Do Front Height Speakers need to Tone and Timber Match L/C/R?

All of your speakers should be tonally matched if possible.

I would try to find speakers that match the fronts but maybe that's just me.

No opinion on heights (never tried them) but for multichannel music surround "matching" is actually not very critical.

When I was running heights, I had some small Polk OWM3 speakers for the heights from a previous setup. In general, even when using very different speakers for LCR (Focals), it was perfectly fine. Since it's mostly ambient information in the heights, you don't really notice the mismatch. However, as soon as a voice would appear in the height channels, it was very jarring and very obvious that the heights didn't match the LCR.

Heights are cool but, they are difficult to do right because it's not always easy (or even possible at all) to mount a matching LCR speaker that high on the wall.

And there are of course many who hold the exact opposite opinion. Perhaps the one thing that we can all agree on, however, is that timbre-matching can't hurt in terms of sound quality.

Most of them haven't actually done the comparison.

Between what and what?

Sure, if the surrounds are good quality speakers you can have an enjoyable listening experience even if they don't match. And if the music is, say, a classical or jazz recording where the surrounds carry mostly room ambience, a timbre mismatch might go unnoticed.

Mut studio surround mixes that treat all five channels equally can absolutely suffer from mismatches. If the same singer's voice sounds different; if an instrument is doubled front-to-back; if the mixer likes his circular pans; if instruments are being positioned between a front and surround speaker... these are all situations where timbre-matching is necessary if the goal is to come as close as possible to reproducing what the mixing engineer was hearing. Because he was almost certainly using five identical speakers.

True, but if some are oak, some are walnut, and some are birch they just won't look all that good together in the same room. .

I guess the difference between timber and timbre remains lost.

On what do you base that? Even if this were true, it stands to reason that surround channels that are treated equally by some mixers in some pieces (or even some movie soundtracks, for that matter), and contain some of the same type of content as the front channels, would be best reproduced by surround speakers that are matched in timbre with the front speakers. I think it is more likely true that some who have done the comparison and disagree may have only listened to titles that put only ambiance and reverberation in the surround channels--either that or timbre-matching in general is just not very critical to them. Admittedly, I didn't think that having a fully timbre-matched system would make a noticeable difference until I installed my own and heard the right material on it.

The main point I was trying to make is that timbre-matching the surrounds to the fronts can't hurt and may help (does help, as far as I'm concerned), if only for some titles or on occasion. For those "on the fence" who do not wish to do their own experimentation but still don't want to miss out on anything, I think that an effort should be made to timbre-match all of their speakers--compromise only if you must. I suppose it's analogous to having a subwoofer that packs a wallop below 30 Hz--you may not always need it and you would not even know what you're missing if you didn't have it, but it's still a good thing to strive for.

Hilarious.

As is the pronunciation, or else people would be writing "tamber" instead. Gee, I hope we've been discussing the right topic all this time, knock on timber.

Between "matched" surrounds and "mismatched" surrounds, for multichannel music reproduction. (I have no opinion as to the relative difference with movies, simply because I do not care.)

(Note: "matched" and "identical" are synonyms here. Likewise, "mismatched" and "not identical" are synonyms. There are some rare exceptions. For example, one could reasonably call the NHT Classic Four and Classic Three "matched" even though they're not the same, but that's because the former is the latter in a taller cabinet with a subwoofer added.)

If the surrounds carry anything but signals designed to increase spaciousness or "special effects" type stuff (trumpets on the balcony, etc.), then the recording is IMO poorly done.

There are a few discrete multichannel disks that I much prefer to listen to two channel expanded over DPL2, because the discrete multichannel mix does asinine things such as put the drum kit's cymbals in the surrounds. While keeping the rest of the drum kit up front!

First, that is not my experience. Second, see supra.

The surround channels are not equal. Last time you went to a concert, did you turn 90deg or 180deg away from the musicians?

I'm not sure what this term "timbre matching" means, besides the marketing implication that one should buy all one's speakers from the same source.

There are two types of speakers: the same speaker, and different speakers.

"Matched" is just so much crap. A horizontal speaker is never going to "match" a vertical one, and the bidirectional out-of-phase things that less informed people call "dipoles" are never going to match monopoles.

There are two kinds of surround recordings. One kind is intended to put you in the audience. The other kind is intended to put you in the performance.

If you are only willing to discuss the first kind, I will concede all of your points and withdraw. And find someone else who wants to talk about the much more interesting second kind.

Any of the Steely Dan or Donald Fagan DVD-A's.
Dire Straits, Brothers in Arms SACD
Seal, Best, DVD-A
The Beatles, Love, SACD
The Who, Tommy, SACD
Fleetwood Mac, Rumors, DVD-A
Many of the AIX REcords discs have *both* presentations and you can change perspective from "audience" to "on-stage." http://www.aixrecords.com/

Many Others, (but likely none approved by DS-21)

Craig
Edited by craig john - 2/26/13 at 6:32pm

If the former are really accurate they're dual mono L/R, no center, with ambient reflections from the surrounds. If the latter are really accurate you're getting pretty much the same content from all the sources, that's what you hear on stage though the monitors. I haven't really heard either type, as I only listen to music in my car or in my shop, always stereo. Like most musicians I know I don't make a big deal about listening to music, it's kind of like going in to work on your day off.

Well, that's a different artistic style with regard to mixing, and other people may want to hear such titles as they were intended to be heard. This means that your original assertion is not generally applicable, and pertains only to the type of material that you prefer to listen to. That's like saying there is no need to match the surrounds to the fronts because you only watch movies in Pro Logic. What about others who wish to play the 5.1 soundtracks instead?

Who says that multichannel music must always and only ever attempt to replicate sitting in a concert hall? OK, you do, but many others do not.

It means that two speakers sound alike, which makes for a consistent soundstage and allows for imaging between the speakers.

Oh please.

Having identical speakers all around obviously makes for the best match (that's why I did this with my current system), but it's not an all-or-nothing proposition. At the very least, speakers that sound similar enough are less likely to distract the listener as sounds move between or are played back simultaneously on two or more speakers.

No argument here, except that I know what a real dipole is but I still call these speakers "dipoles" so that other people can understand what I'm saying. I agree that there is no point in trying to timbre-match dipoles because it can't be done to the point of having any practical benefit (one reason I never recommend dipoles). If somebody wants those, then they can buy them from any other series or manufacturer, as far as I'm concerned.

It totally depends on the material you happen to be listening to at any given moment. Sometimes it matters, and sometimes it doesn't, much like my analogy of having a subwoofer that goes deep and hard--most of the time you won't need it, but when you do, you'll be glad that you have it (otherwise you won't know what you're missing). Whether it's worthwhile depends on a lot of factors, including the situation one finds oneself in. For instance, I've made very few compromises with my system regarding setup, but it's only a 5.1 system (because my sofa is against a wall), and I acknowledge that I am missing out on something, but it's a compromise that I've chosen to live with, due to various factors. Similarly, as much as I've championed having fully timbre-matched systems as well as vertically-oriented center speakers (for the best horizontal dispersion), I've recommended numerous systems that compromise on both, usually due to budgetary and/or physical limitations. The key is to know what is ideal and why; understand what compromises may need to be made, how severe they really are, and why; and then to create the best system you can, based on that.

The question you're asking here is a valid one. Most of what we've discussed in this thread concerns the content of discrete channels, but what's really relevant is the content of the derived channels. I believe that timbre-matching is of minimal importance for surround channels that are derived from a two-channel stereo matrix, because so much of the content is ambiance and reverberation (I still think there is a general benefit, but it's of secondary or tertiary importance). However, channels like rears and heights are either discrete (as rears are with true 7.1 channel soundtracks) or derived as "center" channels between two discrete channels, and therefore may have much of the same content. That's one side of the story, and the other is that rear (or back) channels are usually derived from discrete surrounds that probably share a fair amount of content, whereas the height channels are presumably derived from content that is shared between the surround and front channels, which I'm guessing is much less common. Personally, if I used front height channels (I doubt I ever will, but just suppose), then I would want to timbre-match them out of principle, just in case, but it could definitely be argued that it is less important than timbre-matching the (side) surrounds with the fronts (or even the rears with everything else), however important that may be to you, individually, to begin with.

I'll buy that some of it may be imagined, as with most anything involving perception (sort of like the placebo effect), but there are aspects of what we're discussing that are clearly audible, particularly when the same voices and instruments are reproduced by different speakers, either at the same time or in quick succession. Heck, I recently changed the orientation of my surround speakers and got a noticeable improvement, even though they're both identical to my fronts (due to a combination of known mounting and dispersion issues that I finally corrected); I'm pretty sure that it's a real improvement because I changed them one at a time, as well as back and forth a couple of times, and listened to material that critically tests timbre-matching (which I know because I experimentally isolate channels to find out what's in them).

There are all kinds of people, that's true. But justifying price is dependent on more than just sound quality, such as how much money you have, how much you are allowed to spend in order to keep peace around the house even if you have more than enough (that would be me ), and in many cases decor (that would not be me, except that I can't move the sofa forward like I want to).

You need more experience to know the latter, but if you can trust the experience of others, then you may not need to find out the hard way.

There is nothing wrong with saving money as long as you know what you're getting into. And do you know what? This is just a guess on my part because we all hear a bit differently anyway, but I expect that the Ascend CBM-170 SE you're considering will match your Philharmonic front speakers just fine--not absolutely perfectly, and of course not at the same level of performance (which is probably unnecessary), but they're all very neutral, accurate speakers by design, and would likely work well enough together, despite being completely different designs from different manufacturers. Note that this is not always the case--had you said that you were going to try the Polk RTiA3, for example, I'd immediately let you know that it would be a poor timbral match.

That's right, or in the case of some movies, you're placed in the middle of the action, which happens to include music playing all around you. Yes, it's fake, like virtually all movies are, but some people manage to enjoy it anyway.
Edited by Robert Cook - 2/27/13 at 7:29am

Anything mixed by Steven Wilson, including both his solo albums and those of his band Porcupine Tree. He has also been working with Robert Fripp remixing King Crimson in surround, and did the Thick as a Brick 40th Anniversary mixes.

There is a forum here on AVS that focuses on this topic: http://www.avsforum.com/f/112/surround-music-formats

BIA is the SACD I alluded to above. Another similarly stupidly-recorded disk is the Cream reunion concert on Blu-Ray. The drum kit is on the screen but it pans through all the speakers.

That may fool people who don't pay attention when they listen, I guess.

Speakers sound alike when they're (above the modal region, at least) exactly the same. Otherwise, "timbre matching" is just feckless marketing BS. Fortunately, it's not required in back. Besides the artistic issues, there's the issue of how humans perceive sound. We're just not as good at distinguishing the quality of sounds to our sides and back. Frankly, if someone replaces your back speakers and you don't see it, you'd likely never notice they were changed.

You mean only match. Anything else is a mismatch. It is an all-or-nothing, for the most part. (The exception is that the speakers are identical above the modal region, but some of them have additional bass support, such as the NHT Classic Three/Classic Four mentioned above.)

Fortunately, mismatched surrounds are perceptually unimportant. Front three is critical, though. Otherwise, two channel is a better option than a different left, different center, different right, etc. It doesn't matter which speaker doesn't match the other two. It's wrong regardless.
Edited by DS-21 - 2/27/13 at 7:39pm

Don't be so bloody literal-minded. Recorded music with video doesn't have to sound the way it looks. And recorded music without video can sound however the creators want it to sound, including creative use of a five-channel sound stage.

Depends on what you're listening to. You, apparently, simply refuse to listen to anything that would reveal such differences. Because I have no basis upon which to impugn your hearing.
Edited by rdclark - 2/28/13 at 9:07am

They're trying to pay attention to the music/soundtrack, and the closer the match between their speakers, the better they can do that. Any degree of similarity in timbre is better than clearly different-sounding speakers. The question here is whether it is necessary to match, as much as possible, the surrounds to the fronts, and it stands to reason that for those who listen to titles that have similar content in both the front and surround channels--not you, but other people--it is better to try to match than not to.

Well, then there's that modal region, which makes speakers sound slightly different according to their placement, including the ones at the front--oh well, nothing is perfect, but then there are degrees of similarity that still can help.

Only for those who deal in absolutes like you. Fortunately most people can deal with more than that.

Get serious or get your ears checked, because most people can hear just fine all the way around, otherwise headphones, for example, would all sound the same. I can very easily tell the difference between quality surrounds and crappy ones (when high-fidelity sound is played through them), as well as fairly well matched surrounds (to the fronts) from poorly matched. I've installed enough of both types of systems (and used a few myself for years at a time) to know the difference, and lest you think that this is merely some form of confirmation bias, I was quite skeptical until I finally tried it with material that requires matched surrounds for the best presentation. Even before that (I'm talking about at least six years ago), the 360-degree soundstage sounded more coherent to me with matched surrounds, but I initially dismissed this observation because I thought that it wasn't possible for a typical 5.1 arrangement of speakers (with the surrounds way back to the sides) to work so well together, especially since the match is never perfect due to room acoustics--well, my assumptions were proved to be incorrect by newer observations.

No, unless you have absolutely ideal conditions, there are always degrees of similarity. My current 5.1 system uses the same 2-way, 2-driver bookshelf speaker model all the way around (except for the subwoofer, of course), including the vertically-oriented center, and they don't sound exactly alike in every way, due to placement and room acoustics, even after the recent improvement I've made by reorienting my surrounds (I turned them on their sides for the best dispersion across the seats and to move their tweeters a bit farther away from the back wall they're mounted on; the fronts are all oriented vertically). Even so, titles that utilize the surrounds much like they do the fronts sound a lot better the more closely matched the surrounds are.

Your position on this matter is the extreme one--virtually every aspect of it--so it's the one that requires extraordinary proof. I agree that using identical speakers is better than using different ones that a designer attempted to match in timbre or "voice"--that's why I use identical speakers all around. But I disagree that the match has to be perfect or else nothing because it's never perfect anyway--I can always hear some differences, but if the match is close enough, then our brains can deal with it fine, and the same goes for "timbre-matched" speakers. I also think that matching the fronts, in the grand scheme of things, is relatively more important than matching the surrounds to the fronts, but with the 5.1/7.1 content available today and plenty of people who wish to hear it as intended (not you, I know), I think it's obvious that having surrounds that match the fronts, at least to some degree, is an advantage. On the other hand, your argument that humans just can't hear well enough to the rear and even to the sides for this to matter at all is preposterous to anybody who has normal hearing and life experience in using their ears in the real world, putting aside speakers for the moment.

I like this Robert Cook guy. He makes a lot of sense.

My "surrounds", (and I use Wides as well as Sides) are all excellent "timbre-matches" to my fronts. They use similar silk-dome tweeters and identical mid/woofers. The crossovers are different, and the cabinets are different, but the speakers are all "voiced" by the manufacturer to be very good timbre-matches. Obviously, the placement of the speakers into the acoustics of the room plays a huge role in the perceived frequency response, but the "starting points" are very similar. Audyssey XT32 then steps in and adds a "target curve" that essentially improves on the timbre similarity between channels.

The net result is a very similar presentation of sounds from the surrounds and the mains. As sounds are placed and panned through the speakers, the timbre of the sound doesn't change appreciably. This results in a seamless, holistic and cohesive surround soundstage.

How the director or sound mixer then uses that surround soundstage to present their art is a totally different and subjective discussion. For the purposes of the design of my reproduction system, I can only control the accuracy of it's layout and implementation.

Craig

I agree there can be artistic reasons for the sound on stage not to correlate with the images on screen (if there is a screen involved). A concert video (aside from multimedia interludes) generally does not present such reasons.

Frankly, the way surround is often used is like the way stereo was often used in the beginning - gimmicky hard pans, ping-pong, etc. Ultimately, with stereo artists realized that the music is better served when the technology is used not to be an exhibit unto itself but as a means of presenting the performance. With surround, we're still getting there.

That said, I would love to hear what a band like Radiohead try out multichannel.

Regardless, none of that has any bearing on the reasons why rear speakers needn't match the front speakers. (And why the front three must match for a coherent acoustic scene.) Human beings vary in their sensitivity to aspects of sound such as frequency balance and timing/location cues based on where the sounds are in space. Sensitivity is just higher for sounds in the forward hemisphere (sweep from temple-to-temple including the eyes) than in the rearward hemisphere, and there's evidence for a declining ability to locate differences in height as well.

There is no evidence beyond your assertions based on sighted listening suggesting that is in fact the case.

You don't seem to know much about loudspeaker design.

The modal region is best handled by multiple subwoofers. Which, like the front three mains compared to the surrounds, needn't be identical to one another. (Like identical surrounds, identical subs doesn't really hurt anything, but just as using identical (I'm not going to address the silly case of "timbre matching," which is just a marketing construct) surrounds is an inefficient allocation of resources, using identical subwoofers is usually quite wasteful as well, given that in a properly-optimized system the subs will be playing at very different levels from one another.

I wasn't aware that "deal in absolutes" is a synonym for "can hear."

Check the literature. You can start with Toole's summary in Sound Reproduction before moving into cited articles, and beyond.

You're just plain wrong on this point.

Inapplicable analogy, because headphones are drivers firing directly into the ear-canal. (Let's leave out examples such as the AKG K1000, though given that they have drivers angled in front of the ears they actually support my position even though they have some pretty severe sonic flaws unrelated to imaging)

Oh, so you're selling things. That explains your breathless defense of marketing idiocy.

Obvious visual match, that I'll grant you. Otherwise, you miss my point that the "match" is in fact uncritical. A set of four KEF Q-Compacts in the surround side and rear spots will sound exactly the same as a set of four Tannoy System 800s, even though if used up front the System 800s are clearly vastly superior speakers. (Admittedly, I don't deal in speakers with obvious midrange sound power problems, such as the 7" 2-way with flush-mounted tweeter, and I wouldn't for surrounds either. Perhaps it matters, perhaps it doesn't. My subjective preference is to support competent loudspeaker design over incompetent loudspeaker design regardless. Others may differ)

"The" subwoofer? You'll hear more of a difference in the overall soundfield moving a foot in any direction than you will from differences in the surrounds, because one subwoofer is simply insufficient for high fidelity upper bass reproduction in a small room. (And any room in a domestic home is, for acoustic purposes, "small." "Small room" is a term of art in the literature to distinguish such spaces from clubs, performance halls, etc.) One needs multiples, and one gets better results by setting them up properly.

I find it a little funny, but also a little sad, when people harp on non-issues without taking care of the things that actually matter first.

Actually, my position is just the consensus produced by serious research into the matter. I have no pecunary interest in any of this stuff - i.e. I'm not selling anything. So I have the luxury of seeking truth rather than mindlessly spewing a line. You would do well to consult the literature on human sensitivity to timbral differences by angle before spewing further nonsense.

Thank you for expressly proving my point that surrounds can be vastly different from the mains without sonic penalty. After all, your surrounds are vastly different from your mains, regardless of any assertions from marketing folks like the Triad marketing guy you like to cite - interesting that you tend to citre marketing folks, whereas I tend to cite researchers who have been published in JAES, no? - that there is some sort of match, Your surrounds have, by your own discussion, different crossover circuits and different cabinets. (More specifically, are the baffle widths different? That matters a great deal.)

Now, if one's feels better by matching (i.e. using identical speakers, because anything non-identical is by definition non-matching) surrounds to mains, then do it. Or if one is insufficiently critical a thinker, and buys lines fed to the gullible about "timbre matching," then let that be the crutch that allows one to use mismatched speakers where it doesn't matter, i.e. behind the listener. After all, the goal of this stuff is to end up with something that one is happy with, right?

But those are all analytically distinct points from saying that such things actually matter, and in fact not materially different from saying, "if overpaying for Kimber/Audioquest/Cardas/whatever wire makes you happy, whatever. Enjoy."

Because there's no actual real existing evidence that matching surrounds are sonically important. A few people blathering ignorantly after sighted listening doesn't mean a thing.

PS: I derive no revenue or any other material benefit at all (including discounts, "long term loans" of gear, etc.; my integrity means more to me than a few shekels or free parts would) from "clicks" on my little blog. Obviously, if I did derive any revenue from it, common human decency would not allow me link to it without a disclosure stating that I profited from clicking on the link. But as it is the number of "clicks" is not important to me, so don't trouble yourself about it.
Edited by DS-21 - 3/1/13 at 9:54am

And yet there are mixing conventions for studio-recorded popular music that arose during that era that have become so standardized that we no longer notice them. Drum sets that are as wide as the entire stage. Vocalists that are always exactly in the center. The whole notion of mostly bilaterally symmetrical left/right balance. We've been listening to music mixed this way for decades, and nobody thinks it's gimmicky. Now sure, a good deal of this came from the requirements of disc mastering, but nobody can argue that it's in any way naturalistic. Studio recordings are artificial constructs, and that fact is reflected in the way they're mixed -- regardless of the number of channels they're mixed for.

Probably much like the Flaming Lips (http://www.amazon.com/Yoshimi-Battles-Robots-Deluxe-Edition/dp/B0000B1A2O), in spirit anyway.

[much snippage]

I don't violently disagree with your thesis when it comes to recordings whose surround channels carry mostly ambience, whether it's the architectural acoustics of a concert hall in a live symphony recording, or the rustling of trees in a field (probably pulled from a sound effects library anyway) in a movie. As long as the surround speakers are "high fidelity" in overall performance, an exact match with the front array will be unlikely to bring a noticeable improvement in the listening experience. Different speakers will still sound different, but it's impossible for the listener to know which ones sound right 99% of the time.

For five-channel music mixes, the requirement to achieve accuracy is simply different. This is something I know from experience. There are many, many recordings where matched speakers produce a noticeably different and (I maintain) better listening experience, because it's clear that the mixer intended for a given sound to have the same timbre regardless of which channel it happens to be mixed to at any given moment. And I disagree with your thesis that matching isn't important because people can't hear the difference. I agree that we hear differently front-to-back (I did a lot of binaural recording in the Seventies, and learned a lot about such aural phenomena). But the engineer has the same limitations, and mixes accordingly. If he puts an identifiable individual instrument in the front left channel and a second part played on the same instrument in the right surround channel, he will make darned sure they sound like the same individual instrument and not just two generic versions of it. Matched speakers are the only way they will still sound the same when you play it at home.

These are not important distinctions -- and this is a niche application -- for most users. I would agree that unless studio-mixed surround music is an important application for the user -- and unless he is also willing to meet the additional requirements for speaker placement and potential room treatment that it takes to get five speakers to sound sufficiently similar to one another -- surround speakers should be selected based mostly on how well they work when placed as needed, how well they match up electronically, mounting considerations, and inherent sound quality.

And none of this helps the OP regarding front-height speakers, which I have never used. Although I would imagine that if they share content with any of the other front speakers it would be very important for them to sound the same as the other front speakers; if they sound different, wouldn't they fundamentally change the sound of the front stage? Any sound shared in common between any two of the front speakers would shift in tonal balance if the speakers aren't all the same.