Since you have had the BMRs in your office for two months, I would be interested in how your ears would react if the BMRs were taken out of your office and replaced with the KEF R3s.
-
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!
Mini KEF R3 vs Philharmonic BMR review
- Thread starter Kachda
- Start date
Since you have had the BMRs in your office for two months, I would be interested in how your ears would react if the BMRs were taken out of your office and replaced with the KEF R3s.
ROOSKIE
Major Contributor
Awesome! Thanks for sharing!I was one of the lucky ones who managed to be a part of the BMR road show. Over the past week, I had the pleasure of auditoning and comparing the Philharmonic BMR against my KEF R3. Both were driven by MiniDSP SHD -> March Audio P502 amps with Spotify as source. Just in case any one has any doubts, obviously I am not paid in any form for this review and I did not 'pre-clear' this with @Dennis Murphy in anyway.
@Steve Dallas has done quite an extensive review of this speaker already which you can read here -> https://www.audiosciencereview.com/...semi-objective-review-road-show-stop-1.18828/ . I will just do a brief post with my measurements and review. Word of warning - I am not an expert tester, listener or reviewer, so please take this in any way you want to.
I tried to match levels using REW SPL meter. By luck, both speakers seem to have very similar output for a given volume. Measurements shown below were done when the REW SPL meter was around 70db using Pink Noise. I can't go much louder because I live in an apartment. The room modes are generally similar, but because I had the speakers beside each other, there will be some differences due to that. Something that stands out is that the BMR has about 10db more bass below 40Hz, this can be felt in music with lots of sub-bass, though it won't be the same impact as a subwoofer.
The main difference comes around the 800-1k range. The KEF R3 is about 5db louder in this region, though I didn't feel like this impacted anything in songs. The BMR seems to stay flat and even rise a bit between 1k-2k and then drops off similar to the R3. In songs, this came through to me as some of the female voices being a bit clearer, though it was never jarring.
View attachment 116514
I carried out listening tests in mono first, with left/right channel connected to the BMR/R3. I used the minidsp to route the same signal to both speakers. I also switched them around to make sure room positioning wasn't affecting my impression. Initially, when I had the BMR on my left and the R3 on the right, the R3 seemed harsher and more piercing while the BMR was bassier and more enjoyeable on tracks like Depeche Mode - Enjoy the Silence, Daft Punk - Get Lucky. The R3 seemed to do slightly better on live music. However, this difference completely went away when I switched their positions, which to me is an indication that the speakers are actually quite similar to my untrained ears and the differences I was hearing were due to room position.
In stereo, I can't do quick A/B test, but did my best by switching speakers and listening to the same track on both right after one another. Overall, my impression was on many songs I couldn't really make out any material difference between them. However, on songs with deep bass (Cirez D - Glow, Daft Punk - Get lucky), the extra bass from the BMR definitely made the songs more enjoyeable. The foot tap-ability co-efficient was higher on the BMRs. Other than that though, I couldn't put a big difference between them. In my small apartment with lot of hard surfaces, I couldn't really make a out a big difference between the comparatively narrower directivity of the R3 vs wider directivity of the BMR (or maybe I just don't know how to listen).
Conclusion
If I was buying speakers today with a budget of <2k, and after hearing them side by side in my room, I would probably choose the BMR on pure music enjoyeability. Given that I have already invested money in my R3s though, I am not sure there is enough of a difference to justify selling the R3s at second hand price and paying full price for the BMR. It is a very tempting thought though.
I know folks are asking various things but if you are up for something very easy to test...
For me, a single measurement is not very telling in room.
Take a measurement of the speakers playing pink noise and using the MMM (moving mic method). Make a window about 2-3 feet wide 12-18 inches tall at your listening position (with the spot your head would be in in the middle of the "window") and slowly move the mice around for about 1 minute and record the averages (I like to use about 80-100 averages)using the RTA feature of REW.
Very simple and fast and so far has given me the best general idea of what the in room sound at the listening position really is.
It is not going to tell you the direct vs indirect components but I still really like this measurement.
If you do do this, I would be best to set each speaker in the same spot and remove the one not in use from the area.
There are some good instructions online but these settings in the RTA would be good choices.
After you take the measurements you could smooth them to 1/6 octave of 1/12 or whatever you/others like to see.
Also the way I match volumes is by playing pink noise custom range from 500hz-2500hz and take a SPL reading. So far this has really given a great match in perceived volume and measured volumes. This can be done is REW easily as well as it is an option in the "generator" section. REW has a 500-2000hz default but I use the custom and use the 500-2500 instead.
Last edited:
OP
- Thread Starter
- #23
Hi, the measurements shown in my original post are using pink noise with moving mic. Just that I used stereo signal to get total response rather than mono. If you want to see the individual response, the link In have to @Steve Dallas review includes all kinds of details
May I ask for your apartment listening room size? Coz I am in a literally car park spaced room with extremely close listening and I am getting a genelec 7040 for the sub to use as they fits under the desk without needing me to chop off my legsI am always tempted to get one, and would prefer one even with the BMR. However, given the size of my apt and layout, I would have to carry the sub on my head
that said for the KEF I think they have the new opposed subs with dual 6.5 inch looks great for tiny rooms? No point in tiny rooms with big 12” subs imo
- Joined
- Jan 15, 2020
- Messages
- 6,904
- Likes
- 16,936
Thank you, with level matched above 1 kHz and psychoacoustic smoothing the warmer voicing and deeper bass of the BMR, having higher level below 500 Hz, can be nicely observed:
- Joined
- Mar 17, 2020
- Messages
- 1,071
- Likes
- 4,547
It's certainly an interesting juxtaposition. Some of the variation below 1 k is real, and some I suspect is just due to slight differences in speaker positioning. For example, I'm quite sure the peak at 430 Hz in the BMR response is a room artifact that for some reason isn't getting excited by the KEF. More puzzling is the BMR's fairly wide peak between 2 kHz and 3 kHz. Every measurement that my business partner and I have taken of the BMR shows an on-axis diffraction dip centered at 2800 Hz with only a hint of a rise at 2 kHz. The family of off-axis plots shows the dip filling in, but no signs of a peak. The measurements of the BMR roadshow speaker look just like the one below. So I don't what's up with that. A peak in that area is the last thing I would want to design into a speaker.
OP
- Thread Starter
- #27
My room is L shaped, with the leg of the L being an open kitchen, and the other end open to a hallway. It is also wider than deep. Due to this, my listening distance from the speaker is only about 8-9ft. The width of the room is almost 20ft, but a part of it is used as an office and another end is used to keep a digital piano, so the useable area is about 10ftx10ftMay I ask for your apartment listening room size? Coz I am in a literally car park spaced room with extremely close listening and I am getting a genelec 7040 for the sub to use as they fits under the desk without needing me to chop off my legs
that said for the KEF I think they have the new opposed subs with dual 6.5 inch looks great for tiny rooms? No point in tiny rooms with big 12” subs imo
OP
- Thread Starter
- #28
The slight rise around 2k is also seen in @Steve Dallas measurement, I do not know enough about these things to theorize why these are different to what you have.
Attachments
if so I think you should be able to get the sub!
I am saying so because when I consult Genelec rep for my case (10ft wide X 5ft deep) rectangular room with the speakers placed along the long side, and they are not in the middle, with my listening position being only 2.5 ft away from the 8030Cs, Genelec rep told me that a sub could always make the listening experience better, of coz don't cross it too high as that will get localization problem, just add the sub and try and it should be great
Steve Dallas
Major Contributor
I may be thinking too simplistically here, but I wonder if early reflections augment that region in smaller rooms?
In any case, it is simple to EQ out if desired, therefore not problematic.
- Joined
- Mar 17, 2020
- Messages
- 1,071
- Likes
- 4,547
I think the grid markings must be off. Those peaks are between 5k and 7k. I need to get a REW setup going so I can understand better how to interpret this kind of measurement.
Hi,
The other thing to realise is that you DON’T have to have the mic away at 1m. It’s a nice convenient number that matches speaker sensitivity
eg. x dB @ 2.83V @ 1m.
You might be able to bring your microphone closer- eg. 60-80cm, because if your microphone is longer than the most significant dimension eg. Twice the width of the baffle, or further than the diagonal of a rectangular baffle; then you are effectively in the fair field already, and ok to measure.
The caveat is that a big speaker eg. 5-6 ft tall speaker with lots of drive units eg. WMTMW you have to be further away to make sure all the drive units’ acoustic output sum correctly,
Be aware that you still want the first reflection to be far away as possible; which indoors is usually limited to half the height of the ceiling.
So if your impulse response shows that your gate time is only 3ms, then your frequency resolution is the inverse. 1/0.003 = 333Hz.
So your measurement is fairly good down to 333Hz, but loses its precision completely below 333Hz.
This is how we did things before the Klippel / Klippel NFS acoustic field scanner, that has a frequency response of about 20/5Hz or less!!!
Last edited:
Also check the interface between your microphone and mic holder / stand.
Erin has show reflections between them can causes anomalies in your HF readings.
Are you able to turn off the Psychoacoustic Mode for smoothing?
It may to troubleshoot when comparing to other people’s measurement- to use the same smoothing as Dennis’.
For crossover/speaker design and troubleshooting I use a maximum of 1/24th octave.
So perhaps turn off the psychoacoustic smoothing.
Erin has show reflections between them can causes anomalies in your HF readings.
Are you able to turn off the Psychoacoustic Mode for smoothing?
It may to troubleshoot when comparing to other people’s measurement- to use the same smoothing as Dennis’.
For crossover/speaker design and troubleshooting I use a maximum of 1/24th octave.
So perhaps turn off the psychoacoustic smoothing.
OP
- Thread Starter
- #35
Hi Dennis, I just cut off the response below 1k as they are more room influenced. You can see that the x-axis is from 1k - 20k. There is also a rise between 5k-7k followed by a dip.
- Joined
- Mar 17, 2020
- Messages
- 1,071
- Likes
- 4,547
Sorry--the rise at 2 K was so small my eyes just went to the one higher up. I have no idea what's going on there, other than I'm sure it's not the inherent response of the speaker (unless something got jolted during the BMR's travels).
Yep, I've measured some smaller speakers at 2 feet with no issues. Often when doing preliminary measurements I'll just measure from extra close to give myself a larger window or because I don't want to move stuff out of the way.
Measuring from 2 feet away with 4 feet of distance from walls will give you a 5.5ms gate. I did all the Google Nest Audio measurements at 2 feet, for example.
Granted, measuring from so close can be a bit of a problem for directivity measurements with larger bookshelves, especially vertical measurements. But for a single on-axis curve or even a listening window it is very rarely an issue. As mentioned, 2 times the baffle width is usually fine.
- Joined
- Mar 17, 2020
- Messages
- 1,071
- Likes
- 4,547
That's not to say that the closer plots will be replicas of the one-meter. The phase relationships between the drivers will definitely change as you get closer, and that will affect what you see. I do agree that at times the greater resolution is worth it. Of course, there's nothing inherently right about one meter. I design for a listening distance of 2.5 meters for small monitors and 3 meters to 3.5 meters for larger speakers. That will definitely affect the frequency response, so even a one-meter measurement won't be quite what you're hearing at the design reference point, even ignoring room effects.
Oh for sure. I actually measure all speakers at a few angles at 2m as well(except the couple of times I've forgot), which gives me about a 5ms gate. I just haven't seen a bookshelf speaker yet that measures noticeably different at 1m from 2m within the resolution of the gate, so the latter is what I use. But I always check just to be sure.
Steve Dallas
Major Contributor
Your wish is my command:
KEF R3 vs. Philharmonic BMR Grudge Rematch to the Probably not Death Thread
As none of you recall, I tested the roadshow BMRs that Dennis was kind enough to ship all around the country last year and posted a lengthy thread about the experience. And, I ended up liking them enough to buy a pair. They even unseated my KEF R3s, which I sold shortly after receiving the...
www.audiosciencereview.com
Similar threads
