Poll: speakers parallel or inclined to the listener?

[wgscott]

The speakers flank the fire place which is at a 45 degree angle with respect to the corners, so effectively the speakers are already pointing 45 degrees from their nearest walls before I toe them in.

[mayhem13]

The speakers flank the fire place which is at a 45 degree angle with respect to the corners, so effectively the speakers are already pointing 45 degrees from their nearest walls before I toe them in.

 

IIRC, the track content you mention is all handled by your CM7's midrange driver which is up to 4khz with a fairly steep filter. Try raising your speaker up to where the vertical listening axis is where the mid cone edge overlaps the tweeter faceplate. That mid is one of the best in the business when filtered properly.

[wgscott]

I don't follow, but I want to understand what you mean. Should I lift the speaker up on some platform so that the mid-range cone is at the same height as my ear is from the floor?

[mayhem13]

I don't follow, but I want to understand what you mean. Should I lift the speaker up on some platform so that the mid-range cone is at the same height as my ear is from the floor?

 

Yes. Have someone help measure the distance to the ground from you ear canal while you're in your most casual listening position. Then adjust the speaker height to the same distance from the ground to the top rim of the mid driver. The measurement isn't critical to the mm.

[Bob Stern]

I don't follow, but I want to understand what you mean. Should I lift the speaker up on some platform so that the mid-range cone is at the same height as my ear is from the floor?

A more aesthetic method for aiming the midrange driver at your ears would be to raise only the front of the speakers slightly to tilt the cabinet front surface so that a line from your ear to the midrange driver is perpendicular to the front surface. Thick felt pads might suffice.

 

Before doing this, you should look at the vertical off-axis response measurements of your speaker in Stereophile. Because of the phase relationship among the three drivers, the frequency response may be more irregular if your ears are below the axis than if they are above the axis of the midrange. In that case, it would be safer to err on the side of aiming the speakers too low rather than too high.

 

Incidentally, the latest issue of Stereophile includes Kal Rubinson's report of applying room EQ to his expensive B&W 801 Diamond loudspeakers. He reported that, before equalization, they had a depressed midrange and elevated treble like your B&W's. So this frequency response appears intentional.

[semente]

A more aesthetic method for aiming the midrange driver at your ears would be to raise only the front of the speakers slightly to tilt the cabinet front surface so that a line from your ear to the midrange driver is perpendicular to the front surface. Thick felt pads might suffice.

 

Before doing this, you should look at the vertical off-axis response measurements of your speaker in Stereophile. Because of the phase relationship among the three drivers, the frequency response may be more irregular if your ears are below the axis than if they are above the axis of the midrange. In that case, it would be safer to err on the side of aiming the speakers too low rather than too high.

 

Incidentally, the latest issue of Stereophile includes Kal Rubinson's report of applying room EQ to his expensive B&W 801 Diamond loudspeakers. He reported that, before equalization, they had a depressed midrange and elevated treble like your B&W's. So this frequency response appears intentional.

 

I agree.

Things have been getting worse since the replacing of the Matrix series with the Nautilus.

 

R

[wgscott]

A more aesthetic method for aiming the midrange driver at your ears would be to raise only the front of the speakers slightly to tilt the cabinet front surface so that a line from your ear to the midrange driver is perpendicular to the front surface. Thick felt pads might suffice.

 

Before doing this, you should look at the vertical off-axis response measurements of your speaker in Stereophile. Because of the phase relationship among the three drivers, the frequency response may be more irregular if your ears are below the axis than if they are above the axis of the midrange. In that case, it would be safer to err on the side of aiming the speakers too low rather than too high.

 

Incidentally, the latest issue of Stereophile includes Kal Rubinson's report of applying room EQ to his expensive B&W 801 Diamond loudspeakers. He reported that, before equalization, they had a depressed midrange and elevated treble like your B&W's. So this frequency response appears intentional.

 

Yes, they deliberately design it in, which then raises the question of whether they actually might happen to know better about what they are doing and how to do their jobs than I do.

[semente]

I think that one should not treat manufacturers as some sort of omniscient entity but be prepared to question their design decisions and philosophy.

There are things they ignore and mistakes made unintentionally, or "consciously" to cut production costs (incerease profit margins) or to produce some sort of "house sound".

I would go as far as to say that most commercial products can be improved upon...

 

R

[mayhem13]

Voicing.......it's how the product is finalized with a desired or signature sound. You or I may not prefer the B&W sound but it's certainly NOT bad and the apparent success of the company suggests many a listener don't agree with either of us.

 

WGScott....please don't tilt the speaker. That puts the tweeter waaaay of the vertical axis and will tilt the high end response not to mention shift the phase integration between the two drivers. B&W would have stepped the baffle if the crossover required it to align the driver's acoustic centers but since it's not an LR2 network, not desirable.

[wgscott]

It turns out my ear canals are at the correct level if I slouch a tiny bit in the chair, so I won't do anything drastic. I promise.

[cjf]

Reporting back after a week or so of listening to the speakers after correcting the slightly off center (left/right) tilt and slightly different forward rake there was between the two cabinets. As a reminder, the new position now places the speakers basically level in all directions (or at least as level as my patience would allow for while attempting to make said adjustments).

 

After the corrections the speakers seem less picky about were I place my head while seated in the listening position. In other words, the sweet spot has grown beyond the angle at which I can tilt my head left/right/up/down.

 

At least for me, placing my ears level with the midrange units is pretty much impossible considering how tall the B&W 803D's are. I have what I would call a standard height couch as my listening throne which, by eye, places my ears at the same height as the 1st woofer down from the top of the cabinet which is approx 6" below the center of the Mid Range driver above it.

 

Unless I want to deploy a stack of phone books under my hind quarters there is just no practical way to put my ears at the same level of the Mid Range driver let alone the Tweeters.

 

Despite the above "issues" the fairly big B&W's totally disappear in the room. I'm sure part of this has to do with the fact that I'm feeding them enough power to heat a small home.

 

Like they say, there's no substitute for cubic inches, or in this case Wattage to solve individual driver dispersion shortcomings.

[Bob Stern]

Yes, they deliberately design it in, which then raises the question of whether they actually might happen to know better about what they are doing and how to do their jobs than I do.

Loudspeaker manufacturers commonly design non-flat frequency response intended to impress the customer in the showroom, but not necessarily best for long term listening at home.

 

A dip in the presence region and a boost in the upper treble creates the impression of spaciousness.

 

In any event, my wife and I both feel the CM7 gives music an unnatural tonal balance unless this frequency response is corrected by EQ, so I don't believe B&W's design philosophy merits deference.

[Bob Stern]

Things have been getting worse since the replacing of the Matrix series with the Nautilus.

What about the original Matrix (about 1988) which was reported in audio magazine measurements to have elevated bass, which after some investigation was found to be caused by the inductors in the crossover network being oriented in parallel and hence having unintended mutual coupling. Apparently B&W finalized the design based on a mockup in which the inductors were spaced farther apart than in the production model. Stereophile published several reader letters suggesting shielding the inductors with tin cans, etc.

[semente]

What about the original Matrix (about 1988) which was reported in audio magazine measurements to have elevated bass, which after some investigation was found to be caused by the inductors in the crossover network being oriented in parallel and hence having unintended mutual coupling. Apparently B&W finalized the design based on a mockup in which the inductors were spaced farther apart than in the production model. Stereophile published several reader letters suggesting shielding the inductors with tin cans, etc.

 

As I've said, most commercial products' performance can be improved...North Creek is said to produce a much improved crossover for the speakers (I've no experience with).

B&W did change the crossovers on the subsequent editions of the same speaker and Series 3 no longer sported the APOC circuit.

The Matrix 801 Series 2 reviewed by Stereophile has respectable performance in some of the praramters measures but the final result is still marred by floor-bounce, ill-integrated tweeter (response up by at least a couple of dB):

 

http://www.stereophile.com/content/bw-matrix-801-series-2-loudspeaker-measurements

 

And they don't mention the slightly splashy-sounding metal dome tweeter nor the kevlar colouration from the mid driver (I am currently living with a pair of 802 S3s).

 

R

[wgscott]

Loudspeaker manufacturers commonly design non-flat frequency response intended to impress the customer in the showroom, but not necessarily best for long term listening at home.

 

A dip in the presence region and a boost in the upper treble creates the impression of spaciousness.

 

In any event, my wife and I both feel the CM7 gives music an unnatural tonal balance unless this frequency response is corrected by EQ, so I don't believe B&W's design philosophy merits deference.

 

That was almost a year ago. I'm now using equalization/room correction. The wood floor adds a new twist, but once I put the Persian rug back in place, it sounds better than before. (Even with the echoes, I think it sounded better.)

 

Keep in mind these were probably over-priced at the time at $2K for the pair. (They were soon to be discontinued in favor of the CM8, CM9, and most recently, CM10). To some degree, presumably, you get what you pay for. When I purchased them, I didn't know the difference between an mp3 and a lossless file. I basically bought the least expensive speakers that didn't look and sound terrible in the showroom.

[wgscott]

Here is my initial room response and correction for the CM7s. They do have the dip between 1 and 4 kHz, but there is no accentuated treble. If anything, it rolls off prematurely (I am guessing due to my ClassD Audio amp). This is with the wood floor and Persian rug in place, but the earlier one didn't look significantly different.

 

[semente]

Here is my initial room response and correction for the CM7s. They do have the dip between 1 and 4 kHz, but there is no accentuated treble. If anything, it rolls off prematurely (I am guessing due to my ClassD Audio amp). This is with the wood floor and Persian rug in place, but the earlier one didn't look significantly different.

 

As far as I know, it is normal to have some top end attenuation at the listening spot (due to distance and air absorption).

But if your tweeter is that same as the one on the CM5s, then the roll-off above 15kHz is probably an attempt at smoothening down the tremendous resonance peak at 30kHz.

Fortunately even HighRes recordings have little info above 20k, and it's mostly very low in level...

 

R

[semente]

Looking at the CM5 waterfall, there seems to be a resonance at 4kHz (I can't understand why JA insists on plotting tweeter resonant peaks while chopping off what's really important information):

 

Bowers & Wilkins CM5 loudspeaker Measurements | Stereophile.com

 

Since this is the crossover frequency, it is possible that the dip in the "presence" region (2-4kHz) is being used as a trick to conceal the kevlar cone resonance...

 

R

[wgscott]

the tremendous resonance peak at 30kHz.

 

If it has a resonance peak at almost 2X the threshold of hearing for most adults, why is this problematic?

[semente]

If it has a resonance peak at almost 2X the threshold of hearing for most adults, why is this problematic?

 

I am not sure but I think that when triggered it can affect the tweeter's performance at lower, audible frequencies...

 

R

[wgscott]

... which in turn should be readily measurable.

[semente]

... which in turn should be readily measurable.

 

We'll never know; JA likes to keep the tweeter resonance peak within the graph's limits and that hides all in important information.

But he does mention it:

 

"There is a 20dB-high peak at 30kHz. Fortunately, this resonance lies 5–7kHz higher in frequency than is usual for an aluminum-dome tweeter, and it won't be excited with CD playback. In playing back LPs with a moving-coil cartridge, however, the resonance will be set in motion by ticks. It will also be excited by high-resolution digital playback, with an unpredictable effect on sound quality within the audioband."

 

Makes one wonder why people get so excited about ultra-sonic frequeny content ability in HiRes media when so many tweeters have strong resonances at around 23kHz...

 

R

[wgscott]

It is an argument against having higher sampling frequencies.

 

However, I think he isn't correct saying "... an unpredictable effect on sound quality within the audioband."

 

It is quite predictable. You could even calculate it with the Kramers-Kronig relations.

[Jud]

It is an argument against having higher sampling frequencies.

 

Or rather it would be if that weren't rendered moot by in-DAC oversampling to rates higher than you'll find in 90+ percent of high res downloads.

 

Hard to understand why this simple fact continues to be forgotten or neglected.

[wgscott]

I'm not forgetting or neglecting it.

 

I'm simply suggesting what my tweeters do at 30 kHz is irrelevant to their audible limitations.

 

(Upsampling won't generate content at 30 kHz that wasn't there to begin with).