TheWarrior

TheWarrior

Audioholic Ninja
Someone once said. Evaluating a speaker quality through it frequency response curve is like measuring a wine's quality by its chemical analysis.

Who agrees with that?

I know that Dr. Toole had done extensive testing that showed that Spinorama could predict speaker performace as determined through user preferences. But, I personally have not been able to figure out which speaker will sound good by its frequency response or any other measurement. I have to hear the speaker to see if I like it

Sent from my SAMSUNG-SM-G930A using Tapatalk
A peak in all the off axis measurements confirms a resonance within the speaker itself that colors the sound. I found that audible upon my listening, so I looked up the speaker and found the measurements to explain what I heard, and others did too.

As to your concern with not hearing the same results, how are you listening? Toole conducts tests with a single speaker in mono. It is much more revealing, and is quite relevant given our multichannel - monophonic, sound tracks. Plus you don't get the masking effect of stereo - if a speaker has an issue, you are more easily able to recognize it.
 
B

Beave

Audioholic Chief
I agree that the LS50's bass and treble are superb, but I personally felt the cost of entry did not justify such a glaring, measurable, design error that resulted in inaccurate reproduction of sung voices - simply, there were notes left off the page.
Wouldn't that resonance accentuate content around that frequency? I don't understand how you think "there were notes left off the page." I would expect the opposite - notes unnaturally accentuated a bit.
 
B

Beave

Audioholic Chief
For what it's worth, we had a discussion about the LS50 measurements before and why the resonance at 2kHz seen on the NRC measurements doesn't show up in Stereophile's waterfall plot (or other waterfall plots I've seen of the LS50 from European sites).

The reason is simple, and obvious once you know it, but it took me a while to figure it out:

The resonance doesn't occur directly on the tweeter axis. It occurs off-axis (and to some extent gets stronger the farther off-axis you go). It appears to be an issue with the surround of the midwoofer. Stereophile's waterfall plot is based off JA's measurement of the step response on-axis only. That's why there is no resonance seen at 2kHz. If he did an off-axis step response measurement and then created a waterfall from that, the 2kHz resonance would show up.
 
Last edited:
B

Beave

Audioholic Chief
Also, I saw on another forum a while back that Dr Toole has actually commented on the NRC measurements of the LS50.

He said that resonance is off-axis, but since we listen in reflective rooms and not in anechoic chambers, that the resonance is going into the room and will be audible. It will be more audible in more reflective rooms and less audible in heavily damped rooms.
 
S

shkumar4963

Audioholic
As to your concern with not hearing the same results, how are you listening?
.
My comments were not in response to your comments about the resonance in ls50.

Sent from my SAMSUNG-SM-G930A using Tapatalk
 
TheWarrior

TheWarrior

Audioholic Ninja
Wouldn't that resonance accentuate content around that frequency? I don't understand how you think "there were notes left off the page." I would expect the opposite - notes unnaturally accentuated a bit.
A resonance that comes from the speaker itself is going to color the sound but there are too many variables to say what that would sound like. Considering that resonance covers a whole octave, it seems pretty self explanatory why many people have passed by this speaker.

It was actually Toole that pointed me to those measurements, all I could find were Atkinson's, but the 2khz resonance is very much visible in the on axis measurements - you shouldn't have your midrange abruptly rise 5 dB from 1.5khz - 2khz!
 
Montucky

Montucky

Full Audioholic
Perhaps these speakers are very room dependent. That said, all too often we get far too caught up with measurements and specs. I'll tell you this much, MY ears sure as heck don't have a perfect frequency curve when it comes to incoming information. I know for sure that I have some loss of more than a few frequencies. All of our ears are different, all of our perceptions are different. All we can trust is what WE find pleasing. Sure, a chart is a great starting point and can give us an overall idea of a speaker's qualities, but it's certainly not the final deciding factor.
 
Steve81

Steve81

Audioholics Five-0
It appears to be an issue with the surround of the midwoofer.
Hard to say what the issue is. The UniQ driver in the LS50 is fairly similar to the R100, which performs a bit differently through that range off axis.

LS50
LS50.gif
R100
R100.gif
 
speakerman39

speakerman39

Audioholic Overlord
Hard to say what the issue is. The UniQ driver in the LS50 is fairly similar to the R100, which performs a bit differently through that range off axis.

LS50
View attachment 23449
R100
View attachment 23450
That is rather interesting Steve. Everything that I have been reading raves about how well built the cabinet is on the LS50's. To that end, wouldn't resonant peaks be minimized, if not down right eliminated, due to dampening of the cabinet? Also, the "R" series is said to be more laid-back from what I have garnered here and there.

That is why auditioning is the best way to discern what works and what doesn't as we all hear differently. For example, B&W is said not to measure very well (flat). Yet, I have owned the CM1 S1's and loved them. In fact, they were very impressive in my room. Never had bookshelf speakers that could play that low and sound so darn good doing it. The extension was only part of it. Hence, the quality albeit linearity of the low end was equally impressive. All of that from a 5" driver.


Cheers,

Phil
 
Montucky

Montucky

Full Audioholic
For example, B&W is said not to measure very well (flat). Yet, I have owned the CM1 S1's and loved them. In fact, they were very impressive in my room. Never had bookshelf speakers that could play that low and sound so darn good doing it. The extension was only part of it. Hence, the quality albeit linearity of the low end was equally impressive.
"But...but...but...but...the measurements!!! Don't trust your ears and personal taste, or how those speakers react within that particular space! The measurements are all that matters! Didn't you see the charts?!"
 
Steve81

Steve81

Audioholics Five-0
That is rather interesting Steve. Everything that I have been reading raves about how well built the cabinet is on the LS50's. To that end, wouldn't resonant peaks be minimized, if not down right eliminated, due to dampening of the cabinet? Also, the "R" series is said to be more laid-back from what I have garnered here and there.
Cabinet resonances would be eliminated, but resonances due to the driver wouldn't be. That said, my take on it is that it isn't necessarily a resonance. The LS50's tweeter is voiced a bit hotter than that of the R100, which is reflected in the on axis measurements. That accounts for the rise we see in the off axis measurements at 2kHz, and the dispersion begins to narrow above that as expected, at least until we see the effects of the tangerine waveguide/phase plug, which serves to improve dispersion of the top end.
 
speakerman39

speakerman39

Audioholic Overlord
"But...but...but...but...the measurements!!! Don't trust your ears and personal taste, or how those speakers react within that particular space! The measurements are all that matters! Didn't you see the charts?!"
Well, graphs can be a useful tool when prospecting speakers. However, listening to them in our own homes with our own ears is the best policy. Even when a speaker that measures flat may not be so flat sounding in a given room. Of course, room treatments can be done to deal with such issues. Proper placement of said speakers can also go a long ways as well. In my experience, toe-in can really help, but it can also hinder. It just depends on the speakers and the room that they are placed in.

If a speaker is overly bright, treating the 1st and 2nd reflections points can be of some help. The brightness may not fully go away, but can be somewhat reduced (absorbed). I was not able to measure the B&W CM 1 S1's that I owned, but my guess based on my reading is that they did not measure "flat". However, I certainly did not lose any sleep if they didn't. :):):) But, I sure did love listening to them as they never were fatiguing. Even after hours of listening.


Cheers,

Phil
 
Last edited:
speakerman39

speakerman39

Audioholic Overlord
Cabinet resonances would be eliminated, but resonances due to the driver wouldn't be. That said, my take on it is that it isn't necessarily a resonance. The LS50's tweeter is voiced a bit hotter than that of the R100, which is reflected in the on axis measurements. That accounts for the rise we see in the off axis measurements at 2kHz, and the dispersion begins to narrow above that as expected, at least until we see the effects of the tangerine waveguide/phase plug, which serves to improve dispersion of the top end.
That explains why many say the "R" series is more laid-back when compared to the LS50's. Personally, I already know I can't hear much above 16K. But, being in the 2-3K range that is bound to be somewhat audible. The question is, hower, how audible would it be and could it be treated. Listening in ones own home is the best way to find out inmho. Thanks Steve!


Cheers,

Phil
 
B

Beave

Audioholic Chief
Hard to say what the issue is. The UniQ driver in the LS50 is fairly similar to the R100, which performs a bit differently through that range off axis.
Good point.

I notice the R100 doesn't have the off-axis peak at 2kHz, but it does have a (smaller) off-axis hump around 5kHz that the LS50 doesn't have.

Maybe these are diffraction issues that are different on the two speakers due to the curved baffle vs. the flat baffle?
 
Steve81

Steve81

Audioholics Five-0
Maybe these are diffraction issues that are different on the two speakers due to the curved baffle vs. the flat baffle?
I’d be surprised if diffraction had that significant of an effect on either speaker, but never say never.
 
P

PENG

Audioholic Slumlord
Stereophile likes them so much that group them in their Class A short list. I am fine with that as I forced myself to listen to violin on them for almost an hour already, they sound great to me. When I shopped for speakers in the past I always use violin, solo, sonata concerto etc., and if they didn't sound great I would look elsewhere. Very subjective exercise obviously and I do get influenced by available graphs.
 
KEW

KEW

Audioholic Overlord
Well, I revisited the LS50's today to try to get specific on what it is that doesn't work for me.

I believe it can more or less be summed up by this graph and John Atkinson's statement quoted below:

The blue trace in fig.6 shows the spatially averaged response of my 1978 pair of Rogers LS3/5a's, taken under identical circumstances, but with a 3dB-higher level at 1kHz to equalize the two pairs of speakers' midrange outputs. The peak between 1 and 2kHz that lends this classic speaker its characteristic slightly nasal coloration can be seen, coinciding with a small dip in the LS50's response. The LS3/5a has noticeably more energy in-room in the top three octaves, this audible as extra "air" in direct comparisons with the KEF.
Read more at https://www.stereophile.com/content/kef-ls50-anniversary-model-loudspeaker-measurements#w58jJAL34IMdyAeg.99
In my earlier comments I spoke of a lack of ambiance, and this air is what I am talking about.

In many ways it is as simple as preferring a speaker with more presence! IOW, a more forward speaker over a laid back speaker.

I have a busy schedule the next couple of days, but hope to post my more specific observations on this either Thursday or Friday.
 
KEW

KEW

Audioholic Overlord
The point of this audition for me is to better define (or invalidate if I could not repeat) the aspect of the ls50 that did not sit so well with me. As such I primarily focused on that aspect of the sound.

I started out listening to the ls50's against the Canton Vento 820.2's. This seemed like a logical comparison since both have a list price of $1500. However, the Vento has a 6" woofer which goes pretty deep and from my perspective, aside from making sure it is not muddy or over-blown, bass is not too relevant as I would generally use a sub with either (the Vento might get away without a sub in a smaller room). In any case, I found the large difference in the bass between these two masked many of the more subtle differences.

I gave it some thought and pulled out my EmpTek 41se/B speakers. These little darlings are still among the best speakers I have ever heard. Their limitations are bass (4" Be mid-woofer) and SPL (as you would expect from a 6-3/4" W x 11-1/2" H x 10-1/4" D box). I listened at a distance of 4 ft. at moderate levels which would not challenge the EmpTek's SPL limits while satisfying my objective of listing to the ls50 under near-field conditions.

Norah Jones - Don't Know Why


Norah has an airiness in her voice. As a perverse analogy, it is kind of like the way Gollum says "precious" along with the pure sound of her voice. Unlike Gollum, it lends a seductive quality and is part of the intimacy of a female singer in a night club. This "whisper" is diminished on the ls50. I could describe the ls50 as sounding like she is singing without a microphone in a small venue. Or, perhaps a music hall where the mic was hanging in the air 10' away.

After I wrote these notes, I looked at John Atkinson's measurements and comments as presented in my post above, and believe that fits my subjective experience fairly well.

Ed Palermo's Big Band - RNDZL

This is a modern Big Band playing a re-arranged version of Frank Zappa'a RNDZL. It starts with an aggressive drum lead in and I was impressed with how realistically the ls50's brought the impacts of the drum hits into my room (especially given its size!). I think this is one of the strengths of the ls50's over other speakers in its price range!

At around 0:30 the drummer starts tapping the ride cymbal while accompanying an alto sax solo. The 41se/b presented this as tap followed by the ring of the cymbal. On the KEF the tap is not as prominent and the subsequent ring is almost inaudible.
Strangely, when the drummer does this later accompanying a trombone solo, the KEF does better with it. I don't know if the drummer was being more assertive with the more powerful trombone or if there is something about the harmonics of an Alto Sax vs Trombone that would mask the ride cymbal more with the sax? The sound is still more prominent on the 41se/b, but I would not have an opinion on which was "right" (to my ear).

I also did not like the loss of the upper harmonics of the glockenspiel in this song.

Pink Floyd - Time


The main difference here is how the bells presented near the start. The EmpTek sounded as if they were closer to being in the room, while the KEF placed them "down the hall".
The KEF did a fine job with both the male and female vocals on this song. Switching between the speakers, the relevant difference was the comparative bass lacking of the EmpTek's, and I fully enjoyed finishing out the song listening to the KEF's.

Yes - Heart of the Sunrise

Bill Bruford's drumming is front and center through this song and the drums sounded somewhat veiled on the KEFs as compared to the EmpTeks. I think, again, this is due to the suppression of the cymbals' HF content.
That said, the KEF's otherwise completely trashed the EmpTeks on this song. They fleshed it out wonderfully and made the EmpTeks sound absolutely "thin" in comparison, and I'm talking more than just deeper bass! This is a good place to point out that if you think I am saying the EmpTek is a better speaker, you would be wrong; my objective is to focus on and explain what I found lacking in the ls50!
Interestingly, the airiness in Jon Anderson's voice was diminished, but it just didn't matter that much with his voice. I don't "feel the loss" of it the way I did with Norah. IOW, when I A-B speakers, I mainly respond to my sense of loss, or the thrill of an improvement. If I don't feel it, it is not so much of consequence in my subjective comparison.

Steely Dan - I got the News

Initially, the KEF sounded great on the tight percussive content of Steely Dan, but later into the song, the drums sounded veiled. Upon revisit, I realized when the high-hat is being played, it sounded like the drum kit was tucked in a recess behind the band (not a very significant difference), but at 1:52 where the ride cymbal is played, once again, the KEF's presentation puts a thin veil over the drum set, which for me is a loss.

Chet Atkins (with George Benson) - Sunrise

Again the difference showed in the drum set. The blur of dozens of individual brush strikes on the snare drum is more realistic with the added high frequency energy of the EmpTeks.
The triangle presents as "somewhere back in the back" on the KEF.
I like to close my eyes and envision the musicians/instruments and in this case, the trap set seemed about 15' away on stage with the EmpTek and in an orchestra pit with the KEF.

I think this song would be the one I would choose if I wanted to demonstrate the shortcoming of the KEF's HF presentation! In a way having it happening in female vocals is more damning, but the brushes and triangle would take on a substantially different nature when I switched between the speakers.

Emilie-Claire Barlow – C’est Si Bon

At this point, I think I am beating a dead horse. I will just say I experienced the same as with Norah Jones - the intimacy from the air in Emillie-Claire Barlow's voice is suppressed on the KEFs.

Lyle Lovett - She's No Lady

Lyle Lovett has one of the fullest richest voices I have ever heard, but I really had not recognized that he actually also has some of that air in his voice! However, just like with Jon Anderson (probably more so), it doesn't matter. For me, it is a difference, but neither good nor bad.

Conclusion

Among the music I listened to, with the exception of the air in female voices, ride cymbal, glockenspiel, brushes, and triangle, the KEF presented sounds with a breath-taking clarity that is a solid benchmark in its price class.
I like a lot about the ls50's and fully understand why they are so well liked.
I know I like a forward rather than laid-back speaker, so that is definitely part of the equation.
Furthermore, if I could not instantly switch between speakers in the middle of a note, I'm not sure how much I would notice these differences.
I believe Steely Dan, Yes, and Pink Floyd would all present perfectly well on the KEF without the instant comparison.
It is the "night club" female vocals (Norah Jones and Emillie Claire Barlow) and the Chet Atkins tune where I believe I would still sense a difference given a 2 minute delay between listening to one speaker or the other.
Listening to the big band, now that I know to listen for the ride cymbal, I would notice, but otherwise, I doubt I would pick it out as an issue without the instant comparison. I don't think I would notice the glockenspiel differences were it not for the instant comparison. It sounded great on the KEF!
What I am hedging at is, I think that for many types of music, the KEFs give up nothing.
 
Last edited:
TheWarrior

TheWarrior

Audioholic Ninja
It is the "night club" female vocals (Norah Jones and Emillie Claire Barlow) and the Chet Atkins tune where I believe I would still sense a difference given a 2 minute delay between listening to one speaker or the other.
Listening to the big band, now that I know to listen for the ride cymbal, I would notice, but otherwise, I doubt I would pick it out as an issue without the instant comparison. I don't think I would notice the glockenspiel differences were it not for the instant comparison. It sounded great on the KEF!
What I am hedging at is, I think that for many types of music, the KEFs give up nothing.
So as long as I don't look at anyone else's yard, I'll love the way you trim my hedges?
 
newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top