the front row when sitting there and just in front of and along side the 2nd row when sitting in that row. I can move it anywhere from the front wide speaker for the first row to the side surround location by altering the mixer controls and from the side surround to the rear surrounds with the other mixer. I think all delay controls would noticeably/audibly do is move the center point of the phantom image relative to the 2nd row center chair front-to-back slightly relative to the balance.
I hear what you're saying. I partially agree, since I get a very focused image from my LRs - virtually indistinguishable from sound coming from the center. It doesn't seem like that should be the case, but it seems to be. But that image does collapse if you start offsetting the delay on one speaker. It still come from the general location, but the focus is gone. In fact, with the introduction of auto time-alignment in preamps, the imaging improvements were the first ting I noticed.

Electrically mixing the same signal with different delays is a separate issue, though. That will cause audible phase problems. So, if a sound is panned between the rear and side surrounds and you matrix mix rears and sides, you will likely experience phase issues. That can be avoided by using matching delay settings.
 

·
Registered
Joined
·
3,675 Posts
I think the OP's (and most of us, actually) eyes have glazed over, congratulations. Maybe take your conversations of time-aligning, matrixing, and other esoteric Pro Audio discussions to another thread.
Why use DSP modes like "dialogue lift" if proper positioning takes care of things?
The OP has enough space for an identical speaker on an identical plane for his centre speaker. This is the acoustic ideal, be it bookshelfs, LCRs, or towers. Using bass management properly eliminates the need for speakers to play (much) below the crossover frequency. Why is this so hard for some of you to understand?
 

·
Registered
Joined
·
2,929 Posts
I hear what you're saying. I partially agree, since I get a very focused image from my LRs - virtually indistinguishable from sound coming from the center. It doesn't seem like that should be the case, but it seems to be. But that image does collapse if you start offsetting the delay on one speaker. It still come from the general location, but the focus is gone.
You are talking about two speakers interacting with different (stereo) signals. An array has more than one set of identical signals. For delays under 25ms, you get a single phantom image between the two that pulls to the nearest speaker (time delay included). All I hear is a bird flying around the room in the Amaze demo. It doesn't go in/out of focus as it flies around the room. For the stereo array, I get a more "focused" (clearer sounding vocals) sound with 6 speakers than with 2, possibly because of the partial line source effect canceling out some of the room interactions. I can push one button on a remote to turn the effect back to 2 speakers and the difference is unmistakeable. It's clearer sounding with 6 speakers than two with a wider and deeper image as well.

If you're really worried about it, however, just buy a mixer that has channel delay controls on it and you can set the delay to match the phase of the other speakers as precisely as you want it. I tend to think it's overkill for a small room, but it's not hard to find one with delay controls (might use 1/4" or 1/8" jacks, though so you'd need adapters to match to RCA).

Electrically mixing the same signal with different delays is a separate issue, though. That will cause audible phase problems. So, if a sound is panned between the rear and side surrounds and you matrix mix rears and sides, you will likely experience phase issues. That can be avoided by using matching delay settings.
What phase issue am I supposed to be hearing, exactly? When the bird flies around the room in the Dolby Amaze demo, what do I listen for? There's no electrical delay differences in the output. If you want them, you have to add them externally (like on the mixer). Otherwise, the only delay is for a given channel from the AVR perspective). But channels array to the same channels (i.e. any right front output in the front wide will array with the right main speaker, not the side speaker, only the side speaker part would array with another side surround that has the same signal). The delay difference relative to the listener affects which speaker is perceived as the source of the sound (precedent) and then modified for a reverb effect or large enough level difference. As long as the delay is under 25ms, regardless you won't hear an echo. <25ms = perceived single reflection as very short-lived reverb, >25ms = perceived delay/echo.

I think the OP's (and most of us, actually) eyes have glazed over, congratulations.
If you don't like what I have to say, ignore it. You're neither the person I was responding to or the original poster.

Why use DSP modes like "dialogue lift" if proper positioning takes care of things?
Some people's rooms won't/can't do it that way. I have bookshelves built into the walls that prevent putting three full range speakers at screen height. It cost me $29 total to buy an active mixer to do the dialog lift effect with matched speaker that were ALREADY THERE for front height speakers. Even if I could use an audibly transparent screen, why would I want to spend a small fortune to get large enough towers (or stands) to put the mains at screen height plus buy a more expensive acoustically transparent screen (that needs EQ to get flat response) when the problem is solved for the price of a $29 mixer if you have front heights and/or a center height already installed? It even works for 85" OLED TVs as the phantom image doesn't care that there's a solid object sitting there.

The OP has enough space for an identical speaker on an identical plane for his centre speaker.
So everyone else reading the thread does too? Maybe when he realizes he can save a few thousand dollars to put towards something else more worthwhile when a $29 mixer does the same damn thing he'd reconsider for that matter. It's a waste of money if you already have the speakers there to simulate a phantom image in the middle.

This is the acoustic ideal, be it bookshelfs, LCRs, or towers. Using bass management properly eliminates the need for speakers to play (much) below the crossover frequency. Why is this so hard for some of you to understand?
Acoustic ideal is a thought concept, not reality. The reality is an array of two identical speakers sound like ONE speaker in a different phantom location. Commercial cinemas have been using arrays since their inception, particularly for surround sound.
 

·
Registered
Joined
·
3,527 Posts
For the stereo array, I get a more "focused" (clearer sounding vocals) sound with 6 speakers than with 2,...It's clearer sounding with 6 speakers than two with a wider and deeper image as well.
We're using the terminology differently.

When I speak of a "focused" image, I don't mean clearer sounding vocals. That's a separate issue. I mean that the perceived sound source is a point, as opposed to a large sound source. When authoring Atmos positional audios, recording engineers actually have a size adjustment that will expand the sound source across more speakers to create larger sound sources. 3D positional audio in video games works in a similar way, whether for stereo or multichannel. A voice or a gunshot emanates from a point, but a waterfall emanates from a volume. So, a wider and deeper image has less focus.

If you are a fan of wider and deeper imaging, that's fine. With stereo playback, speakers and rooms are often designed with that goal in mind. Your speakers and room are your surround sound processors. Ignoring personal preference for a moment, with Atmos, the intent is to have multichannel playback system that is capable of reproducing tight focus when called for and envelopment when called for.

What phase issue am I supposed to be hearing, exactly? When the bird flies around the room in the Dolby Amaze demo, what do I listen for? There's no electrical delay differences in the output.
If you mix the same source with itself but add a delay to one of the signals you will hear phasing. You can get guitar FX pedals called "phasers" that do that intentionally.

If you let an AVR set delays to each speaker, it will time align all speakers, in order to produce the best imaging and minimize phasing. But, if you mix two of the channels that already have different delays set, when a source pans between two speakers, the mixed channel will end up with the same source mixed with itself but with slightly different delays. So, it will phase and detract from imaging. The audibility of the phasing depends on the content, but phasing from electrical mixing is much more audible than phasing from two speakers mixing in the room.
 

·
Registered
Joined
·
2,929 Posts
@rcohen - The bird in the amaze demo is the same "size" image with or without the extra speakers. Whatever larger mage you're referring to, it's clearly minimal. By wider with the extra array in stereo, I mean spacing for out of phase images, not increased size of a sound object. The right speaker test tone is still at the right speaker not the right wide. A singer's voice dies not get "larger" in size when I turn on the additional speakers. The extra front arrivals are the same as the first wall reflection, but without the residual repeat bounces. You get more ambience, but with a clearer image instead if a less clear image with fewer room interactions.

All I'm saying us don't knock it until you've heard it. What theory suggests to someone isn't the same as reality. I'd turn the extra speakers off when guests aren't using the extra chairs if the sound was worse. I don't pretend to easily hear ss#1 from the front row (it does more for rows two and three), but the sound is clearer and more ambient with front wides and heights getting some sound and the frequency response graph shows better response as well so I'm happy with it.

I'm not sure I'd prefer discrete front wides as much material and upmixers don't even use them and I'd lose that great sounding stereo music playback.
 

·
Registered
Joined
·
3,527 Posts
@rcohen - The bird in the amaze demo is the same "size" image with or without the extra speakers. Whatever larger mage you're referring to, it's clearly minimal. By wider with the extra array in stereo, I mean spacing for out of phase images, not increased size of a sound object. The right speaker test tone is still at the right speaker not the right wide. A singer's voice dies not get "larger" in size when I turn on the additional speakers. The extra front arrivals are the same as the first wall reflection, but without the residual repeat bounces. You get more ambience, but with a clearer image instead if a less clear image with fewer room interactions.

All I'm saying us don't knock it until you've heard it. What theory suggests to someone isn't the same as reality. I'd turn the extra speakers off when guests aren't using the extra chairs if the sound was worse. I don't pretend to easily hear ss#1 from the front row (it does more for rows two and three), but the sound is clearer and more ambient with front wides and heights getting some sound and the frequency response graph shows better response as well so I'm happy with it.

I'm not sure I'd prefer discrete front wides as much material and upmixers don't even use them and I'd lose that great sounding stereo music playback.
I think you're right. I just have to experiment to see what works best for me. A matrix mixer DSP would be useful to have handy for a variety of purposes (surrounds, shakers, karaoke.)
 
41 - 59 of 59 Posts
Top