Yeah, I'd love to see the JBL remeasured.
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Worst measuring loudspeaker?
- Thread starter Purité Audio
- Start date
Actually, a lot can go wrong with overseas suppliers and manufacturers as well as quality control and rough handling of the products during shipping. I agree it should not happen, but you cannot compare this to a DIY or home-manufactured product.
The scaling might change the visual impression of linearity issues, which are anyways to be taken with a grain of salt, if there is no additional information on the directivity, and we don´t know how it actually sounds (have heard the 3-way model solely).
Nevertheless I would say, despite from heavy smoothing the issues between 500Hz and 2K persist, and they would make me cautious as such behavior is often a result of either resonance, diaphragm breakup, cancellation or directivity issues which in most cases are audible and cannot be EQ´ed.
If we assume that on-axis response and in-room response have to somehow work together to create a tonally balanced impression, I would assume the bumps at 800 and 1,2k do not make any sense, as this is usually the band in which an 8" just begins to narrow down directivity, and the neighboring bands are transition to the horn and the horn playing alone, which makes them recessed in the room anyways. Many 8"+horn 2-ways sound presence-heavy, like with boosted 1K region, so boosting this band further does not really add up.
In other cases, such behavior is indicative of a soft-diaphragm, yet powerful 8" cone which shows breakup in its highest octave, but it would be speculation to presume this is the case here. I would anyways shy away from such concepts as they in my understanding offer not advantage over a 3-way, particular not with a price tag of 15 Grand.
That is a rather unlikely explanation, as FR with changing vertical position would always be most pronounced at crossover point (1.6k) as well as in the upper treble, and result in pretty narrow-banded issues or a continuous slope (for the treble). Both is not the case here, issues are rather broad-band and limited to 1-6K.
That is clearly a narrow-banded issue and might be resulting from either problematic horn design, high crossover frequency or lobing/phase issues (or a combination thereof). A similar behavior can be ruled out for the JBL in question (1.6K crossover).
Something is clearly amiss here, either with the speaker or the measurer ?
Nasty load tooSorry, Hi-Fi News.
Keith
JBL Summit Ama loudspeaker
With its three new Summit series loudspeakers, JBL is continuing its journey through the Himalayas. The Makalu and Pumori floorstanders, plus the standmount Ama auditioned here, join its existing ‘Project’ Everest DD67000 [HFN Aug ’14] and K2 S9900 [HFN Aug ’10] flagships in taking their name...www.hifinews.com
2.8ohm/212Hz minimum is lower than JBL’s rated 3.1ohm/239Hz. Moreover, there are substantive +57°/–71° swings in phase angle that incur a minimum 1.1ohm/100Hz EPDR that remains sub-2ohm from 63-700Hz.
Any room correction can correct for this so I would not call it 'broken' but indeed at this price level it just should not happen
Solely looking at frequency response, surely not. Difference between the two units? Definitely yes.
That is an interesting position, as it would apply to the majority of speakers covered in this thread - you can correct almost any frequency response with the help of room correction, and most of speakers don´t deserve to be called ´broken´ or ´worst measuring´?
Have to disappoint you here, as deviations between specimen of a speaker, particularly those equipped with a horn, usually come with other side effects which cannot be corrected by EQ.
I have experience with thousands of speakers in a lab under anechoic conditions, and I can assure you: evidence is overwhelming in this case that the cause is the speakers, not the measurer. Everything a lab engineer could do wrong, like microphone position, polarity, measuring distance, can be ruled out here looking at the two graphs, as it would leave typical traces in either the upper treble or the transitional band. Nothing like that here.
It is a bit ironic. For years we have been told that Harman/JBL is the only manufacturer in the world in possession of the holy truth of loudspeakers, defining ´target curves´, the only entity in the world capable of doing truly scientific research on sound quality, and the moment someone is coming up with a speaker deviating from any of these ideals, chances are high it will be called rubbish and ´bad from scientific point of view´.
Now we have an expensive reference-line product by JBL which is clearly deviating from anything Dr. Toole and Dr. Olive have ever postulated, the latter hinting that the internal research facility was not working on loudspeakers for 10+ yrs already, being finally discontinued, the mother company eventually buying B&W who are known to be kind of the forefront of ´frequency response does not matter´ movement among loudspeaker manufacturers.
How does this all add up? Why did JBL assumingly drop their ideals of the past? Make a wild guess!
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,931
- Likes
- 6,880
Blue is the FR at the recommended listening angle, with the speakers axis crossing behind the listeners head.
Waxx
Major Contributor
I would like to see this measured in a more controlled setup also, because it's results make me think something's wrong with the measurments. Not that i think these series are top of the line, they are clearly targetted to the audiophile market seen the pricing, and JBL does it often right, but not always. But the measurements show stuff that i hardly can think is due to the speakers (in my experience with building speakers myself). Hifinews is also not really a site i trust on this (in general), It's to much a marketing/paid review site than an objective source like here or Erin.
I did hear the Everest (old model) next to the 4367 and liked the 4367 (subjectivly) a lot more. I don't think that reference line is the best in their lineup. There the styling is clearly more important than technical perfection. But this result is not what i would expect. For me the ultimate JBL's are still the 4367 or the M2, and that was it already before I saw this measurments.
It would be great if this speaker ends up at Amir or Erin's Kippel, with more standardised controlled measurments to confirm or correct what we see here.
I did hear the Everest (old model) next to the 4367 and liked the 4367 (subjectivly) a lot more. I don't think that reference line is the best in their lineup. There the styling is clearly more important than technical perfection. But this result is not what i would expect. For me the ultimate JBL's are still the 4367 or the M2, and that was it already before I saw this measurments.
It would be great if this speaker ends up at Amir or Erin's Kippel, with more standardised controlled measurments to confirm or correct what we see here.
It would be unfortunate to draw conclusions out of context, since this magazine (HI-FI News ) presents measurements whose scale we do not know. Just compare the MoFi V10 Master Edition measurements published by the magazine with Erin 's Klippel measurements that speaks for itself. That said, I completely agree regarding the 4367.
See, this is much worse.
HFN's scale is much more compressed horizontally and more expanded vertically than what we see from ASR/EAC/Stereophile. 30 dB top to bottom, but also there's an entire extra octave shown above 20k.
MarkS
Major Contributor
- Joined
- Apr 3, 2021
- Messages
- 1,818
- Likes
- 2,938
Graphs show +/- 3dB from the lower limit (200Hz) to 10kHz, which is quite good, and then a big peak at ~15kHz. This is due to beaming, common to all large planars. Off-axis and power-in-room should be much more reasonable.
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,931
- Likes
- 6,880
That is an extremely generous reading. There are different ways of getting ±3dB, and this...
...is not a good one. Big hole, big peak, big hole, lots of jaggy resonances. Plus channel to channel variances.
And BTW beaming in a flat panel does not cause a peak on-axis. The energy is still going wide, but it is being cancelled by out-of-phase energy from the other side of the panel. The peak is most likely severe resonance.
I would really miss the 1-2 kHz, the peak above 10 kHz is not (more) my working area
, so would more help than harm in my place.
MarkS
Major Contributor
- Joined
- Apr 3, 2021
- Messages
- 1,818
- Likes
- 2,938
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,931
- Likes
- 6,880
@MarkS thanks for backing up my statement. Nowhere does it say what you said, that the frequency response needs to spike as a result of this beaming.
To repeat what I said, the energy is still being emitted broadly, but cancelling off axis by out of phase energy from the other side of the panel. It is not being 'redirected' to on axis, as you seem to be implying.
cheers
To repeat what I said, the energy is still being emitted broadly, but cancelling off axis by out of phase energy from the other side of the panel. It is not being 'redirected' to on axis, as you seem to be implying.
cheers
Similar threads
