3-4 dB dip at crossover region: what should I listen for to hear it?

You don’t need a spl meter and generators, you just need to listen to a frequency sweep. Many audio test albums have those tracks.

A couple of comments:

Frequency response is measured with continuous signal at 1 watt of input.  This does not take into account what happens when things really get moving and music is played through them.   The linearity of a steady state output at 1 watt may be quite different  from a transient of a few microseconds at, say, 80 watts. Music played at a "normal" listening levels may exhibit little perceived (or measurable) differences in frequency response.  

At the crossover point, 2 drivers (typically) are playing the same signal at, or about, the same volume level.  These drivers may be vastly different in design: materials, cone area, motor structure, or technology thus having varied characteristics in resolution, dispersion, speed, dynamic headroom, etc.  The "weirdness" that can occur to dynamic signals in the narrow range of the crossover points can't be predicted with standard frequency response measurements.

You've got to give them a listen.  In your room.

OP:  I dont' know about better, but I heard the Klipsch Heresy's with a Rogue integrated and really liked them. 

What's an example of a currently produced speaker in the 4349's price range (give or take) that has similar dynamics and sound quality but measure better? 

@erik_squires, thanks, that makes sense.

Funny you mention Altecs -- I don’t have much experience with horn speakers other than some old Altec Lancings my friend runs with a pair of MC30s. The 4349s remind me of those speakers more than any others I’ve heard. My friend’s system was my first exposure to hifi beyond the Bose 301s and Sony Receiver I had at the time, so its sound is engrained in my memory.

@james633, yeah, I think I'll start a thread for the 4367s and 4349s -- I'd  be interested in reading about both models.

Any speaker or combinations of them will have their own peaks & dips, it's own response to an input....it's 'voicing' can either be a 'love it' or eq it to 'make it sound OK' in ones' space....

...and the graphs will reflect hat.....not that the unit will sound 'better' or 'worse', it just shifts the perception to the listener v. the means of measuring of such.

I've messed about with various 'n sundry....and you can only 'tweak' so much before the basic 'characteristics' of a speaker in a very subtle fashion until it turns to muck....mho, of course....

OP: To be clear, your speakers don’t exhibit what I'd normally call horn colorations.  While not perfect, the overall response in the horn region is pretty smooth compared to say an old-school Altec. The complaint ASR makes is that the two drivers don’t blend perfectly at the crossover.

Horn colorations are the peaks and dips in the usable range of the horn, not how they mesh with the other drivers.

I would be interested in a dedicated threat about the 4349s as you get used to them and compare them to your other speakers. I am thinking about buying 4 of them for surrounds for use with my 4367s. So the “audiophile” sound quality is not critical for what I plan to use them for I still enjoy reading opinions. 
 

I know I have liked my 4367s enough they have changed my opinion on what is important for hifi. Dynamics have moved from important to most critical. Interestingly I notice it most on subpar recordings. Old kind of cheesy music like Devo and B-52s have punch and snap that I never knew was there. Van Halen jump (the original not the remastered trash) has well, jump. While audiophile recordings like Vanessa Fernandez “here but I’m Gone” snare drum hits are pretty stunning. 

I don’t think you will hear it. I have a pair of Klispch RP8000f (own a pair of JBL 4367s too) and they RP8000f has the same kind of thing at the crossover but much worse on axis. Once in room I really don’t “hear” the dip much. In the Klispch it comes across as a slight lack in female vocal presents.

You can probably effect it with toe and distance from the side walls. If you have a computer for a source just boost that frequency with EQ and see how it sounds.

I think JBL should have used a 2” driver and crossed over lower. I know they prototyped a 2” model. Maybe it was too late or cost too much to hit the price point of the 4349. I have never heard them but the 3” driver in the 4367 is pretty amazing. Based on the white paper the duel ring diaphragms have less break up at high frequency compared to the 4” titanium drivers but have more distortion at the lower end and probably why they crossed over the 1.5” driver in the 4349 so high. I know Greg Timbers said the new driver was more about cost control than performance vs the 4” metal domes and he thought the end result was about the same. The pro side that they D2 driver was designed for was all about power handling and most the pro speakers that use the D2 are 3 ways with mids so the driver is not pushed as low.

on a side note, I see a comment about horn coloration. This to me is caused defraction back down the feed throat which the 4349 should have nearly zero due to its advanced design (computer aided).

White paper here if anyone is interested.
https://www.audioheritage.org/vbulletin/attachment.php?attachmentid=83563&d=1552949127

@rischa 

It won't be dramatic, but female voices will be a little darker. Same for violins. Contrary to opinion 3-4 dB is quite a lot. You may not notice anything wrong as is. You would have to compare it with a corrected speaker. Digital preamps with room control and EQ allow you do make comparisons like this. 

Probably.  Each shape of horn will have its own set of peaks and dips.

@toddalin, thanks for posting the graphs. Is this generally what's behind the "horn coloration" I've heard mentioned? 

@toddalin That was my memory/expectation as well.

As I said, dips in the upper midrange are characteristic of the breed when using horns.

JBL 4348:

4345:

4355:

4333 (L300 pro equivelent) I have a pair of L200 modified to L300 specs and the dips drive me crazy. I don’t even listen to them anymore since I created the Mermans.

4313: 12" 4-way with no horn:

"I’ll also note that the JBLs in the video were the 4367s, the model above mine. According to Erin at Erin’s Audio Corner, the 4367s measure and sound great. Erin is also one of the guys on the ASR 4349 thread who thought the 4349s measured terribly (I like Erin and subscribe to his Patreon)."

Do they really? I hear slight the depression and Stereophile certainly found it:

In home measurement:

The ASR measurements suggest that the midrange dip is less pronounced in-room. Using a 12in driver in a two way design is always going to be compromised. Dips in frequency response are not always interpreted by the brain as dullness - a drop in the midrange may be heard as brightness because it can make the higher frequencies seem elevated relative to the midrange.

The depressed upper midrange is no stranger to JBL professional monitors all the way back.  Some are worse than others.

@toddalin thanks for the explanation and video. I thought both sets of speakers came off pretty bad in rhat video, but - and feel free to accuse me of being biased - the JBLs were much more tolerable. The Klipsch sounded quite shouty to me, whereas the JBL had depth and warmth (warmth relative to the Klipsch that is).

I’ll also note that the JBLs in the video were the 4367s, the model above mine. According to Erin at Erin’s Audio Corner, the 4367s measure and sound great. Erin is also one of the guys on the ASR 4349 thread who thought the 4349s measured terribly (I like Erin and subscribe to his Patreon).

@jheppe815, thanks for the information -- very reassuring. My top end is pretty smooth so maybe these have more hours on them than I thought. Overall I like them, but I need more time to adjust to their sound. They’re SO dynamic. I’m used to the much more delicate sound of my Omega High Outputs. I’m about to switch the 4349s over to one of my tube amps and play some classical guitar. If it sounds as real/natural/correct as my Omegas, that will be a great sign.

If anyone thinks I should start a dedicated thread for rhe 4349s, let me know. Not sure how much interest there is.

The range in question is above the fundamental pitch of most singers and it is the overtones that add "brightness" to the sound.

Perfect example. Listen how her voice (maybe guitar too?) is a bit more "laid back" with the JBLs.

1500 is upper midrange. If you do a search for frequency chart of musical instruments,  you should find what you are looking for.  Erik,  I have built using buttererworth, bessel, linkwitz riley and chebyshev. I believe that you were looking for bessel,  this has a very smooth impulse response and I agree with you, I suspect the designer was most likely dealing with some other issue that they used this dip to deal with it. 

Wow.

I agree with Corelli. 

@rischa - I remember distinctly how good the 4349's sounded right out of the box.  However, the high frequency compression driver did smooth out in short order as the drivers broke in.  I didn't have a very long break in period where the cabinets sounded vastly different, though I put quite a few hours on right away.  I run JBL 2269H 18" drivers as subs, so low frequency extension or break in isn't something I could comment on.

As I work for an AV company, my business partner came to audition the system as well and we used a calibrated mic and software to see what was going on in the room.  Both of us were very aware of the various graphs that have been published showing that 1.5k dip and we wondered if we could hear that.  There is a slight (and very slight) downward curve in the 1800 to 2200 hertz area on the graphs as we measured the room / system, but of the various people that have heard the system (all of them musicians and all of them with very good systems better than mine), no one has stated "hey, what's with the dull / flat vocals" or anything like that.  I'm certainly not grabbing for my 1k or 2.5k eq knobs on my preamp and boosting those frequencies.  I find these cabinets very enjoyable and have no want to get rid of these for something else.

I have a second system with reasonable components and as I go back and forth between systems, I don't find the 4349's severely lacking in any way.

@toddalin, thanks, I will listen for this and compare with my other speakers. Can I ask what you mean by "overtones"?

With a dip of that magnatude in that area, you will hear a "dullness/flatness" in the vocals as though the overtones are subdued.  Many JBLs suffer from this. 

@jheppe815, glad to hear you’re enjoying your 4349s after two years. They’re definitely growing on me. Can I ask if you experienced a long break-in period? I actually bought mine used but I don’t think the original owner played then much beyond a brief audition period, as I know he had a few different speakers on hand to try at the same time as these.

Here’s mine:

Yes, I’ve asked myself several times how I ended up with PA speakers, lol. I think they look great, though.

@rischa - I bought my JBL 4349’s two years ago. I read extensively about these speakers before I bought them as you are probably doing.

I’d recommend simply to just listen to them and put away all the reviews, graphs, etc. I very much enjoy my 4349’s. Yeah, probably not an "audiophile" grade speaker in many people’s minds (a good friend has asked me why I have a pro audio "PA in my living room"), but I’ve had two years of really enjoyable listening. They will play at whisper quiet levels with much detail yet play at very high volumes that can be great fun. I too wondered about the dip in response that was noted in reviews, but that has not been noticeable by me or others that have heard the system in detail.

That cable comment is popular with me!
Also the remarks about room acoustics. I was going to write something of the kind myself. Room acoustics are at least as important as the speaker technology for the final sound.

I just want to add to @erik_squires comment. If you test your in-room response at the listening position you will quickly figure out why you are not hearing the narrow dip shown by ASR. By the time you put your speakers in your room and you add the reflections and absorptions in a normal listening environment the frequency response varies by a lot more than the dip you are concerned about. The audiophile forums are filled with stories by listeners spending huge dollars on room treatment to control these reflections/absorptions and finding that they made the sound worse.

And now for an editorial comment that will likely be unpopular but I just cant help myself. I think it's ironically wonderful that this thread is about not being able to hear a several dB dip in a speaker's frequency response but many members on this forum will report hearing dramatic differences between interconnects. When they describe what they hear they often use terminology that indicates large frequency response differences between two kinds of cables which, of course, don't show up in any kind of testing. This is an interesting hobby.

mlsstl’s comment dovetails with something I wanted to add. I use unamplified acoustic music as my "personal reference," too. I do like rock, even loud rock (e.g., Tool), but mostly listen to so-called "classical," and I play cello and acoustic guitar; my wife plays piano and my daughter violin. We hear live acoustic instruments in my audio listening space every day. We also sing, my daughter professionally.

Still, here’s a lesson of some kind, I think. I’ve got five pairs of high-end speakers, and two systems (one mostly for movies, in the library, and the main rig for music). Every now and then, I set up one—or even two—of the "extra" speaker pairs in my main listening room in such a way that I can fairly easily switch between them and my favored pair. And let me mention that my favored pair (Scientific Fidelity "Teslas" made in the late 1990s) are rare probably because Corey Greenberg in Stereophile killed the company with a very negative review when they first came out. One of the other pair are highly regarded Von Schweikerts, and another pair won all kinds of awards from Stereophile and other respected places, measuring flatter and with less distortion than any speaker at any price ever measured to that point in the anechoic chamber of Canada’s National Research Council (that’s a hint about its identity). However, of all my speaker pairs, it’s that last one I like the least. And remember: my ears are trained to prefer natural acoustic instruments in the very same acoustic space as my audio system utilizes.

So what’s the "lesson" here? Maybe that listening to recorded music is just a different experience from listening to "the real thing" (if that means: live acoustic instruments, voices, etc.). To the extent that this is true, measurements may actually be misleading—as they are for that speaker I just mentioned but didn’t name.

@mlsstl  You are not wrong at all. 

By the way, I'm not saying a juiced frequency response is necessarily something never to be done.  Between old Wilson's that had a similar crossover hole at ~ 2.4 kHz to the Dali's with the 2-3 DB extra treble lift, or the Dynaudio's with a W shaped response.  These speakers all have fans and listening habits which make them the ideal speaker.  There are also speakers and reviewers that have become enamored of certain colorations and call it neutral.  Ugh.

@erik_squires -- "What actually happens is that the ragged response makes some notes pop more than others, tricking your ear into believing it to be more revealing, and it is, but only in some ways."

Just an added comment -- I find that some speakers that are "impressive" on first listen -- those that make certain instruments or voices "pop" -- often turn out to be the most wearing to listen to over time.  I use unamplified live acoustic music as my personal reference when deciding if I like a particular piece of equipment, but recognize that lots of people listen for different characteristics, particularly if they are primarily fans of rock, pop, EDM, RAP or such.  But, that's the nice thing about this hobby, there is something for everyone. 

Just, for the record, using a cheap SPL meter and frequency generator is so 1990s.  Get Room EQ Wizard with a calibrated mic (~$80) or OminMic and do it right. 

Alternatively, use your phone with AudioTools with something like the Dayton Audio phone mic

OP:  You are most welcome. 

@erik_squires , thanks for your input on this thread!

If you can’t hear it, you can’t hear it. don’t force yourself to.

MANY speakers have ragged response graphs far worse than this, and they are often credited with being "revealing" and making the reviewer go through their entire music catalog again as if they are hearing the songs for the first time. Amazing!!

What actually happens is that the ragged response makes some notes pop more than others, tricking your ear into believing it to be more revealing, and it is, but only in some ways.

This is overall a very smooth measuring speaker, with the potential for excellent off-axis performance and dynamic range. JA at Stereophile has certainly praised a number of speakers that measured far worse.

Thanks to everyone for the responses. So it seems like a narrow dip in spl at one point in the range is relatively inconsequential, at least in this instance. That's good to know.

@roadcykler, your post is surprising on an audiophile forum. Whay do I want to hear the dip? Because I want to be able to correlate the measurements with my listening experience. Or in this case, to learn that the the dip has little effect on what I'm hearing. If none of this was important, I'd just buy some old Pioneers at the goodwill and be done with it. I like learning about audio almost as much as I like listening to music -- it's how I put together a system that presents the music at its best.

One more thing. Several here have objected to measurements in principle, appealing to what your ears like as the only relevant standard. Well, yeah, of course. But that doesn't mean measurements are irrelevant. I don't want to open up this always contentious can of worms, but the fact is that measurements are used by the scientific engineers who design the equipment, and they do correspond, in "objective" ways, to our "subjective" ear-experiences. Duh. If you simply reject the relevance of measurements altogether, then you reduce the audio experience to one of taste alone. If you do that, this forum becomes nothing more than a place to share your enthusiasms. 

Speaking for myself, I appreciate posters like Erik Squires because they provide more than mere opinions, more than mere personal preferences. There are correlations between measurements and subjective listening which, at least in principle, bypass the pitfalls of mere judgments of taste.

Not sure if this is right or relevant (better informed posters like Mr. Squires, please correct me if I'm wrong), but such "dips" (i.e., "3-4 dB at about 1.5 kHz") are not analogous to, say, missing a shade of blue between Cerulean and Cobalt. What I mean is that the missing shade of blue will jump out as a gap to the eye. The "missing" decibels "around" 1.5 kHz, however, are as it were blurred together with frequencies just below and just above. It's not a discreet gap. I learned this by using a crossover to send "only" frequencies below 50 Hz to my subwoofer. I found that if I shut off my main speakers, I still got sounds out of the sub that were clearly above 50 Hz. A knowledgeable friend described this slope phenomenon to explain why. If this is a poor explanation, I would appreciate gettin' schooled by someone better informed.

BTW, the SPL meter linked above (by elliot...) is only A-weighted. But C-weighting is much more relevant for music, as it doesn't discriminate against low frequencies. If you are listening to music loud (say, 90 dB or more C-weighted), and switch your meter to A-weighting, the level will drop significantly (to 80 dB or even less). 

Why do you want to listen for something that many consider a flaw? Just enjoy the music and if you can't because of some perceived problem, buy some different speakers and do not look at the frequency response. Sheesh.

Believe it or not I actually would like to hear it, if only so I can better relate measurements to performance. The thing is, I’m not sure these speakers are right for me (too soon to tell - I always take a long time to adjust to new gear), so while I’ve got them I might as well learn from them. These are my first horn speakers so I bought them as much for the experience as anything else.

You don't want to hear it, because then you can't un-hear it and you'll be looking for new speakers.  If you like the sound, close all ASR webpages, and just listen to the music.

You don't want to hear it, because then you can't un-hear it and you'll be looking for new speakers.  If you like the sound, close all ASR webpages, and just listen to the music.

I just read the review and Amir from ASR actually put that speaker on his recommended list. Beyond measurements, he was pleased with the sound.

Today sound pressure meters are very inexpensive

get one with a tripod fitting

next a CD (not LP) with test tones, many single frequencies, and you set the mic at seated ear height at your listening position.

They don't have to be laboratory accurate, just give you relative differences, i.e. + ____ db or - _____ db than the prior frequency.

My test tone CD has 29 frequencies to compare. Expensive, you could find something else

You are measuring the space as well as your speakers. Slight relocation, slight toe-in adjustments differences can be found.

@panzrwagn I may have been thinking of the S. Harsch crossover alignment.

@panzrwagn no.  Linkwitz Riley sum to zero at the crossover if alihned correctly.   This is more obscure. 

Linkwitz-Riley

@erik_squires I think you were thinking of Linkeitz-Riley crossovers which are nominally -6dB at the crossover points.

And as far as the ASR crowd, I have a sneaking suspicion the individuals commenting have never heard the JBL 4349 or heard of Linkeitz-Riley filters, let alone the theory behind them.

If you want to know exactly what the crossover characteristics are, it’s quite easily done. Get a used audio frequency generator and oscilloscope and plot the curve with some graph paper or Excel.