JBL 4309 Review (Speaker)

[Spocko]

The question was never only what produces the best measurements. The methodology is that measurements show accuracy and, when on the Harman curve for in-room performance, interact with the room to produce the most pleasing results. All the best research is based on subjective impressions by listeners, often trained, in controlled listening tests.

The danger that this site helps avoid is getting sucked into the eternal upgrade cycle. A world class headphone stack is now relatively cheap. A great speaker, amp, preamp, DAC stack can be done for 3-4k, and much less if you're willing to use powered speakers. Want to go down to 20hz? Then it costs a bit more as you need a great sub.

Actually, I firmly believe that the "eternal upgrade cycle" is part of many audiophile's journey - it's an essential component that holds their interest. Buying new gear is in and of itself exciting - reading the reviews, doing the research, setting it up, upgrading other components to match - this is something many actually enjoy.

 

[Doodski]

Actually, I firmly believe that the "eternal upgrade cycle" is part of many audiophile's journey - it's an essential component that holds their interest. Buying new gear is in and of itself exciting - reading the reviews, doing the research, setting it up, upgrading other components to match - this is something many actually enjoy.

It's primeval behavior for sure.

 

[hardisj]

It's definitely possible I'm wrong, and that's part of why I'm responding.

FWIW, I pretty much agree with everything you've said.

 

[ROOSKIE]

Here is a side by side comparison with a $79 speaker that you gave a 0/5, and described as "fairly uneven" and with "multiple resonances".


View attachment 158691

That's a $2,000 speaker you rated 5/5 against a $79 speaker you rated 0/5. The commentary is also very different. My point is, subjective impression rules the day when it comes to panther score. Frequency response, price, etc. are obviously a much smaller factor. Like I said, I like it this way. We essentially have both a subjective(panther) and objective(Olive) score for every speaker.

I have spent a fair amount of time with that Sony speaker. On sale for $75 a pair, it is totally fine vs a crappy sound bar or similar .
Otherwise I think it sounds bad. Especially when turned up anywhere above "medium loud".
I can barely enjoy it and would never be happy with.
I would take everyother speaker I have owned that is "hifi" oriented over it(many). Granted it is the least expensive suggested retail wise but some were close in price.

 

[Newman]

You still have a big hump at 110Hz

It’s 2 dB at 110 and 1 dB by 100 and 150 Hz. “Big hump?”

a 2dB magnitude fairly wide depression from 200-500/600Hz

It’s 2 dB at 300 Hz and 1 dB (!) by 250 Hz and 320 Hz - that’s not wide.
The PSY curve I posted is only 1 dB at 300 Hz and a fraction of a dB by 250 and 320 Hz.

Take a look at my PSY curve of the FR to maybe put it into subjective context.

The thing I would quibble with @amirm about is letting a loudspeaker off the hook for engineering imperfections that are maybe not damaging the sound (based on one sighted audition), but not letting electronics off the hook for engineering imperfections that are almost certainly not damaging the sound. He justifies the electronics headless panther with “this could have been so easily avoided”, and I think that’s reasonable, but engineering defects in a speaker, not a cheap one either for its size, still get a soccer-ball panther if it sounds good in a sighted listen.

 

[Newman]

Also let’s not forget that @amirm did confirm that EQ-ing the 1000Hz and 1800Hz areas did improve the subjective SQ.

YET AGAIN confirming the science that points to smoother FR being preferred. So, people who claimed, earlier on, that the good subjective sound ‘naked’ is evidence that smoother FR is not that important, really need to get a grip.

 

[napilopez]

For one, I don't think it's fair to say ~"so remove a 4dB positive magnitude 300-400Hz wide resonance" right at one of the most critical sections (800-1100Hz). I think a lot of speakers could look quite a bit smoother if you remove their biggest flaw.

Not commenting on the rest of your comment to avoid being caught in a debacle lol, but just thought I'd bring up the peculiarly consistent impression I have that anomalizes from roughly 700-Hz to 1500 Hz are the ones that least bother me. I don't know why. But speaker after speaker issues around here just don't seem to affect me.

That may be complete hogwash. I have no real explanation for it. But I notice deviations below roughly 700 Hz and above ~1.5 kHz more than anything happening in between. It's very weird. I do however notice that I can hear more a broad hump around here than I do a combination of dips and peaks.

The only other thing I'd caution is that I'm not sure much of the things people are calling resonances are actually resonances, as in with audible ringing in the time domain. You can clearly see several small resonances in the sony, but the more jagged response of the 4309 makes it hard to definitively call much a 'resonance.' I normally only consider it an issue when the resonances rises clearly over the surrounding response, even if its a small blip. If its peaks and dips it's harder to tell and arguably harder to hear.

That's where the impedance graph and slow sine sweeps are handy. Note the fairly clean impedance graph except for at about 320 Hz.

Also not trying to defend choosing this speaker over another with cleaner measurements. but at the same time I feel it would be ridiculous not to acknowledge that I had a clear preference for this speaker over a similar measuring prettier one from harman itself, and I think it's always good to discuss what deviations are problems and which are less so.

Somewhat to @Newman's point, it's easy to say a speaker is likely to be good and enjoyable by most people when the response is textbook perfect. But when it's not, I think it requires extra care to work out which things matter most.

 

[richard12511]

Not commenting on the rest of your comment to avoid being caught in a debacle lol, but just thought I'd bring up the peculiarly consistent impression I have that anomalizes from roughly 700-Hz to 1500 Hz are the ones that least bother me. I don't know why. But speaker after speaker issues around here just don't seem to affect me.

That may be complete hogwash. I have no real explanation for it. But I notice deviations below roughly 700 Hz and above ~1.5 kHz more than anything happening in between. It's very weird. I do however notice that I can hear more a broad hump around here than I do a combination of dips and peaks.

That's super interesting, and really what I'm curious about.

The only other thing I'd caution is that I'm not sure much of the things people are calling resonances are actually resonances, as in with audible ringing in the time domain. You can clearly see several small resonances in the sony, but the more jagged response of the 4309 makes it hard to definitively call much a 'resonance.' I normally only consider it an issue when the resonances rises clearly over the surrounding response, even if its a small blip. If its peaks and dips it's harder to tell and arguably harder to hear.

By resonance, I'm simply going by something that shows up in all three of the top curves of the CEA-2034. You may be right that some of those aren't resonances. The points I listed as resonances simply occurred in all 3 curves.

That's where the impedance graph and slow sine sweeps are handy. Note the fairly clean impedance graph except for at about 320 Hz.

Also not trying to defend choosing this speaker over another with cleaner measurements. but at the same time I feel it would be ridiculous not to acknowledge that I had a clear preference for this speaker over a similar measuring prettier one from harman itself, and I think it's always good to discuss what deviations are problems and which are less so.

I think you may be misunderstanding the purposes of my posts somewhat. Both you and Amir have come to very positive conclusions about this speaker. I believe you. I'm coming at this from a different perspective. I'm looking at this as "the CEA-2034 suggests this is a poor speaker, but it's been given top marks by 2 very objective reviewers. What is the cause of this discrepancy?. What is the Olive score missing?" Current points are:

Overall tonality tilt is more important than number of resonances.
Directivity is more important than the Olive score gives it credit for.
FR errors from 700Hz to 1.5kHz don't matter as much(this is based on your first part)

I'm not trying to completely refute y'all's subjective opinions.

 

[richard12511]

FWIW, I pretty much agree with everything you've said.

Thank you!

 

[amirm]

Around 0.5dB sound pressure level difference I would define as dynamic compression, since such a level change, with a correspondingly small Q, is perceptible as a tonality change - this is simply my definition to be able to compare loudspeakers.

I don't believe dynamic compression is a factor in home speakers. PA speakers, sure. Home speakers? No. My measure of dynamics is how loud I can play before the speaker produces clearly audible distortion. This will dwarf any 0.5 dB change in frequency response. In bass frequencies the impact of room is so large that it dwarfs any such changes anyway (and for all we know, it may tilt toward sounding better, not worse).

Even for PA use, using flat response makes little sense for music because spectrum is not remotely flat. This is why folks have come up with m-noise and such.

Even if you believed this matters, you need to first have a controlled situation were you can assure that our two point measurements are actually producing objectively correct results. Just because there are differences in frequency response, it doesn't mean that compression has happened and to that degree, especially if you are zooming in so much as to worry about 0.5 dB differences.

 

[MaxBuck]

I'm guessing I wouldn't like the sound these things produce. Not that it matters; I'm not in the market for speakers.

 

[pozz]

FR errors from 700Hz to 1.5kHz don't matter as much

For what it's worth, this is roughly where our psychoacoustic mechanisms compete. ITD sensitivity falters, being strongest below 500Hz, and ILD sensitivity takes over, strongest above 1.5kHz. Localization ability takes a hit in between.

It's our very own crossover

 

[Inner Space]

Also let’s not forget that @amirm did confirm that EQ-ing the 1000Hz and 1800Hz areas did improve the subjective SQ.

YET AGAIN confirming the science that points to smoother FR being preferred. So, people who claimed, earlier on, that the good subjective sound ‘naked’ is evidence that smoother FR is not that important, really need to get a grip.

Yes, he said "The effect was so subtle that I had to resort to blind testing of the EQ to be sure it was making a positive difference" ... which is a fancy way of saying "not that important", I guess.

 

[beefkabob]

Actually, I firmly believe that the "eternal upgrade cycle" is part of many audiophile's journey - it's an essential component that holds their interest. Buying new gear is in and of itself exciting - reading the reviews, doing the research, setting it up, upgrading other components to match - this is something many actually enjoy.

Oh, that's sad. Engineer geeking is one thing. The method a DAC uses may be mentally pleasing. But on the magical sound quest, it's sad.

 

[Newman]

Yes, he said "The effect was so subtle that I had to resort to blind testing of the EQ to be sure it was making a positive difference" ... which is a fancy way of saying "not that important", I guess.

It’s generically true to say that audiophiles with high standards specialize in making mountains out of molehills, and those with somewhat lower standards say “it’s not that important” (your words).

Rather than judging audiophiles for having standards, IMO the real quest for truth in audio is more about determining whether the listening difference is objectively there at all, or whether personal sighted biases have distorted the impression and, in the case of reviewers, led them to mislead a whole audience, one hopes accidentally. ASR can do a great service here by helping us to root out the BS flowing from the keyboards of so many sighted listening equipment reviewers.

 

[Inner Space]

It’s generically true to say that audiophiles with high standards specialize in making mountains out of molehills, and those with somewhat lower standards say “it’s not that important” (your words).

Let's alter a couple of words: It’s generically true to say that audiophiles with academic standards specialize in making mountains out of molehills, and those with practical standards say "it's not that important."

Which is borne out here every day. Some guy mentions a SINAD of 120dB, ten other guys say it's not that important above 100, or 80, or honestly, in the real world, pal, you can get away with 60. Etc, etc. Good advice, generally. Good practical advice.

As is, within reason, FR doesn't matter as much as you think it does. Chasing the impossible dream of flatness is like chasing a SINAD of 150dB. We don't need to, anyway. It's not that important. Within reason, plus or minus single-figure dBs, with a lump or bump here or there, works perfectly fine.

I say so not because I'm stubborn, but because I'm thinking of it musically. All of some and some of all reproduced music depends on line, and flow, and touch, and nuance, all of which are impossible unless the brain can somehow compensate for violent FR variation unrelated to musical intent. Possibly you think different. I'm happy to disagree. Over and out.

 

[pozz]

I say so not because I'm stubborn, but because I'm thinking of it musically. All of some and some of all reproduced music depends on line, and flow, and touch, and nuance, all of which are impossible unless the brain can somehow compensate for violent FR variation unrelated to musical intent.

Kind of lost you there.

 

[Inner Space]

Kind of lost you there.

Think of a recording of a favorite musical figure - part of a cadenza, maybe, or a pedal steel guitar, moving sweetly up the octaves. That's how you experienced it and how you remember it. But what hit your eardrum was at least doubling and halving in intensity at unpredictable intervals along the way. Yet somehow we normalize it and perceive subtlety and nuance as if the FR swing has been subtracted. I'm assuming that part happens in the brain, because I'm not real clear where else it could.

 

[GWolfman]

Agree with what a few others have said, surprised by the low score and high praises. Interesting find.

 

[pozz]

Think of a recording of a favorite musical figure - part of a cadenza, maybe, or a pedal steel guitar, moving sweetly up the octaves. That's how you experienced it and how you remember it. But what hit your eardrum was at least doubling and halving in intensity at unpredictable intervals along the way. Yet somehow we normalize it and perceive subtlety and nuance as if the FR swing has been subtracted. I'm assuming that part happens in the brain, because I'm not real clear where else it could.

Oh I see. You're talking about rapid changes in envelope and spectrum. It makes sense. We are pretty sensitive to stuff like this. You can look it up in the neuro lit, where it's called frequency following response and envelope following response. Basically measurements of the auditory centers in the brain which show how closely the neural representations track the acoustical event. Turns out, pretty close.

I'd tend to agree about lumps and bumps on FR, but I think the real goal is knowledge in the end. How precisely we can understand things and how much it all matters if a bit of impracticality is allowed. Not easy questions. Giving a good answer is harder than making a buying decision, in that sense.