-
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 F208 Tower Speaker Review
- Thread starter amirm
- Start date
That's pretty good. You may want to try to adjust Dirac's target curve in order to push that wide dip around 200Hz to something like this:
Last edited:
I have bad experience with "correcting" that recession around 200Hz by EQ. This is typically consequence of several first reflection nulls, and right treatment is to move speakers and listening spot! EQ might lead to boomy, chesty, bassy sound.
It may be due to SBIR, in which case it won't react well to EQ attempt. In that case moving the speakers or the LP is indeed the only option to get rid of that. However, it seems a little bit too wide to be SBIR, so I'd give it a try with EQ and see what happens.
Good way to tell would also be to check if that dip is in the minimum phase region.
Last edited:
This looks really good, and I'm sure it sounds equally well!
Btw, that slight loss of energy around 520Hz is visible in left and right channel response you posted before so you may try to work that out too.
It sounded like it had too much bass. So I am giving it a go with the below, based on the +2 to -8db curve.
Response up to 2kHz is looking really good and I would leave it as it is. What may explain your impression of "too much bass" is actually treble roll-off after 2kHz which you may consider changing to something like shown on the graph below. This would add treble so overall tonal balance will be improved.
Looks very good to me, better than mine 
...for now
btw: to people obsessing on 0.5dB difference between Hypex and Purifi modules on the highs, look at that!
...for now
btw: to people obsessing on 0.5dB difference between Hypex and Purifi modules on the highs, look at that!
Looks very good to me, better than mine
...for now
btw: to people obsessing on 0.5dB difference between Hypex and Purifi modules on the highs, look at that!
it’s with the standard stereo Purifi from Nord and SHD, and I agree!
vavan
Senior Member
corrected link as couldn't edit original post
ehabheikal
Senior Member
- Joined
- May 22, 2020
- Messages
- 442
- Likes
- 161
This is a review and detailed measurements of the Revel F208 speaker. It was kindly purchased by a member and drop shipped to me for measurements. Revel F208 costs $2,500 each or US $5,000 for a pair.
Note: our company, Madrona Digital is a dealer for Revel speakers. Feel free to read as much bias as you like in this review.
The unit as tested is in piano black which gives a luxurious feel to the unit. Alas, the unit was too heavy for me to lug it up to my listening room so you have to settle for a plastic wrapped version of it in my test "lab:"
View attachment 62502
There was a sturdy sheet of plastic covering the front which I took off but left the rest of the plastic on it to protect it as you see in this shot:
View attachment 62503
I went through a bit of hell trying to figure out why the speaker would not plays the highs at first. After hours of trying to figure it out, and losing out a day, I realized it was a simple thing: the bar wires I was using would not always allow the shorting bar to do its job. I have to use a bare wire as to lower the interference around the speaker as the robotic arm moves the mic around. Anyway, I have to take the speaker to the owner in an hour so no time for listening tests either.
This speaker was the heaviest I have tested and despite having that wonderful lift you see in the first picture, it was a killer to unpack and move around.
I played with the bass compensation and it does what it says: if you set it to boundary, it lowers the bass level. I did not test the tweeter one but I suspect it does the same. An array of power resistors behind that panel makes these simple level changes.
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.
All measurements are referenced to the tweeter axis with frequency resolution of 2.7 Hz. I used a high number of scans (over 1000 measurements).
Spinorama Audio Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker can be used. 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 62506
If you look at Harman measurements, you see that they show a lot more bass output and smoother response:
View attachment 62507
After spending countless hours on the measurements, I am convinced Harman measurements like above are in error. My in-room response that doesn't use any of the Klippel mathematics is also in conflict with their measurements. If you look at the measurements for Revel Salon 2 speakers, you see that the older one performed in 2007 is the same as above but the 2017 one is very similar to what I got for F208: https://speakerdata2034.blogspot.com/2019/03/spinorama-data-revel-ultima.html
View attachment 62508
Notice the sharp drop in low frequency in the second measurement below. And bit of scalloping in high frequency response which I am also getting in F208.
I was puzzled that the on-axis response was not as smooth as I expect Revel speakers to have. That disappeared when I looked at the rest of the measurements starting with predicted in-room response:
View attachment 62509
Wow! With no DSP this speaker managed to get such balanced and smooth response. It appears that the off-axis response is tuned to counter the on-axis response as we see here:
View attachment 62510
This makes the speaker fairly room independent and sharply increases your chances of putting this in any room and enjoying accurate sound reproduction.
Distortion levels are very much controlled:
View attachment 62511
Ignore the rise in very low frequencies as that goes beyond the speaker's ability to reproduce. The rest shows no "acoustic events" other than a bit of wiggliness between 100 and 200 Hz:
View attachment 62512
In this calibrated graph where the range is 50 dB, I like to see nothing but blank space for distortion and the F208 gets there with the exception of the lower frequency range.
Impedance test shows dipping to 3.5 ohm so you better have a high current amplifier that doesn't mind such loads:
View attachment 62513
Horizontal directivity diagram shows very wide response which means you don't have to sit in a vice to get good sound:
View attachment 62515
And if you left the side wall reflections bare, you should get wider image. The inclusion of that mid-range gives good dispersion to whopping 100 degrees!
Vertically the situation is not as good as we typically see in non-coaxial designs:
View attachment 62516
I am in a hurry so could not optimize the CSD plot so here it is:
View attachment 62517
I wanted to see if I could diagnose the up and downs in high frequencies so ran this 3-d plot:
View attachment 62518
The problem could also be the mid-range still going as we see in the hot area below the tweeter.
Speaking of that, i had time to run the lower bit by itself:
View attachment 62519
But am out of time to calibrate the tweeter response and overlay it.
Conclusions
I expected textbook on-axis response but did not quite get that out of Revel F208. But the speaker came to its own with stellar predicted in-room response which is what ultimately matters as you hear both direct and indirect sounds. Off-axis summed response also looks superb. People ask me if they should buy Revel speakers. I always tell them they should for two reasons:
1. You can put them in just about any room and they sound great due to off-axis response being carefully designed.
2. They are the only speakers designed and released with double-blind listening tests.
I wish I had done my listening tests so I could make my own judgement with F208 but I don't have time as I have to pack and drive them to the owner in a few minutes. But based on measurements, I think they come very close to my Revel Salon 2. If so, they should sound wonderful.
I am going to put Revel F208 speaker on my recommended list. I know, shocker.
-----------
As always, questions, comments, recommendations, etc. are welcome.
Killed my back moving this speaker around so had to hire someone to get rid of all the weeds as @Thomas savage is not available. He is charging me $100/day so on top of the expenses for this site, I need money for that. So please donate as much as you can using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
It would be nice that if for low distortion speakers like this one you can extend the distortion graph down below 40db
Am I understanding things right that the summation of the on-axis and off-axis response is better than the direct on-axis response? I ask because I recently purchased a couple of GIK bass traps that are full-band. I did so because our couch is about 2 feet from the back wall (we basically have our room arranged the opposite of what it should be from an audio perspective--it is about 19.6 feet long and 11 feet wide--our TV is mounted on the wide side to give the room a "larger" feel); the idea was to put a trap mounted horizontally behind the couch to catch the direct reflections that were occurring 2 feet from behind us.
Well... So I did that, and I haven't been crazy about the result. I can post measurements, if need be. I get the impression that something like an audyssey dynamic volume function is enabled.
I currently have the traps mounted on the sidewalls to catch first reflections of the F208's. My right tower is pretty close to the wall (2 feet, 2 inches), and the left tower has a bit of room before the side wall (5 feet, 3 inches). I may actually like the sidewall mount location better than directly behind us (at the moment). I plan on doing some measurements today to see what I come up with.
So, back to my original question: given the original measurements Amir gave us in the original post, would having the sidewall bass traps catching first reflections deteriorate overall accuracy?
Well... So I did that, and I haven't been crazy about the result. I can post measurements, if need be. I get the impression that something like an audyssey dynamic volume function is enabled.
I currently have the traps mounted on the sidewalls to catch first reflections of the F208's. My right tower is pretty close to the wall (2 feet, 2 inches), and the left tower has a bit of room before the side wall (5 feet, 3 inches). I may actually like the sidewall mount location better than directly behind us (at the moment). I plan on doing some measurements today to see what I come up with.
So, back to my original question: given the original measurements Amir gave us in the original post, would having the sidewall bass traps catching first reflections deteriorate overall accuracy?
About summation of on-axis and off-axis responses what you see in CTA2034 spinorama called listening window is a good header to get understanding, on-axis is hammered out of one specific directive curve at 0º where listening window is directivity avarages of -30º to +30º horizontals and -10º to +10º verticals and includes on-axis, and in that CTA2034 specification call for those numbers in 10º steps then listening window in CTA 2034 spinorama is a avarage of 9 directivity steps and should cover a few listening seats in a normal room.
Think that you haven't been crazy happy about the result use traps or any treatment for F208 can be seen in Amir's spinorama, below is a copy of Amir's data and it can be seen this speaker is pretty perfect in anechoic domain and scores a high number of 6,08 in the preference rating race which place it as the fifth best there, positive highlights is on-axis response is close to listening window and any curve in spinorama is some fair smooth within few dB and dispersion is nice and wide, so think when F208 is situated inside a real room what will happen is below 500Hz the nice anechoic response will be ruined and its that spoiled area below 500Hz for F208 that need some analyze and correction so anechoic response is kept as good as possible, more on theoretical correction below these graphs.
In below animation have modeled F208 inside a room like yours using 19,6 feet for longest room dimension and 50% leakage to establish room pressurization gain and 3 times non coherent boundary distances to establish some room boundary gain, doing so show room gain is not a smooth thing so anechoic response below 500Hz is ruined and in that almost any musical instrument and voices have their area of where keynote happen is below 500Hz then sound performance is general spoiled, well above 500Hz where harmonics add up to instrument and voices keynote is still smooth but we need to analyze and EQ the non smooth room gain below 500Hz to get back the fantastic smooth F208 anechoic response plus the benefit that room gain adds a lower frequency system reach..
Below animation is to show the above non smooth non ideal room gain curve that need a measurement analyze to find some EQ settings using good old minimum phase filters to smooth out the good enough else room gain boost..
For above modeled example of room gain have used center between F208's two 8 inch woofers as distance to floor but in midrange takes over at 270Hz then midrange position at about 37 inches height will also mean something for frequencies below 500Hz, below animation can show that in area below 500Hz its best if wife and physical look of speaker placement for the eye can live with the compromise ensure boundary distances to the three closest boundarys relative to front of woofer position are as different as possible, the three different distances ensure EQ boost is not more than a +2dB boost where two same boundary distances take +5dB boost and three same boundary distances is a huge 11,5dB boost to repair, ignore one closest boundary distance below is called rear wall its the the one closest behind the speaker that some call front wall.
Last edited:
Amazing. Thank you so much for explaining things in a way even I could understand!
this was incredibly thorough. If I’m understanding things right, the fact that my towers are at significantly different distances from the walls actually helps things below?
Thanks, ensure position in space right on the front baffle in the middle between the two 8 inchers never have same distance to closest sidewall or frontwall will make the boundary interference suckout less deep if system is non EQ room corrected, but also if those suckouts is room corrected with EQ then headroom is much better for transducers and poweramps boosting about +2dB instead of boosting some huge +5-12dB down there, using room EQ correction for system forgot say if left verse right speaker is not situated symetrical relative to room shape then if one wants the huge stereo kick illusion sit exactly positioned in sweet spot of the triangle and enjoy dream sound then left verse right speaker needs each their analyze and unique EQ filter correction for that stereo triangle to happen and also think its important when one subjective quality check the room EQ correction that one listen one mono channel at a time as Amir do in his listening tests, because having two channels fire when one is supposed to listen technical critical for the quality of EQ filter then two channels illusions in brain hinders hear the real thing as clues for naturalness or genre of track material to be non system dependant etc, drawback listen to mono track can be track material can miss some instruments or voices that happen on track in the other channel but that phenomen can be hindered temporary route L/R channels to channel under test, a dirty quick quality test for if left verse right speaker channel perform acoustic the same in-room performance is play a mono track while sitting in sweat spot and then panning the ballance from left to right and vice versa should give same acoustic experience.
Can be wrong here but whatever reflections from F208 should be good enough quality seen in spinoramas PIR curve and sound natural if they EQ corrected for room nasties, so think if you offer the time down the road get left and right channel each their room correction below 500Hz then try that trap exercise again and see if room EQ makes a difference there.
Similar threads
- Poll
