• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Revel F328Be Speaker Review

When placed too close to you, the voices sound like they are way above you
I can hear this if I'm sitting on the edge of my couch too close to my f208s
 
Suggest dont be so negative look up F3 or F6 low end roll off for floorstanders and instead trust Revel designers know what they are doing based their research on subject so probably deserve a user smile instead of those negative user notes, reason is for example for book shelve category the designer cant know whatever distance to any or the closets boundary will be but actual for floorstanders they absolute know precise the distances to nearest boundary is the floor and can calculate a most reasonable roll off then that would sound pretty realistic right without using subwoofers.

In below visual there is calculated floor boundary gain for woofer most close to the floor then middle woofer then upper woofer and finaly some pressurerization room gain with 50% leakage for a room having longest room dimension of 20 feet, think animation suggest for the clinic anechoic F328Be curve there will be good 30Hz low end reach covered before more room boundary's probably at longer distances and nasty room interferences add their finale signature to acoustics..

Boundary_and_roomgain_x1x2_800mS_EDIT_2.gif
 
Last edited:
For example, compare the areas circled in red on the left and right images, then compare the areas circled in blue. Look at the broad(low Q) peaks and valleys, where they are, and how deep/high they are. If you ignore all the high Q jaggies of the graph on the left, it's almost like you're looking at the exact same speaker. The image on the left - to me - looks like a higher resolution copy of the image on the right. Is it possible that all the extra jagginess(and most of the lower Olive score) is attributable to the extra measurements that Amir took for this speaker?
Very plausible. And that's an argument for standardizing on a number of measurement points.
 
Guess we need another f208 remeasured:cool:

Honestly, I wouldn't mind :D. Most would probably be opposed, though, and understandably so.

It would be interesting to me just to hear Amir's impressions of how the two compare. Also, he didn't provide the beam width graph for the F208 so it's tough to compare the dispersion. @amirm is it still possible to provide the beam width image for the F208? Or would you have to remeasure to get that? Also, you touched on it in the review, but could you provide any more details about the Salon2 comparison?

The spin for the F328Be actually looks a little better than the Salon2, but based on what I've heard(from Revel engineers), the Salon2 still mostly beats it in their blind tests. My guess is that the off axis graphs that @MZKM provides might look a little better on the Salon2 than they do for this speaker. Enough to offset the slightly less flat LW. Would love to hear more of your subjective thoughts, though, as the Salon2 is a speaker I've been very close to purchasing at least a couple times.

I would love to see a blind listening test between the F208, F328Be, and Salon2. I bet it would be fairly close.
 
Last edited:
Dsp. No passive can equal a top tier DSP . that said just use some room eq and DSP with these to tailor even further.
Still impressed by the KEF R3 measurement for a top passive and would love to see a R series or even KEF Refrence series floorstander on these pages.

+1 for those KEFs & while we are wishing, I would add the KEF Blades in that list too.
 
Nice to see a floor standing speaker join the mix. It’s what I‘ve been waiting for.

Congratulations on your 100th speaker test.

Hopefully you can find a friendly neighbour to assist with the moving task and we see more of same. Maybe some top of the line active monitors from, ATC, Genelec, and Neuman.
 
$16,000 for a pair...Why can't I get scenes from the Great Gatsby out of my head??? :)
 
So this speaker isn't too bad. It only fell .16 short of the JBL 308 mkII in the formula for scoring. Looks like the 308 goes flatter into the bass a ways.

If you add a sub to the Revel it slightly exceeds the JBL 308 score.

So yes, that formula still bothers me.
 
Last edited:
Few speakers have left such a lasting impression on me than those Dynaudio Consequences. I think it was the mid 80s when I heard them at an audio show. They were about $15,000pr at the time IIRC. They were just amazing and re-scaled domestic loudspeaker reproduction for me going forward.

Never heard a pair of Jamo Oriels, but their 707ii and 507a and the Concert V/VII of the early 90s were based on that research and design. Internal twin woofers to about 120Hz, ports and bass/mid/treble front facing. Low distortion designs where the woofers are true woofers, not trying to do midrange as well. This Revel has what appear to be woofers, mid woofer, midrange and treble and also very low distortion.

There are also plenty of speakers where the woofer sits above the tweeter in small bookshelf designs. Think older Mission, Jamo etc.
I heard the Consequences as well and thought they were among the best (along with the original Appogees and the Avalon Eidolons) I've ever heard.
 
100 speaker reviews! Wow does this mean you've been averaging one every 3 days this year? Even paid reviewers don't go at this rate I think.
It is hard for me to accept that rate as well. Then again I do a speaker review every other day or so. That it averages 3 days sounds about right.
 
Guess we need another f208 remeasured:cool:
Yeh, when I have copious amount of free time, I will need to go and remeasure the first set of speakers. As if that is ever going to happen.... :)
 
Suggest dont be so negative look up F3 or F6 low end roll off for floorstanders and instead trust Revel designers know what they are doing based their research on subject so probably deserve a user smile instead of those negative user notes, reason is for example for book shelve category the designer cant know whatever distance to any or the closets boundary will be but actual for floorstanders they absolute know precise the distances to nearest boundary is the floor and can calculate a most reasonable roll off then that would sound pretty realistic right without using subwoofers.

In below visual there is calculated floor boundary gain for woofer most close to the floor then middle woofer then upper woofer and finaly some pressurerization room gain with 50% leakage for a room having longest room dimension of 20 feet, think animation suggest for the clinic anechoic F328Be curve there will be good 30Hz low end reach covered before more room boundary's probably at longer distances and nasty room interference adds their finale signature to acoustics, heights of woofers relative to floor is not precise but a reasonable eye balled estimation..

View attachment 92665

+1. My HDI 3600s, which have vaguely similar low end response per Amir's Klippel measurements, have an F3 of ~27Hz in room running full range. Even my old LS50s, up on 30" stands, managed to hit mid 40s in this room before farting out. Now that I'm running those same LS50s near field in my office, extension to even 80Hz seems to be a struggle, but at the same time I have to cut a good 6-10dB from 100-250Hz because of reflections from my desk. Boundary gain is a thing.
 
This is a review and detailed measurements of the Revel F328Be floor standing speaker. I purchased a single unit through our company (Madrona) with kind discount from Harman Sales (I asked for a review sample from the PR group but was told "don't call us, we will call you."). The retail cost of F328Be is US $8000 each or $16,000 for a pair. This makes it the most expensive speaker we have tested to date.

I chose the color black and I must say, this is one gorgeous speaker:

View attachment 92617

It comes in an exquisite finish with sparkles in it in the top panel:
View attachment 92618

Even the Beryllium tweeter exudes beautify (in the eye of an audiophile of course):

View attachment 92619

Here is a closer shot of the speaker itself:
View attachment 92620

As you see, there are three 8 inch woofers backed by two large ports in the back. There is then a mid-range and tweeter.

Measurements that you are about to see were performed using the Klippel Near-field Scanner (NFS). This is a robotic measurement system that analyzes the speaker all around and is able (using advanced mathematics and dual scan) to subtract room reflections (so where I measure it doesn't matter). It also measures the speaker at close distance ("near-field") which sharply reduces the impact of room noise. Both of these factors enable testing in ordinary rooms yet results that can be more accurate than an anechoic chamber. In a nutshell, the measurements show the actual sound coming out of the speaker independent of the room.

I performed over 2000 measurement which resulted in error rate of less than 1% throughout the range. This took over 4 hours to capture all the data which resulted in a 1.5 gigabyte file (imagine doing this work manually!).

Temperature was 60 degrees F. Measurement location is at sea level so you compute the pressure.

Measurements are compliant with latest speaker research into what can predict the speaker preference and is standardized in CEA/CTA-2034 ANSI specifications. Likewise listening tests are performed per research that shows mono listening is much more revealing of differences between speakers than stereo or multichannel.

Reference axis was between the midrange and tweeter (a guess).

Revel F328Be Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker is and how it can be used in a room. This so called spinorama shows us just about everything we need to know about the speaker with respect to tonality and some flaws:

View attachment 92621

This speaker was very difficult to measure in low frequencies. The combination of two ports and three woofers created an overly complex low frequency sound field. The initial runs produced clearly incorrect measurements below 200 Hz. After five sets of measurements and optimization, I dialed out almost all the error. The results show that little step between 100 and 200 Hz which Harman data does not. I suspect with a bit more optimization that could have been filled in a bit more. This aside, for a passive speaker we have excellent flatness of response especially considering how many sound producing elements there are here.

Efficiency is superb for a Revel speaker but do keep in mind that we need power in low frequencies the most and there, we don't have nearly the 91 dB that we have in the upper range. This of course applies to all speakers.

Look at how the dashed blue line (early window) nicely parallels the black dashed green line indicating excellent directivity (direct and indirect sounds match). This makes the speaker room friendly as the reflections don't change the tonality of the speaker much. And also allows easy equalization. Let's look at the sum of important early reflections:

View attachment 92625

Amazing that all the varying reflections sum to such a smooth response. Putting everything together, we get a prefect predicted in-room frequency response:

View attachment 92626

It took me forever to measure all the drivers and ports in near-field. Turns out most of them were duplicating each other's response so here is a sample set:

View attachment 92627

Notice the port tuning down to 30 Hz! This speaker should produce very deep bass. And clean too when you look at these incredible distortion measurements:
View attachment 92629

Are you seeing what I am seeing? Even at 96 dB SPL, you hardly have any bass distortion. Indeed during measurements, I could hear the most perfect sweep ever of any speaker I have tested. Most of the time I hear the pitch change/distort during the range. Not here. The sweep was so buttery smooth. Yes, that is an objective technical term!

Here is the distortion in absolute level:

View attachment 92630

I picked 50 dB as a rather arbitrary limit long time ago with speaker after speaker missing that mark in low frequencies. Not the Revel F328Be. With the exception of one frequency, it stays well below that.

Horizontal beam width is very smooth:

View attachment 92631

Notice how the line in pink is also smooth (-12 dB).

Shown as a contour map, we see the same nice response:

View attachment 92632

Due to inclusion of a mid-range, the vertical response is also good compared to our typical 2-way speakers:

View attachment 92633

Since this is a tall speaker, you may want to tilt it down a bit if you are sitting too close to it.

Finally, here is our impedance measurement:

View attachment 92634

That is a pretty low impedance so make sure you have a good, high-current amplifier to go with this speaker.

There is a small resonance visible but I don't see a sign of it in the frequency response so it must be quite minor.

Revel F328Be Listening Tests
Truth to be told, I measured this speaker weeks back. But then it sat in the garage because I could not figure out how to carry it to our loft to listen to it. I almost post just the measurements but then I thought questions of sound and comparison against my Revel Salon 2 would come from every angle. Asked my wife if should could help me drag it upstairs and she said no way as did my back. Didn't know what to do for a while until I realized there was another option: create a setup in our living room which is at the same level as the garage. So I built a quick system out of Matrix Element i streamer and DAC plus Purifi Amplifier. The picture you saw at the start of the review was the setup except I had the speaker further into the room.

First impression was: "this is the sound I am always looking in a speaker." It sounded very familiar to me as it should. What surprised me though was the deep bass. I had not heard such clean and deep bass out of any speaker I have tested so far. My own Salon 2 speakers produce them of course but nothing I had tested would come close. The F328Be was there. It is a wonderful kind of bass you don't get out of a sub in how clean and integrated it is. It doesn't jump out to slap you in the face. At the same you wonder why you had never heard these notes with smaller speakers.

I ran through my "speaker killer" tracks which are full of these sub bass frequencies and the F328Be produced them like it was an everyday affair. Even at super elevated levels, the sound was exceptionally clean. No hint of bass distortion let alone bottoming out. I could see the triple woofers moving but no deep excursion to make them distort.

The one sensation you should know about is that these are tall speakers. When placed too close to you, the voices sound like they are way above you. In a way this is a natural presentation of the size of a singer. But at the same time, you don't want it to keep reminding you so. Push them out and this effect reduces.

I actually listened to the F328Be out in the open first before pushing it into the corner. The latter did cause a bit of boominess due to room mode. I don't have current measurements for our living room so I did not try to dial them out. It is the nature of speakers that play low that they excite room modes more, requiring DSP correction for best sound.

So do the F328Be speakers outperform my Salon 2? With no ability to listen to them in the same spot, I can't say for sure. I can say that at no time I wished I had the F328Be instead of my Salon 2. Its sound simply asymptote up the level of Salon 2 but didn't exceed it. You are spending $6,000 less with 328Be so coming this close is impressive.

In some sense I am the wrong guy to review this speaker, coming from Salon 2. It was impossible to impress me with a speaker with similar tonality and performance.

Conclusions
It was great to test the premier speaker in Harman's newest line. The F328Be is in entirely different class than any other speaker we have tested. The difference comes in impressive bass performance that is essentially distortion-free. This brings a level of clarity and dynamics that is addictive. You can play as loud as you possibly want and the speaker comes along happily. No sign of any change in speaker response. It is just at home playing audiophile tracks and rock and roll.

Needless to say, I am happy to recommend the Revel F328Be. If you want to buy a set of speakers and be done with it, you know what to get now!

------------
As always, questions, comments, recommendations, etc. are welcome.

This was our 100th speaker review and measurements!!! Hooray!

motion-graphics-number-animation.gif


Hard to imagine testing so many speakers since January. When I bought the Klippel NFS measurement system, I was wondering how it would ever pay for itself. I figured if I tested 100 speakers, it would still cost $1000 per so we could have just asked a service to do the work. But here we are in less than a year and we are already passing that benchmark. Here is to another 100 speaker reviews!

Appreciate any donations using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Thanks for your efforts.

Here some thoughts on the EQ and a comparison with the F208 and the official data from Harman

Spinorama no EQ:
Score: 5.37 with sub: 7.74
The bass and the 200Hz bump looks strange quite different from the Harman data.
Especially considering that @amirm seems to really like it.
What I especially like is the directivity down to about 200Hz that no bookshelf can manage (unless beam forming or other solutions are applied).
I wonder if that has something to do with the impressions of Amirm. The 200Hz bump/shelf removed it would make the PIR much closer to the Harman in-room target.

Revel F328Be No EQ Spinorama.png

Directivity, better stay close to the axis of the tweeter
Horizontal 5k notch
Vertical: better stay at tweeter height

Revel F328Be 2D surface Directivity Contour Only Data.png

Revel F328Be 3D surface Vertical Directivity Data.png

Revel F328Be 3D surface Horizontal Directivity Data.png


EQ design:
I have generated two EQs. The APO config files are attached.
  • The first one LW is targeted at making the LW flat.
  • The second, Score, starts with the first one and adds the score as an optimization variable, i.e. looks into the PIR.
  • The EQs are designed in the context of regular stereo use i.e. domestic environment.
  • This, I believe, might show the limit of the LW only EQ as the off axis suffers from this hard target as seem on the PIR.
  • The mixed Score/LW optimization, the EQ score, process is much closer to the out-of-the-box speaker and, assuming Harman did their double blind tests, seems more reasonable. I could be wrong...
Score EQ LW: 5.88, with sub: 7.69
Score EQ Score: 6.42, with Sub: 8.23

APO Config files attached, although I don't have much hope that someone will report on their performance...

Code:
Revel F328Be APO EQ LW 96000Hz
November112020-134749

Preamp: -3 dB

Filter 1: ON PK Fc 53 Hz Gain 3 dB Q 0.42
Filter 2: ON PK Fc 225 Hz Gain -1.25 dB Q 4.18
Filter 3: ON PK Fc 1345 Hz Gain 0.83 dB Q 2.98
Filter 4: ON PK Fc 1841 Hz Gain 1.15 dB Q 5.85
Filter 5: ON PK Fc 4131 Hz Gain 1.91 dB Q 2.51
Filter 6: ON PK Fc 7746 Hz Gain 1.11 dB Q 4.27
Filter 7: ON PK Fc 17370 Hz Gain 2.24 dB Q 0.42

Revel F328Be APO EQ Score 96000Hz
November112020-110528

Preamp: -3 dB

Filter 1: ON PK Fc 50 Hz Gain 3 dB Q 0.41
Filter 2: ON PK Fc 203.5 Hz Gain -0.66 dB Q 5.2
Filter 3: ON PK Fc 1858 Hz Gain 0.89 dB Q 3.96
Filter 4: ON PK Fc 2418 Hz Gain 1.38 dB Q 12.75
Filter 5: ON PK Fc 4322 Hz Gain 1.53 dB Q 5
Filter 6: ON PK Fc 7422 Hz Gain 1.11 dB Q 5.27
Filter 7: ON PK Fc 26824 Hz Gain 2.42 dB Q 0.59

Revel F328Be EQ Design.png

PIR-LW-ON Zoom
Revel F328Be Zoom PIR-LW-ON.png

Regression - Tonal
Revel F328Be Regression-Tonal.png

Spinorama EQ LW
Revel F328Be EQ LW Spinorama.png

Spinorama EQ Score
Revel F328Be EQ Score Spinorama.png

Radar with EQ Score
Revel F328Be Radar EQ Score.png

The rest of the data is attached

Now let's see how it compares with the F208.
The resolution of the F208 data seems lower so I interpolated the F328Be data to match that of the F208 to have a better comparison.
However doing so still yields a less smooth data set, I am not quite sure how the NFS export/measure at different resolutions.

First with different resolution:
There is a definite family resemblance but the F208 exhibits much more LF output.
The F328Be as a smoother SPDI, better Xover?
The Bump/shelf seems shifted higher up, close to 300Hz, in the F208 data but still there, interesting.

Revel F208 vs Revel F328Be No EQ.png
With the similar resolution
Revel F208 vs Revel F328Be No EQ F208 Resolution.png

the EQed version with different resolution
Revel F208 EQed vs Revel F328Be EQed.png

I had to rerun the F328Be as the EQ was off when I changed the resolution.
Revel F208 vs Revel F328Be EQed F208 Resolution.png

EDIT:
@amirm Added the comparison with the data scanned from the published Spinorama
LF and HF have significant deviations.
Revel F208 Scan vs Revel F328Be NFS.png



I'll go back to check how the PIR compares with the different targets when I have time.
 

Attachments

  • Revel F328Be Reflexion data.png
    Revel F328Be Reflexion data.png
    254.8 KB · Views: 91
  • Revel F328Be Raw Directivity data.png
    Revel F328Be Raw Directivity data.png
    912.4 KB · Views: 101
  • Revel F328Be Normalized Directivity data.png
    Revel F328Be Normalized Directivity data.png
    503.1 KB · Views: 102
  • Revel F328Be No EQ F208 resolution Spinorama.png
    Revel F328Be No EQ F208 resolution Spinorama.png
    153 KB · Views: 90
  • Revel F328Be LW data.png
    Revel F328Be LW data.png
    265.9 KB · Views: 95
  • Revel F328Be 2D surface Directivity Contour Data.png
    Revel F328Be 2D surface Directivity Contour Data.png
    297.3 KB · Views: 98
  • Revel F328Be APO EQ LW 96000Hz.txt
    394 bytes · Views: 109
  • Revel F328Be APO EQ Score 96000Hz.txt
    396 bytes · Views: 107
  • Revel F208 Scan vs Revel F328Be NFS Radar.png
    Revel F208 Scan vs Revel F328Be NFS Radar.png
    75.5 KB · Views: 107
Last edited:
Very nice speakers.

Also very interesting reading the comments, particularly about the distortion.


Coincidentally I measured my old new speakers on the weekend. The good speaker (long story about the bad speaker) is well behaved at 80dB, and not too bad even at 90dB, obviously it probably goes up quite a lot at 98dB.....

However, the Revel F328Be's retail for AUD$31,000 a pair. My 25 year old JVC speakers second hand cost me between AUD$5,000-10,000.

I must say I am quite impressed that something so old and used measures reasonably........ What do you think @waynel?
 

Attachments

  • DISTORTION AT 80Db.jpg
    DISTORTION AT 80Db.jpg
    722.3 KB · Views: 171
  • DISTORTION AT 90dB.jpg
    DISTORTION AT 90dB.jpg
    780.9 KB · Views: 157
The bass and the 200Hz bump looks strange quite different from the Harman data.
Especially considering that @amirm seems to really like it.
As you know, the in-room bass response has little to do with anechoic so what I am hearing is not what we are seeing.

On Harman comparison, yes, I could not match their low-end. This has been a recurring problem. I have begged them multiple times to send me one of their samples so that we can go back and forth until we get to the bottom of this but they have not come through.

As I mentioned, I have a way to optimize the low frequency measurements a bit more. I just dread having to put this monster on the stand and spend the 4+ hours to measure it again.
 
this is entry product of high-end, lol
It is actually invisible to folks wanting high-end speaker. It is too rational of a design and way too cheap to be in play. Even my Salon 2 is scuffed at.
 
Here some thoughts on the EQ and a comparison with the F208.

Spinorama no EQ:
Score: 5.37 with sub: 7.74
Are you able to stich in the Harman measurements in bass or all around to see what the score would be?
 
Back
Top Bottom