This is a review and detailed measurements of the JBL 104 Powered Monitor (speaker). I purchased this recently for just US $90 including free shipping from Amazon. JBL shows the list price as US $129.
These are hefty, egg-shaped speakers:
That's the master speaker that houses the amplifiers for both channels. Here is its back panel:
By computer speaker standards, they are a bit large. Compared to a JBL LSR305P Mark ii, they are quite a bit smaller.
They feel quite solid for something so cheap. Really, hard to imagine you are getting a brand name speaker with dual power amplifiers shipped to you for just $89. But do they sound good? Let's measure and find out.
Passive Speaker Measurements
Thinking there is just a passive crossover, I would just power the slave unit with my own amplifier. That way, I can set the voltage to it exactly to 2.83 volts as the CEA-2034 standard stipulates. So I did and ran the full suite of measurements. These are the spinorama measurements:
Boy, this is a disaster. No bass. No low mids. And highly irregular highs. I set one up to compare to the JBL 305 and my listening results don't match anything like this. I play clip after clip and it just doesn't jive with the measurements. It then occurred to me that the amplifier in the master unit may have an EQ it applies to what it feeds the drivers. So I measured it all over again which I now mark as the "104 active monitor."
Active Speaker Measurements
CEA2034 stipulates that you have to set the levels this way:
I can't go to 1 meter let alone 2 meters. So I set the level at 1/3 meter which translates into 95 dB. When I played this, it sounded a lot louder than I remembered the passive speaker levels. I don't know how they arrived at this being the equivalent level as the 2.83 volt SPL depends on speaker sensitivity and this one does not (since it is a measured value). Anyway, measurements are done and this is what I got:
Check out that big boos! Not only does it push up the bass but also good bit of the response to about 1.7 kHz where the crossover is. We have an on-axis cancellation around 7.4 kHz or so. A deep dive into polar plot near that frequency shows the problem:
And let's show off what our software can do by looking at that in 3-D all around the speaker:
Expected in-room response assuming you listen in far-field, would be like this:
The sound should be bright as we normally like to see this tilted down.
We are done with key metrics.
Advanced Speaker Measurements
Early reflections looks almost like estimated In-Room response:
The coaxial driver produces nearly identical vertical and horizontal dispersion which accounts for above:
And here it is at all angles:
Our horizontal and vertical slice contour plots show what we already know (and highlighting the uneven response):
Accuracy of Results
I was interested in measuring the JBL 104 because Harman has a frequency response measurement for it:
Yes, I know. Our measurements look nothing like this. Or do they? Look to the left: the scale is from 20 to 115 dB! That is a huge scale which helps to compress the graph a lot. We too can play that game:
Neat trick, ha?
CEA-2034 standard rightly requires 50 dB scale which is what I have used in this review and will do so in the future. That way we can properly see the variations in frequency response.
The JBL graph is quite smooth which tells me it lacks resolution and hence the fact that we have more exaggerated ups and downs but overall, the correlation is quite good.
Informal Listening Tests
The character of JBL 104 depended on the music you played. On some content, it sounded like a baby version of jBL 305P Mark ii. The latter of course had deeper bass and warmer sound but I was surprised how close the 104 came. On other content though such as female vocals and anything that had "lisping," the JBL 104 became too bright and annoying. This was after a moment of "ah, that sounds detailed."
Overall quality was tons better than junk computer monitors.
Conclusions
I don't know how we got fixated on testing so many JBL products but here we are. The JBL 104 drivers by themselves are quite poor. Heavy EQ in its amplifier though tries to make-up for some of that. Objective results are still nothing to write home about. Subjective listening depends on the content naturally as you either excite the peaks or don't. If you do not, it sounds alright and performs well above its price point. Can't recommend it though if you want a true hi-fi experience.
------------
As always, questions, comments, recommendations, etc. are welcome.
Drove 200 miles roundtrip in pouring rain and tons and tons of traffic to pick up 7 more speaker to test! If you think I am going to pay for all gas and time spent on the road, you are very much mistaken! Please pay up by donating using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
These are hefty, egg-shaped speakers:
That's the master speaker that houses the amplifiers for both channels. Here is its back panel:
By computer speaker standards, they are a bit large. Compared to a JBL LSR305P Mark ii, they are quite a bit smaller.
They feel quite solid for something so cheap. Really, hard to imagine you are getting a brand name speaker with dual power amplifiers shipped to you for just $89. But do they sound good? Let's measure and find out.
Passive Speaker Measurements
Thinking there is just a passive crossover, I would just power the slave unit with my own amplifier. That way, I can set the voltage to it exactly to 2.83 volts as the CEA-2034 standard stipulates. So I did and ran the full suite of measurements. These are the spinorama measurements:
Boy, this is a disaster. No bass. No low mids. And highly irregular highs. I set one up to compare to the JBL 305 and my listening results don't match anything like this. I play clip after clip and it just doesn't jive with the measurements. It then occurred to me that the amplifier in the master unit may have an EQ it applies to what it feeds the drivers. So I measured it all over again which I now mark as the "104 active monitor."
Active Speaker Measurements
CEA2034 stipulates that you have to set the levels this way:
I can't go to 1 meter let alone 2 meters. So I set the level at 1/3 meter which translates into 95 dB. When I played this, it sounded a lot louder than I remembered the passive speaker levels. I don't know how they arrived at this being the equivalent level as the 2.83 volt SPL depends on speaker sensitivity and this one does not (since it is a measured value). Anyway, measurements are done and this is what I got:
Check out that big boos! Not only does it push up the bass but also good bit of the response to about 1.7 kHz where the crossover is. We have an on-axis cancellation around 7.4 kHz or so. A deep dive into polar plot near that frequency shows the problem:
And let's show off what our software can do by looking at that in 3-D all around the speaker:
Expected in-room response assuming you listen in far-field, would be like this:
The sound should be bright as we normally like to see this tilted down.
We are done with key metrics.
Advanced Speaker Measurements
Early reflections looks almost like estimated In-Room response:
The coaxial driver produces nearly identical vertical and horizontal dispersion which accounts for above:
And here it is at all angles:
Our horizontal and vertical slice contour plots show what we already know (and highlighting the uneven response):
Accuracy of Results
I was interested in measuring the JBL 104 because Harman has a frequency response measurement for it:
Yes, I know. Our measurements look nothing like this. Or do they? Look to the left: the scale is from 20 to 115 dB! That is a huge scale which helps to compress the graph a lot. We too can play that game:
Neat trick, ha?
CEA-2034 standard rightly requires 50 dB scale which is what I have used in this review and will do so in the future. That way we can properly see the variations in frequency response.The JBL graph is quite smooth which tells me it lacks resolution and hence the fact that we have more exaggerated ups and downs but overall, the correlation is quite good.
Informal Listening Tests
The character of JBL 104 depended on the music you played. On some content, it sounded like a baby version of jBL 305P Mark ii. The latter of course had deeper bass and warmer sound but I was surprised how close the 104 came. On other content though such as female vocals and anything that had "lisping," the JBL 104 became too bright and annoying. This was after a moment of "ah, that sounds detailed."
Overall quality was tons better than junk computer monitors.
Conclusions
I don't know how we got fixated on testing so many JBL products but here we are. The JBL 104 drivers by themselves are quite poor. Heavy EQ in its amplifier though tries to make-up for some of that. Objective results are still nothing to write home about. Subjective listening depends on the content naturally as you either excite the peaks or don't. If you do not, it sounds alright and performs well above its price point. Can't recommend it though if you want a true hi-fi experience.
------------
As always, questions, comments, recommendations, etc. are welcome.
Drove 200 miles roundtrip in pouring rain and tons and tons of traffic to pick up 7 more speaker to test! If you think I am going to pay for all gas and time spent on the road, you are very much mistaken! Please pay up by donating using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
