• 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

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,424
Location
NYC
The one measure I don't see here is the one NRC of Canada does called "deviation from linearity." It measures the difference of FR in db when the device under test is playing at 70 and 90 db respectively. I think this test is a useful proxy for the dynamic range of the speaker. It could also be expanded to see what the deviation from linearity is when the speaker is playing at, say 80 and 100 db.

The speakers almost always show no change
I posted the F228Be spinorama few posts back, it's a Klippel measurement by Harman. Here it is again.
I think the 228Be should not be so different in bass extension so maybe it would be interesting to overlay it to the F208 ASR spin.

View attachment 62635

Do we know Harman is using the NFS? It's possible they are simply using Klippel measurement equipment, but not necessarily the NFS?
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
I have both the Salon2 and F208. I prefer the F208S. They are less clinical sounding and perhaps let more of the emotion of music through.

Interesting. I don't doubt it, as these spins are near perfect for a passive speaker. Still, I would imagine that the Salon 2s can play quite a bit louder, which may be very important, depending on the listener.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700

vavan

Senior Member
Joined
Apr 29, 2019
Messages
341
Likes
212
Location
Kazan, Russia
I was not using calibrated UMIK-1 but another mic that had problems in the HF range, hence the roll-off after 10kHz
I also have in room hf roll-off with f208s and umik
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
I find that speaking of subs is confusing in the context. If we talk about music (not movies) there is very little content at 20Hz, maybe the lowest organ note but not much. The F208 is a full range speaker so it should be able to play full range music without a sub, as a consequence I'd leave subs out of the equation. The R3 is a bookshelf and is not meant to be a full range speaker so bit of surprise here.

Also we are discussing this matter only because there is a huge difference between Harman measurements and ASR's (in bass response), I am not really sold on the fact that "Harman measurements are wrong" and also specs are wrong so would like to better understand. In the end we spend 50% of the time on this forum praising Harman/Revel stuff, the only that do blind tests, the only that publish official spins, the only baked by research and now all in a sudden they are just wrong?

IME, even if a speaker can play full range, it still needs to be crossed over to subs to get an even response around the room. I suppose you could fix much of that with EQ, but that eats into headroom. Best to just cross over to multiple subs and EQ that. I think @QMuse posted a really pretty bass response with no subs, but my experience is that examples like that are rare.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
What went wrong with the CSD measurement?
As I said, I did not have time to investigate. That graph is room compensated and I don't recall if that created the peak there or it was there before that. Will check later.
 

BYRTT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 2, 2018
Messages
956
Likes
2,452
Location
Denmark (Jutland)
This one is bit funny in that Revel list XO point for woofer to 270Hz and Amir analyze curve as below which is more similar a 600Hz 8th order roll off or something :p there about.

8.png
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
Thanks for posting that. As he shows, the packing is very well thought out. Without it, it would have been impossible for me to unpack and pack it by myself. Other speakers put zero effort in packaging and protecting the drivers. They can be nightmare to open and repack.

I did ask my wife to help with the second one as I was in a hurry to return them to the owner. She officially quit after helping me saying I need to hire help!
 

Gatordaddy

Active Member
Forum Donor
Joined
Apr 1, 2020
Messages
110
Likes
186
IME, even if a speaker can play full range, it still needs to be crossed over to subs to get an even response around the room. I suppose you could fix much of that with EQ, but that eats into head room. Best to just cross over to multiple subs and EQ that. I think @QMuse posted a really pretty bass response with no subs, but my experience is that examples like that are rare.

IMO, subs appear to be the more cost effective (and purportedly higher fidelity) solution to get target-response bass down to 20 hz, especially at this price point. These look very well designed to get either get high output into large rooms, or to provide adequate bass and output into less-large rooms without subs unless you are a HT nut.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
This one is bit funny in that Revel list XO point for woofer to 270Hz and Amir analyze curve as below which is more similar a 600Hz 8th order roll off or something :p there about.
Ah, that is my fault then as the room compensation will compensate for any droop below the stated frequency. And I forgot to update that field which I think defaulted to 700 Hz from my last measurement.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
@amirm: I noticed that the spin dataset you published is, somewhat surprisingly, 10 pts/octave instead of the usual 20:
Yes, that is on purpose. Meant to stated that but with being in a hurry to leave and return these speakers, I forgot. :)

The default in Klippel NFS is 10 points/octave. Long time ago I updated that to 20 points/octave to match the research from Sean Olive on speaker preference. For this speaker I had to use high order expansion to get the bass response to be correct. That in turn causes some errors that show up as many jagged peaks in the frequency response and especially power DI. I did not have time to tune this parameter enough to balance the two needs so took the quick shortcut of reducing the resolution to basically soften those jaggies.

Klippel is working on a solution for this so it should be less work in the future.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
Could the lack of bass as compared to the official graphs be something as simple as the boundary and normal inputs on the back rotary knob being mixed up?
No. I took the picture of the terminals right after the measurements were performed. As I noted in the review, I played with boundary switch and took one sweep. It just lowered the level some (as it should).
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
F228Be spinorama, could you overlay it to the F208 spin as measured by amirm?

View attachment 62614

I happen to have Klippel NFS generated version of F228be as well, performed by someone else which I got as part of my due diligence before purchasing the Klippel NFS. These measurements show the same steeper roll off I am seeing in F208. In addition, I have two anechoic measurements of the same, neither one of which show that bass response published by Harman above.

Since the anechoic rooms have room modes that impact both the microphone and speaker, perhaps Harman attempts to optimize that for best bass performance.

I think at this point we need more information from Harman on conditions of their testing to resolve this difference. I have spent an eternity trying to confirm my work. Klippel has verified that what I am seeing is correct as well (for Revel F35).
 

Robbo99999

Master Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
6,874
Likes
6,672
Location
UK
IME, even if a speaker can play full range, it still needs to be crossed over to subs to get an even response around the room. I suppose you could fix much of that with EQ, but that eats into headroom. Best to just cross over to multiple subs and EQ that. I think @QMuse posted a really pretty bass response with no subs, but my experience is that examples like that are rare.
I don't think it was that pretty, I think 1/6th smoothing was used judging by the look of the whole curve so that would have hidden a lot of peaks & troughs in the bass response....but it did extend further into the bass than the spinorama suggested.
For example this is a one sixth smoothing plot for my JBL 308p Mkii (well it's 1/6th on the bass and more relaxed on the treble), and you can tell it's the same kind of look to the bass in terms of smoothing as QMuse's measurement. I know that if I use Var Smoothing on the whole plot that the bass will show a lot more fluctuations, definitely 1/6th smoothing being used on that, so it's not as beautiful as it looks.
miniDSP Tquarter Harman Bass and Half Harman treble 250-20000 speaker delay.jpg
 

BYRTT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 2, 2018
Messages
956
Likes
2,452
Location
Denmark (Jutland)
Here's my first try measuring. I'm a bit of a REW noob though, so take it with a grain of salt.

I watched this video as a reference for setting up the mic and REW. Would appreciate any pointers.....

Thanks here overlaid to Qmuse's measured in room curve from that dealer shop plus Amir's anechoic on axis and PIR (predicted in room), looks you have tough low end reach from boundary's that tight corner and imagine you still grinning ear to ear listening to them very much beautifull speakers ...:)

About pointers for in room measurements there is threads here at ASR and over in review thread for Kali Audio IN-8 there where some video links to MMM, else think via PM ask @QMuse or @thewas_ for their good guidance.


Airs_F208_PIR_plot.png
 

Robbo99999

Master Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
6,874
Likes
6,672
Location
UK
Here's my first try measuring. I'm a bit of a REW noob though, so take it with a grain of salt.

I watched this video as a reference for setting up the mic and REW. Would appreciate any pointers.

View attachment 62701
Well, my first port of call would be to choose a Harman Curve as target, and overlay that on your plot there. By the looks of it you've got a wide trough from 230Hz down to 130Hz, which is similar to my JBL 308p Mkii's now that I have them in a new position, I have the same dip as you. I'd use 2 Low Shelf Filters to create a broad rise in the frequency in that gap between those 2 frequencies....so I'd do something like 230Hz Low Shelf Q1 +5dB...then I'd do another at 130Hz Low Shelf Q1 -5dB(minus) to bring it back down. That would pump up that region, then I'd run measurements again and then let REW iron out the peaks. I did something similar for my JBL's except I only went with -3dB on the bottom Low Shelf as I wanted to retain a 2dB Low Shelf for all of the low frequencies sub 100Hz to give the speaker more bass extension, but you don't need that.

If you can't really play with speaker positioning then go ahead and EQ them, otherwise find best mixture of speaker positioning combined with how willing you are to compromise the room layout that gives you the best measurements...and then EQ it from that point.

In fact find the ideal toe in, listening position distance, and distance between speakers for maximum imaging, and then EQ it from that point...that's in addition & part of the broader coarser speaker positioning I talked about in previous paragraph.
 
Last edited:
Top Bottom