James Loudspeaker QX806BE - In Room & Dirac corrected measurements

[Deleted member 49736]

I have a pair of James Loudspeaker QX806BE speakers in my kitchen. These are in-wall speakers with a quad 1" Beryllium tweeter array and dual 7.5" woofers. After a ridiculous amount of work, I was able to pull off the more complicated flush installation; standard installation leaves them sticking out 3mm. They are powered by a NAD M33 integrated streaming amp that uses a pair of the excellent Purifi 1ET400A amp modules. However, in this case, the most important feature of the M33 is Dirac Live multi-phase "room" correction. The speakers sound absolutely awful without correction. They reminded me of a PA horn. As you can see by the measurements, the crossover is very poorly executed. The subjective results after using the latest version of Dirac Live (V3.3.3) are, in a word, amazing. They magically transformed from midrange monsters to high bandwidth cream pies with that je ne sais quoi beryllium is famous for. Still, they're nowhere near Focal, Revel or Magico Beryllium equipped offerings I've heard. The quad tweeter array does have an endearing quality as a result of its extremely high efficiency. It also has great bass extension for it's size and driver arrangement, the sealed box helps with that. The box itself is a modern masterpiece of machined and welded aluminum. This thing is a hefty beast for an in-wall speaker at 35lbs each. The bottom line is, the speaker is so poorly tuned from the factory, it requires serious correction. If I was going to redo this project from scratch, I would go with the Sonus Faber Palladio PL-664 or B&W CWM7.3S2 in that order. That said, Dirac Live is incredible.

Red = Before
Orange = After

 

[HarmonicTHD]

I have a pair of James Loudspeaker QX806BE speakers in my kitchen. These are in-wall speakers with a quad 1" Beryllium tweeter array and dual 7.5" woofers. After a ridiculous amount of work, I was able to pull off the more complicated flush installation; standard installation leaves them sticking out 3mm. They are powered by a NAD M33 integrated streaming amp that uses a pair of the excellent Purifi 1ET400A amp modules. However, in this case, the most important feature of the M33 is Dirac Live multi-phase "room" correction. The speakers sound absolutely awful without correction. They reminded me of a PA horn. As you can see by the measurements, the crossover is very poorly executed. The subjective results after using the latest version of Dirac Live (V3.3.3) are, in a word, amazing. They magically transformed from midrange monsters to high bandwidth cream pies with that je ne sais quoi beryllium is famous for. Still, they're nowhere near Focal, Revel or Magico Beryllium equipped offerings I've heard. The quad tweeter array does have an endearing quality as a result of its extremely high efficiency. It also has great bass extension for it's size and driver arrangement, the sealed box helps with that. The box itself is a modern masterpiece of machined and welded aluminum. This thing is a hefty beast for an in-wall speaker at 35lbs each. The bottom line is, the speaker is so poorly tuned from the factory, it requires serious correction. If I was going to redo this project from scratch, I would go with the Sonus Faber Palladio PL-664 or B&W CWM7.3S2 in that order. That said, Dirac Live is incredible.

Red = Before
Orange = After
View attachment 223867

Just a hint. Remember Toole et al. states that a flat anechoic frequency response is preferred by most listeners (of their test group). That usually leads to a slight roll off (downward tilt of ca 0.5 to 1dB per octave) for in-room frequency responses. EQing to a flat in-room response usually is perceived as too bright or too little bass.

Just in case you want to explore an alternative.

 

[thorvat]

Just a hint. Remember Toole et al. states that a flat anechoic frequency response is preferred by most listeners (of their test group). That usually leads to a slight roll off (downward tilt of ca 0.5 to 1dB per octave) for in-room frequency responses. EQing to a flat in-room response usually is perceived as too bright or too little bass.

Just in case you want to explore an alternative.

Tilt of the in-room response curve actually depends on the listening distance. For shorter distances (1-2m) curve is pretty horizontal while on longer distances (3-4m) it is approximately as you stated. For even longer distances curve gets tilted even more. IIRC the estimated in-room response of CEA 2034 is given based on listening distance of 3m.

 

[HarmonicTHD]

Tilt of the in-room response curve actually depends on the listening distance. For shorter distances (1-2m) curve is pretty horizontal while on longer distances (3-4m) it is approximately as you stated. For even longer distances curve gets tilted even more. IIRC the estimated in-room response of CEA 2034 is given based on listening distance of 3m.

Correct

 

[heraldo_jones]

Dirac user since v1 here, best room correction I have ever tested, better IMO than the highly praised Trinnov hardware system.

 

[Deleted member 49736]

Just a hint. Remember Toole et al. states that a flat anechoic frequency response is preferred by most listeners (of their test group). That usually leads to a slight roll off (downward tilt of ca 0.5 to 1dB per octave) for in-room frequency responses. EQing to a flat in-room response usually is perceived as too bright or too little bass.

Just in case you want to explore an alternative.

It’s funny how often this comes up when I share my in-room measurements. The only time a tilted target curve sounds good to me is with low and mid-fi gear. Bright falls into the category of subjective terminology. Dome material plays a huge role in the perception of brightness. Electrostats are usually very flat, and almost never accused of being bright. Tweeter level being too hot and using EQ to target a flat response are very different things.

Dirac suggested a target curve with a -1.5db tilt on its “treble” adjustment. Having the benefit of storage for 5 Dirac filter presets. I noticed little to no difference in A-B testing between flat and -1.5db. The flat tube is nowhere near harsh, so it’s the logical tune to go with.

As far as bass, my DIY enclose for a JL 8w7 is in progress.

Tilt of the in-room response curve actually depends on the listening distance. For shorter distances (1-2m) curve is pretty horizontal while on longer distances (3-4m) it is approximately as you stated. For even longer distances curve gets tilted even more. IIRC the estimated in-room response of CEA 2034 is given based on listening distance of 3m.

Listening distance approx 11ft. A little over over 3m.

Dirac user since v1 here, best room correction I have ever tested, better IMO than the highly praised Trinnov hardware system.

It’s also better than I’ve ever been able to do with 10 band PEQ. I look forward to trying DLBC when I finally get a device that supports it. I also really want to try Dirac in a car stereo.