Revel PerformaBe F226Be Floorstanding Speaker Review

[richard12511]

Can you help me understand the max output stuff a bit more. 106dB long term max seems not very loud, or maybe I was just expecting much more than that. You mention that "Keep in mind that these speakers play louder than 106dB. The test setup and conditions are described in the appropriate section above. This simply gives me a way to provide an apples-to-apples comparison against other test subjects.". Is this caveat in reference to long term vs peak max spl? Do you think these speakers would be capable of reference level peak output(105dB) with a listening distance of 4m?

 

[napilopez]

JBL speakers don't have wide dispersion, just Revels do.

? When did I talk about dispersion width?

In any case, not totally true. JBL tends a bit narrower on their horn designs, but The JBL L100 and L82 both are fairly wide dispersion designs, and even the HDI-1600 maintains revel-ish horizontal dispersion despite the beefy waveguide.

Revel M106(MZKM's graph from Amir's measurements):

JBL L82(my measurements):

70 degrees off axis is about 5 dB down from 1K to 5KHz. This is an asymmetrical speaker so this is the 'far' side of the tweeter. The other side is a little wider from 2K up, but has a crossover dip:

Revel M16(MZKM's graph from Amir's measurements):

JBL HDI-1600(amir's measurement):

Pretty much the same as the M16.

 

[Vini darko]

Quality review thankyou. Seems you get what you pay for with these.

 

[sweetchaos]

Outstanding effort and incredible review, Erin!

 

[hardisj]

Can you help me understand the max output stuff a bit more. 106dB long term max seems not very loud, or maybe I was just expecting much more than that. You mention that "Keep in mind that these speakers play louder than 106dB. The test setup and conditions are described in the appropriate section above. This simply gives me a way to provide an apples-to-apples comparison against other test subjects.". Is this caveat in reference to long term vs peak max spl? Do you think these speakers would be capable of reference level peak output(105dB) with a listening distance of 4m?

There was no issue listening to them at 105dB at 4m. In fact, I believe I even mentioned that specifically in my subjective portion... you'd have to double check me.

The max SPL testing is a stress test. About 1 minute of multitone noise followed by 2 minutes of rest. The idea is to test for compression and distortion using this hot/cold procedure. The multitone signal is intended to mimic music but is more intensive than the majority of music people listen to. I linked a video of this testing in my review but also again below so you can see (hear) just what the signal is. As you can imagine, this signal can be more punishing for a speaker than regular music would be. Even though it is more stressing than most music, it is part of the IEC standard for rating max SPL and I find it's a good way to come up with a single output level without having to look at different distortion measurements. And the idea is that anyone can take my methods and apply them to their own testing and should be able to come up with roughly the same numbers.

 

[theyellowspecial]

Thanks and great review. I hope Amirm includes the 360 dispersion plots in his future reviews.

 

[phoenixdogfan]

Add my voice to the many likes. When I get my Purifi Eval 1 working, these would be the match made in heaven. Add a miniDSP SHD Studio, and any good $200 balanced output DAC, and I could use my SB2000 and get a sound that's about as good as it gets in a 12 x 15 foot room.

 

[CDMC]

Really doesn’t get better than this.

 

[Chromatischism]

Any idea as to the cause of the vertical asymmetry?

 

[sfdoddsy]

Another class-leading review Erin, and more class-leading measurements from Revel.

I can't help wondering, however, if they would be even better with s smaller midrange unit.

Revel traditionally use a SB Acoustics 6.5 woofer in their bookshelf speakers like the M126Be.

Then in the three-ways they use 6.5 or 8" woofers and a 5.5 midrange. My query is why they don't go down to the 4.5" model SB make in the same series? Theoretically, it would be an even easier dispersion match for the tweeter and require a minimally higher crossover to the woofers.

My current three-way speakers use the same tweeter and bass drivers as the F208, but with the smaller SB Acoustics mid. They sound and measure very nicely without the waveguide and would probably do even better with it.

 

[Juhazi]

In an RMAF talk, John Atkinson mentions how since vinyl can have high amplitude ultrasonics, it can cause some IMD in the audible range with tweeters, mostly metal dome, that ring <30kHz. It wouldn’t be an issue with >44.1kHz music though as the ultrasonics are such low amplitude.

I'm much more worried of cartridge resonances and distortion.

 

[ctrl]

Any idea as to the cause of the vertical asymmetry?

If the speaker is not symmetrical in the vertical direction, the radiation in the vertical direction isn't symmetrical - this is quite normal.

The tweeter and also the midrange driver are located in the upper part of the baffle, so there is hardly any baffle at the top, but a lot of baffle at the bottom.
This leads to different edge diffraction and baffle interference and thus to different vertical angular frequency responses above and below the measurement axis.

Above the measurement axis of the tweeter there is only a small baffle, so there is a widening of the radiation there, because below the measurement axis of the tweeter the baffle prevents free sound radiation - (and if the measurement axis is between the tweeter and the midrange driver, the distance from the tweeter to the measurement microphone is minimal).

Below the measurement axis, the distance of the midrange and woofer to the measurement microphone is minimal and the angle of the drivers to the measurement microphone is optimal, therefore the sound pressure is slightly higher than above the measurement axis. Therefore the sound radiation in the low frequency range widens towards the bottom.

Conversely, above the measurement axis, the distance of the woofers to the measurement microphone is at a maximum and the measurement angle to the microphone is unfavorable. Therefore the sound pressure level is lower.

In addition there are all kinds of interferences due to phase shifts.

 

[stunta]

@MZKM Is it possible to generate the preference rating for this speaker from this review?

@hardisj Thank you for the super-human effort. Hats off to your dedication to audio science. I am not technical in this field at all, but have you at some point validated that your testing aligns with amirm's testing? IOW, if @amirm tested the same speakers, would the results be close enough?

 

[iec 60318-4]

Thank for your amazing review!

 

[tuga]

A bit too much presence, a high distortion peak at the top of the midrange passband and a tweeter that's shelved up by some 2dB though it very gradually steps down into line above 7kHz.
I was expecting better to be honest.

I wonder if that weird ripply response off-axis between 1 and 4kHz is the cone showing its ugly face or a diffraction/wave-guide issue.

It would have been interesting to see free-field measurements of the 5.25-inch midrange.

 

[MZKM]

@MZKM Is it possible to generate the preference rating for this speaker from this review?

@hardisj Thank you for the super-human effort. Hats off to your dedication to audio science. I am not technical in this field at all, but have you at some point validated that your testing aligns with amirm's testing? IOW, if @amirm tested the same speakers, would the results be close enough?

@pierre has it calculated from the Harman Spin, which is very similar except the bass extends a bit more; he got a 6.8 if I recall.

Erin's data files are a pain in the a** to work with, as each measurement angle is a separate file. If people request it and Erin provides it, I usually go through it for speakers that don't have score already, but since the Harman Spin exists and Pierre has already done the score on that, I don't feel the need.

 

[tuga]

The F228Be measured by JA also exhibits the same strange "ringing" behaviour between 1 and 4kHz:

https://www.stereophile.com/content/revel-performa-f228be-loudspeaker-measurements

 

[tuga]

The "ringing" is visible in Erin's impedance plot too:

 

[napilopez]

@MZKM Is it possible to generate the preference rating for this speaker from this review?

@hardisj Thank you for the super-human effort. Hats off to your dedication to audio science. I am not technical in this field at all, but have you at some point validated that your testing aligns with amirm's testing? IOW, if @amirm tested the same speakers, would the results be close enough?

He did so during his review of the buchardt s400. The results should be quite comparable. As extended testing has shown here, measurements are quite comparable across reliable sources.

A bit too much presence, a high distortion peak at the top of the midrange passband and a tweeter that's shelved up by some 2dB though it very gradually steps down into line above 7kHz.
I was expecting better to be honest.

I wonder if that weird ripply response off-axis between 1 and 4kHz is the cone showing its ugly face or a diffraction/wave-guide issue.

It would have been interesting to see free-field measurements of the 5.25-inch midrange.

Confused, where are you seeing tweeter shelving? The listening window, which is what revel optimizes for last time I checked, tilts down slightly. The on-axis presence peak disappears by 10 degrees off-axis, so it is essentially meaningless.

 

[tuga]

Confused, where are you seeing tweeter shelving? The listening window, which is what revel optimizes for last time I checked, tilts down slightly. The on-axis presence peak disappears by 10 degrees off-axis, so it is essentially meaningless.

The "presence" peak happens on-axis but I agree that one one doesn't have to listen on-axis.

The shelving is obvious in the PIR and Erin's actual in-room measurement, less so on-axis.