JBL 4349 Review (Studio Monitor Speaker)

[Newman]

May I remind you how dB works? dB of sound pressure level (dB SPL) is defined as 20 log10 p1/p0. In other words 6dB means four times the sound level produced for the same electrical signal.

Did you miss the bit where I said “1/4 the power”? Or just not understand it?

Also, you may want scan the sensitivity specs of the speakers that are on the market. 97dB/1w/1m is an extremely high sensitivity!

My 4-ways average between 105 and 110. Please reset your scale.

cheers

 

[amirm]

I will check my claims with REW in the next days, maybe my measurement methods were wrong.

Pretty sure it is. See my video on this topic:

 

[Francis Vaughan]

And that piece by Bach was written to test organs

I'd not heard that idea. Not convinced in truth. It isn't known when he wrote it, estimates start from when he was 19, and run until his death. There is even some question as to whether he actually wrote it at all. There is no record of his ever performing the work (no surprise here, such things weren't recorded in general), and the only copy of the manuscript is by a student of another teacher who himself transcribed Bach's work. So, whilst a nice story, I suspect it has been made up well after the event.
Organ works of the time didn't contain registrations (the actual stops to use at any moment), that was up to the individual organist and constrained by the organ being played.

 

[richard12511]

Couldn't you easily test for this phenomenon by just running frequency sweeps at a bunch of different SPL's? So you'd run a 70dB sweep, then 80dB, then 90dB, and you'd overlay the frequency responses to see if they matched. You could even increase the length of the frequency sweeps if you were concerned that the heating effect wasn't being given enough time to occur. Seems like it would be a pretty easy thing to test for?

Erin has started doing this with his latest reviews.

 

[amirm]

Couldn't you easily test for this phenomenon by just running frequency sweeps at a bunch of different SPL's? So you'd run a 70dB sweep, then 80dB, then 90dB, and you'd overlay the frequency responses to see if they matched. You could even increase the length of the frequency sweeps if you were concerned that the heating effect wasn't being given enough time to occur. Seems like it would be a pretty easy thing to test for?

I ran a bunch of tests like this. At the end, drawing conclusions were hard so I abandoned them.

 

[Duke]

View attachment 125315

@amirm, by any chance did you try listening to the single 4349 from about 20 to 30 degrees off-axis?

The reason I ask is, I THINK this might sound a little bit better than listening on-axis (or approximately so). The direct sound from the horn would not be quite as loud, and I think there would be less spectral discrepancy between the first-arrival and reflected energy (which may or may not be desirable).

 

[amirm]

@amirm, by any chance did you try listening to the single 4349 from about 20 to 30 degrees off-axis?

Not horizontally but I sat below the vertical axis.

 

[Robbo99999]

I ran a bunch of tests like this. At the end, drawing conclusions were hard so I abandoned them.

Isn't the conclusion just that frequency response has anomalies at different SPL's, if that speaker happens to exhibit that........it's just a "flaw"/characteristic of that particular speaker, so no need to draw further conclusions? Ideally you'd want your speaker to have identical (within reason) frequency sweeps at different SPL's.

 

[Robbo99999]

Still confused?

Two separate idealized transfer functions; one for Headphones and the other for Speakers. Never did I say they were the same other than both being idealized standards.

This speaker tests much closer to the idealized speaker transfer function than most or all headphones approach the standard Headphone transfer function.

I think it is odd that speakers are judged more harshly and held to a higher objective standard.

Thanks DT

Ah, now I know what you mean, I thought you were saying this speaker was following the headphone Harman Curve closer than the speaker Harman Curve! Yes, ok, I see the parallels you're drawing now, you're saying speakers are held to more stringent targets than headphones when it comes to reviews on this site. Well, the reality is that headphones are a bit of a Wild West when it comes to freqeuncy responses, they can get very funky and far off target.....so I suppose it's more reasonable to expect speakers to adhere closer to the ideal.....the science behind "best" speaker frequency response is also a lot more solid than that of headphones, due to the fact that people's individual ear & head anatomy will determine what their "ideal" headphone frequency response would be, so you can't be quite as stringent with the "perfect" standards for headphone frequency responses, so it makes sense to be stricter with that of speakers.

 

[richard12511]

My 4-ways average between 105 and 11

I agree with @sarumbear that 97dB definitely qualifies as high efficiency. At least when talking hifi speaker averages. Sure, there are much more efficient designs(like your 4 ways, but those are very rare).

 

[Vuki]

I have seen loads of comments over the years from the first watt matters most and we don't use as much power as we think (Pass et all the SET lovers) and you need far more power than you realise (Sanders) but using an amp with a calibrated power meter shows far, far more power being used than I would have ever guessed in this 6 year old video I only saw recently. It is also amazing how much less a bit of chamber music used.
I suspect there is a clue to Amir's enjoyment of these in here.

What's strange to me in that video was that at 500W/bass heavy music I couldn't see bass membrane moving. From my expirience with that kind of power and music there should be at least 1cm membrane excursion.

 

[amirm]

Isn't the conclusion just that frequency response has anomalies at different SPL's, if that speaker happens to exhibit that........it's just a "flaw"/characteristic of that particular speaker, so no need to draw further conclusions? Ideally you'd want your speaker to have identical (within reason) frequency sweeps at different SPL's.

You would assuming the test is repeatable. But that aside, to the extent all speakers show some deviations, how do rank them?

 

[TheBatsEar]

This is a review ...

I reviewed your room. Amir! That dust! All it needs is a damp cloth and ten minutes to get rid of it.

My mother would skin me alive is this was my place. ;-)

 

[DualTriode]

@amirm, by any chance did you try listening to the single 4349 from about 20 to 30 degrees off-axis?

The reason I ask is, I THINK this might sound a little bit better than listening on-axis (or approximately so). The direct sound from the horn would not be quite as loud, and I think there would be less spectral discrepancy between the first-arrival and reflected energy (which may or may not be desirable).

Duke,

I believe that there is a lot to the on vs off axis thing you are speaking of.

First a couple of related thoughts;

Auditioning a single mono speaker in the lab with time gated measurements plus doing single speaker listing tests is a good place to start.

Doing the math looking at the beginning of beaming vs frequency for this speaker’s the angle off axis of +- 6dB Frequency Response is near +-55degrees near the same as the Controlled Dispersion angle of the Compression Driver with attached waveguide. An engineer selected this crossover frequency. Too bad both woofer and CD ran out of gas below and above the crossover frequency.

The size of the room and the selection of the +-6dB angle of coverage are critical to the proportions of direct sound and reflections at the sweet listening seat on the couch.

For intermediate listening distances (2 meters plus) I like Controlled Dispersion angles in the range of 100 to 120degrees. I toe in the speakers to cross the axis of speakers in front of the listening position. If I move left I am more off axis of the left speaker and the direct SPL of the left speaker is reduced. Also as a result of moving left I am farther from the right speaker but closer to the right speaker axis, the right speaker SPL increases. Moving to the right all the opposite things happen.

With 110 to 120 +-6dB degree coverage you can get up and move around, the illusion of being there seems to be real.

I have done this with 60 degree, 80 degree, 90 degree and other Controlled Dispersion Waveguides.

I can tell you that controlled dispersion is important to the illusion of being there.

Your Mileage May Vary. You know personal preference and all that.

Thanks DT

 

[DSJR]

It was the high level I often listen at that they couldn't cope with at Paul Green's in Bath. I probably played Frank Zappa and Mahler!
The Nightingale NM point 5 which had that popular mid unit Peerless K040-MRF also fitted to the Gale GS 401 (and with a 1" dome tweeter and 8" bass unit) was much, much more to my taste and I ended up buying them! They were a small version of the NM1 which had a bigger cabinet and KEF B139 bass unit.
Just put the Halleluwah live recording from 1972 on...

A pal of mine had NM .5's I remember. He went to the dark side (Isobariks) before redeeming himself somewhat (ATC 100A's). Now retired to Devon, the ATC's are jammed in corners and don't work, but as he's not really into 'HiFi' any more (golf and travelling are more his 'thing' I gather), I dare say he'll significantly downgrade. I've made suggestions but no reply so we'll see... 'Cottagy' small sitting rooms are all very well and cozy in the winter, but bloody 'ell, they're annoying for decent audio gear - and I too prefer the effortlessness of a big low distortion box rather than squitters plus subs which takes up even more room - sorry!...

 

[DSJR]

I have seen loads of comments over the years from the first watt matters most and we don't use as much power as we think (Pass et all the SET lovers) and you need far more power than you realise (Sanders) but using an amp with a calibrated power meter shows far, far more power being used than I would have ever guessed in this 6 year old video I only saw recently. It is also amazing how much less a bit of chamber music used.
I suspect there is a clue to Amir's enjoyment of these in here.

The issue here is a three way passive speaker of only moderate efficiency and a 'nice' tonal balance in possibly a large space. I'm not doubting the evidence there, but I remember driving that model with 200WPC or so (8 ohms - 300WPC at 4 ohms) and they didn't 'sound' loud at all until I tried to talk to a colleague and had to all but shout. Now, is this the promoted clean natural sound of the speakers used, or are they compressing the dynamics slightly into a pleasant 'wall' of sound? I'm a fan of this brand, but need to be a little bit of a devil's advocate to put some balance in. I only heard the 4367's briefly but fell in love with them, the feeling being that low stands, a few metres of listening distance and again, plenty of power to override the passive crossover would pay dividends.

I have memories of the slim small Meridian M33 active speakers. The sound to me was somehow contrived and 'all of a level' until one sat and got used to them. very odd as they were actually very clear I remember.

There's just something about low distortion large bass drivers hardly working that's so damned attractive.

 

[Frank Dernie]

What's strange to me in that video was that at 500W/bass heavy music I couldn't see bass membrane moving. From my expirience with that kind of power and music there should be at least 1cm membrane excursion.

Indeed but the video quality was astonishingly poor, so maybe it was moving further but IME the model 40s are overblown in the deepest bass, and I didn't hear that characteristic sound so maybe the power was in the mid bass. Anyway both the amount of power being used and the difference between the electronic music and the chamber music was more than I would have expected, The model 40s are not particularly efficient though, 86dB for 2.83V.

 

[Robbo99999]

You would assuming the test is repeatable. But that aside, to the extent all speakers show some deviations, how do rank them?

How to rank them? I can imagine doing say 75dB / 85dB / 95dB, if speaker A shows identical frequency response in all but 95dB and speaker B shows identical at all 3 levels then speaker B is better than speaker A from that point of view. I don't really see why you'd need to rank them on that variable though, could just be another measurement that folks can use to assess the qualities of a speaker and if it suits their usage needs or not. I'm not really the advocate for this test though, I was initially commenting on another persons post on how to determine/test for such variations.

As to assuming the test is repeatable, I don't know how reasonable that is. You'd think that at the same room temperature and after a period of warm up that it would be repeatable, but I'm not sure. Obviously you'd keep your measurement mic in exactly the same position for all tests, so that shouldn't be an issue - I know if I don't move the mic at all when measuring my speakers then the frequency response is identical each time when run at the same level.

 

[Robh3606]

My 4-ways average between 105 and 110. Please reset your scale.

cheers

That's way up there. How do you manage that? You running full range horns??

Rob

 

[Robh3606]

You would assuming the test is repeatable. But that aside, to the extent all speakers show some deviations, how do rank them?

As previously posted you can measure in 10db increments and see how much power compression there is. They have been doing this for years with PA drivers.

Rob