• 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!

Philharmonic BMR Monitor Semi-Objective Review - Road Show Stop 1

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
I would definitely agree for a narrow or constant directivity speaker; that would be far too bright. This speaker though is very wide and will show much less of a slope in-room due to all the reflections. It's hard to say without listening.

Given that we know this is a neutral speaker, I think it's best to just let the natural roll off happen. The speaker naturally rolls of ~6dB from 6-18kHz?(sitting far away, so kinda hard to see exactly). The Dirac curve only falls ~1dB? or so, so unless I'm misunderstanding something, Dirac is turning the BMR into a speaker that's 5dB hot at 18kHz.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
I just scrolled up and I see the response above 10 kHz dropping. I'm surprised with the ribbon and all. His room must be bigger than mine, because here is a popular waveguided dome tweeter in room measuring less than 17x11 ft:

Lol, the name of the file gave it away :p. Great looking response, though.
 
OP
Steve Dallas

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
I just scrolled up and I see the response above 10 kHz dropping. I'm surprised with the ribbon and all. His room must be bigger than mine, because here is a popular waveguided dome tweeter in room measuring less than 17x11 ft:

Both of my rooms are near cubes. Absolutely terrible.

The charts in this post show the speakers pointed straight ahead and pointed straight at the MLP. The slope of the roll-off is nearly identical in both measurements. This seems to indicate it is not room-dependent. It is also not a problem, as it is close to the target curve, and there is not much content up there anyway.
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,756
Likes
16,218
Dirac does help with those unfortunate nulls.
Would be interesting to measure and plot both with and without Dirac at the same amp level, as in my experience Dirac and almost all DRC products just pull the total level down up to 6-7 dB so the dips aren't as deep.
 
OP
Steve Dallas

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
Would be interesting to measure and plot both with and without Dirac at the same amp level, as in my experience Dirac and almost all DRC products just pull the total level down up to 6-7 dB so the dips aren't as deep.

There is definitely a drop in level. I am not sure it is as simple as pulling everything down to the lowest null, however. I believe it tries to avoid that by adjusting phase first. It will also boost up to 10dB in some areas, which is usually only effective after adjusting phase. Or, at least that is what messing around with RePhase and REW have taught me.

This is getting way off topic, but I believe you are correct, as it applies to AcourateDRC. The instructions have you pull down the target to 1.5 or 2dB above the lowest null, then it does its calculations. This is the Revels as measured and with the target in place. The correction is calculated from this:

Measurement.png
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
Would be interesting to measure and plot both with and without Dirac at the same amp level, as in my experience Dirac and almost all DRC products just pull the total level down up to 6-7 dB so the dips aren't as deep.

On my Dirac system, I do have to turn it up a bit once I turn it on. I don't think it's 6-7dB, but it's definitely quieter.
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,756
Likes
16,218
There is definitely a drop in level. I am not sure it is as simple as pulling everything down to the lowest null, however. I believe it tries to avoid that by adjusting phase first.
In my room its phase correction didn't really manage to do fill any null like for example one that happens at the summation L+R at a specific bass frequency due to the different room phase response of them except mainly linearising the phase above but also other DRC products failed there.

On my Dirac system, I do have to turn it up a bit once I turn it on. I don't think it's 6-7dB, but it's definitely quieter.
Yes, that depends also on how deep the dip is and how much it fills it.

As said though some REW measurements with Dirac on and off are always very interesting, as maybe it works more effectively in your rooms.
 
OP
Steve Dallas

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
Worked with Dirac a little more and successfully created a target that closely follows my usual Harman-like target. Getting closer, but it still sounds... warm.

Correction limited to under 1K and under 10K. My preference is still for under 1K.

This:
Dirac Target to 1KHz.png


Results in this (stereo):
BMR Dirac Harman Target 1K vs 10K 16.png


Up to 1K correction overlaid with target (stereo):
BMR Dirac 1KHz vs Harman Target.png


Up to 10K correction overlaid with target (stereo):
BMR Dirac 10KHz vs Harman Target.png


I'll work on that target curve more when I have time...
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
In my room its phase correction didn't really manage to do fill any null like for example one that happens at the summation L+R at a specific bass frequency due to the different room phase response of them except mainly linearising the phase above but also other DRC products failed there.

I haven't evaluated the effectiveness of many of these software correction tools... but how well automated room correction programs work down in the bass is incredibly dependent in the room and placement situation.

Phase correction to "fill" nulls is more like navigating around (rather than through) nulls/nodes by manipulating the time.

1610669189181.gif
 
OP
Steve Dallas

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country

alexis

Member
Audio Company
Joined
Jun 3, 2020
Messages
97
Likes
159
He says he took some measurements. Hopefully he posts them here. I am not on Reddit and don't care to be, so I can't ask about it.

If you scroll through pictures posted on Reddit, the last one has a Frequency responses diagram for both.
 
OP
Steve Dallas

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
If you scroll through pictures posted on Reddit, the last one has a Frequency responses diagram for both.

Right, but only to 500Hz. We see his room modes excite in the same manner, and the BMR has more much more bass energy, which are things we already know. He talks a lot about the midrange. I'd like to see some measurements up there. [shrug]
 

Bobulator

New Member
Joined
Jan 26, 2021
Messages
4
Likes
4
BMR Roadshow Review - My subjective listening comments and observations on the Original BMR compared to new BMR v2. (Crosspost from AVSforum)

I had the good luck to still have on loan an original pair of BMRs, in fact the specific pair that Erin of Erin’s Audio Corner reviewed a few months ago, to be able to compare with the new Roadshow BMRs. I don’t have a lot of test equipment or detailed knowledge, so this will primarily be a subjective review. That being said, I do have an Omnimic rig, so I did do a FR test of both, and the comparison graph and more info on it is further down below.

First my listening impressions: As you can imagine, they sound very similar, sharing the same BMR mid and tweeter. The biggest difference is in the bass of the new SB woofer. It sounds fuller, richer, with more low and mid bass bloom than the original Scan Speak (SS) unit. I really like this new driver, as I felt there was a bit of warmth missing in the original system. However both of these systems have exceptional clarity, imaging, and soundstage definition for well recorded material. Tremendous air, delicacy, [Insert your favorite audiophile adjectives as desired]. Simply amazing!

More of a concern to me comes with listening to less well mixed recordings, like most all pop/classic rock music from the 60s up till maybe 10 or 20 years ago. I grew up in this era, and enjoy listening to this music, unfortunately the essentially ruler flat frequency response of these high resolution monitors makes these often poor mixes hard to listen to. I wouldn’t go as far to say ear bleeding like some speakers I’ve heard, because the phenomenally low distortion drive units and xover integration keep the sound as clean as possible, it is purely the recordings themselves to blame. If you don’t listen to this type stuff, but focus on more modern pop, or pretty much any well recorded jazz or classical, this will not likely be an issue for you. This is why my personal preference is for a bit of the controversial BBC dip in system response, a couple db dip in response generally in the 1-4 kHz area.

On a side note, I was very fortunate to purchase a pair of Avalon Ascent MKIIs recently from a local audio club member who was downsizing, and my first comparison was between one of the new BMRs and one of them. (my preamp has an old school balance control to switch between speakers) The difference was striking, the Avalons almost sound like they have a towel draped over them! However, I can listen to them all day on any type of music, both modern well recorded, and classic rock poorly mixed without fatigue. To be fair, you have to add another 0 to the price of these compared to the BMRs! I’ve read Avalon doesn’t like reviewers to test FR, and I suspect I now know why, probably some of that BBC dip dialed in which would look bad on a graph. In fact, looking at a FR I ran on them and indeed there is a broad dip in the 1.5 to 4k Hz region. They also roll off a bit faster past 15k. Yet somehow they have similar amazing imaging, with a deeper soundstage and even somehow seems to image past the speakers.

Back on topic. Dennis said the new units are somewhat more efficient, I heard this mainly in the bass as mentioned. To my ears it almost sounds like the higher efficiency SB, crossed over a bit higher than the SS is the main difference in the slight additional volume of this unit when A/B ing them. I don’t think it is enough to substantially change the amp requirements for adequately powering either BMR version.

Below is my Omnimic graph comparing the 2 speakers. For this test, I had my mike about 1m from the center of the drivers, and put each speaker on the same 21” stand with the front baffle at exactly the same position. The mike was about 33” from the floor. The test was run in my family/media room, roughly 22’ square with cathedral ceiling, with 3x large door cutouts and a stairway down, all walls are tongue and groove pine. Large rug on the floor, and normal furniture. Equipment used for the test and listening is an Oppo 103, fed into an Adcom GFP-565 preamp (run thru bypass outputs, skipping tone controls, etc.) into a Classe DR-25 amp. Used sine sweep track 2.

1611627927525-png.3083464



The blue line is the original BMR, the red line is the new version. You can ignore the absolute curves, due to room interactions, the main takeaway is the difference in the curves. You can see the couple dB increase for the new BMR in the FR up to 100 Hz, this is audibly noticeable when listening to music. They track pretty closely 100-200 Hz, then the new BMR pulls up again and stays a bit smoother till about 1.5 kHz. Pretty much ruler flat above that for both. Not a huge difference in their curves, but the slightly higher bass output the new BMRs makes them more pleasant to my ears.

As mentioned my ears prefer a gentle slope down or a BBC dip (I believe this is due to the Fletcher-Munson curves, google for more info), but perhaps then some of the laser like clarity would be sacrificed. In the old days, some manufacturers had “presence” controls or switches that would allow users to pad down this region, but these days perhaps DSP is the tool to play with to tailor the sound to users preference in their listening room.

Perhaps another concern of some is the loudness capability of these systems with relatively modest 6-7” woofers. Diameter of the woofers from the center of the surrounds are 125mm for the SB vs 140mm for the SS. So I suspect the SS may be able to get louder eventually than the SB given enough power, but my ears would be bleeding from the presence region by then. I could listen comfortably at a pretty loud 85-90 db (per the Omnimic meter), with virtually no compression or other deleterious effects. At higher levels, there is a bit of chestiness with the SB woofer I didn’t hear with the SS unit, but I was getting seriously loud. (wife was out of the house) There also was some port chuffing, not unexpected at these high levels, really only noticed on the last track of Dennis’ test CD - a 34 Hz tone.

Some notes on the cabinets: the new BMR is gorgeous! Super smooth gloss finish over a wine colored veneer. The finish is so thick (looking at the corners it looks like almost 1/16” thick) it has a milky color, I wish it was more clear. Due to the mirror finish, you can hardly see the veneer beneath, a previous reviewer’s spouse said it looked like the speaker could have been made out of plastic, I concur. The older BMR had the curved cabinet, it seemed a bit more solid with the knuckle rap test. I prefer its cherry veneer over the newer wine colored veneer, but that’s personal preference. For me a brown stained walnut veneer, with the beautiful new BMR finish (but clear, unmilky color) would be killer!

Summarizing, in my opinion the new BMR is slightly more efficient, with better, warmer bass, giving a more balanced sound, but still has the laser sharp imaging of the original BMR. It’s still a bit bright to me on poorly mixed recordings like classic rock, but overall a simply amazing design, with updates making it even better.
 

Dennis Murphy

Major Contributor
Technical Expert
Joined
Mar 17, 2020
Messages
1,071
Likes
4,535
BMR Roadshow Review - My subjective listening comments and observations on the Original BMR compared to new BMR v2. (Crosspost from AVSforum)

I had the good luck to still have on loan an original pair of BMRs, in fact the specific pair that Erin of Erin’s Audio Corner reviewed a few months ago, to be able to compare with the new Roadshow BMRs. I don’t have a lot of test equipment or detailed knowledge, so this will primarily be a subjective review. That being said, I do have an Omnimic rig, so I did do a FR test of both, and the comparison graph and more info on it is further down below.

First my listening impressions: As you can imagine, they sound very similar, sharing the same BMR mid and tweeter. The biggest difference is in the bass of the new SB woofer. It sounds fuller, richer, with more low and mid bass bloom than the original Scan Speak (SS) unit. I really like this new driver, as I felt there was a bit of warmth missing in the original system. However both of these systems have exceptional clarity, imaging, and soundstage definition for well recorded material. Tremendous air, delicacy, [Insert your favorite audiophile adjectives as desired]. Simply amazing!

More of a concern to me comes with listening to less well mixed recordings, like most all pop/classic rock music from the 60s up till maybe 10 or 20 years ago. I grew up in this era, and enjoy listening to this music, unfortunately the essentially ruler flat frequency response of these high resolution monitors makes these often poor mixes hard to listen to. I wouldn’t go as far to say ear bleeding like some speakers I’ve heard, because the phenomenally low distortion drive units and xover integration keep the sound as clean as possible, it is purely the recordings themselves to blame. If you don’t listen to this type stuff, but focus on more modern pop, or pretty much any well recorded jazz or classical, this will not likely be an issue for you. This is why my personal preference is for a bit of the controversial BBC dip in system response, a couple db dip in response generally in the 1-4 kHz area.

On a side note, I was very fortunate to purchase a pair of Avalon Ascent MKIIs recently from a local audio club member who was downsizing, and my first comparison was between one of the new BMRs and one of them. (my preamp has an old school balance control to switch between speakers) The difference was striking, the Avalons almost sound like they have a towel draped over them! However, I can listen to them all day on any type of music, both modern well recorded, and classic rock poorly mixed without fatigue. To be fair, you have to add another 0 to the price of these compared to the BMRs! I’ve read Avalon doesn’t like reviewers to test FR, and I suspect I now know why, probably some of that BBC dip dialed in which would look bad on a graph. In fact, looking at a FR I ran on them and indeed there is a broad dip in the 1.5 to 4k Hz region. They also roll off a bit faster past 15k. Yet somehow they have similar amazing imaging, with a deeper soundstage and even somehow seems to image past the speakers.

Back on topic. Dennis said the new units are somewhat more efficient, I heard this mainly in the bass as mentioned. To my ears it almost sounds like the higher efficiency SB, crossed over a bit higher than the SS is the main difference in the slight additional volume of this unit when A/B ing them. I don’t think it is enough to substantially change the amp requirements for adequately powering either BMR version.

Below is my Omnimic graph comparing the 2 speakers. For this test, I had my mike about 1m from the center of the drivers, and put each speaker on the same 21” stand with the front baffle at exactly the same position. The mike was about 33” from the floor. The test was run in my family/media room, roughly 22’ square with cathedral ceiling, with 3x large door cutouts and a stairway down, all walls are tongue and groove pine. Large rug on the floor, and normal furniture. Equipment used for the test and listening is an Oppo 103, fed into an Adcom GFP-565 preamp (run thru bypass outputs, skipping tone controls, etc.) into a Classe DR-25 amp. Used sine sweep track 2.

1611627927525-png.3083464



The blue line is the original BMR, the red line is the new version. You can ignore the absolute curves, due to room interactions, the main takeaway is the difference in the curves. You can see the couple dB increase for the new BMR in the FR up to 100 Hz, this is audibly noticeable when listening to music. They track pretty closely 100-200 Hz, then the new BMR pulls up again and stays a bit smoother till about 1.5 kHz. Pretty much ruler flat above that for both. Not a huge difference in their curves, but the slightly higher bass output the new BMRs makes them more pleasant to my ears.

As mentioned my ears prefer a gentle slope down or a BBC dip (I believe this is due to the Fletcher-Munson curves, google for more info), but perhaps then some of the laser like clarity would be sacrificed. In the old days, some manufacturers had “presence” controls or switches that would allow users to pad down this region, but these days perhaps DSP is the tool to play with to tailor the sound to users preference in their listening room.

Perhaps another concern of some is the loudness capability of these systems with relatively modest 6-7” woofers. Diameter of the woofers from the center of the surrounds are 125mm for the SB vs 140mm for the SS. So I suspect the SS may be able to get louder eventually than the SB given enough power, but my ears would be bleeding from the presence region by then. I could listen comfortably at a pretty loud 85-90 db (per the Omnimic meter), with virtually no compression or other deleterious effects. At higher levels, there is a bit of chestiness with the SB woofer I didn’t hear with the SS unit, but I was getting seriously loud. (wife was out of the house) There also was some port chuffing, not unexpected at these high levels, really only noticed on the last track of Dennis’ test CD - a 34 Hz tone.

Some notes on the cabinets: the new BMR is gorgeous! Super smooth gloss finish over a wine colored veneer. The finish is so thick (looking at the corners it looks like almost 1/16” thick) it has a milky color, I wish it was more clear. Due to the mirror finish, you can hardly see the veneer beneath, a previous reviewer’s spouse said it looked like the speaker could have been made out of plastic, I concur. The older BMR had the curved cabinet, it seemed a bit more solid with the knuckle rap test. I prefer its cherry veneer over the newer wine colored veneer, but that’s personal preference. For me a brown stained walnut veneer, with the beautiful new BMR finish (but clear, unmilky color) would be killer!

Summarizing, in my opinion the new BMR is slightly more efficient, with better, warmer bass, giving a more balanced sound, but still has the laser sharp imaging of the original BMR. It’s still a bit bright to me on poorly mixed recordings like classic rock, but overall a simply amazing design, with updates making it even better.

Thanks for the write-up Bobulator. I'm going to try and make a point of not commenting on or taking issue with subjective impressions of the BMR's in these reports from participants in the road show. I don't want to discourage anyone from saying exactly what they think. However, I did want to make a couple of points about the posted measurement, since it's so hard to keep straight exactly what methodology is being employed on the various measurements we see on this site. The 'lator's plot was made using the OmniMic suite of hardware and software, which has some peculiarities. First, the plot is made using the "blend" feature of OmiMic, which gradually shifts the gating from anechoic to room mode beginning around 800 Hz and below. (That's too high a transition point in my view, and the same designer's Praxis software, which I use, doesn't transition until around 300 Hz). So all of the bumps and grinds below about 800 Hz are modes in the 'lator's particular room.

Second, the default smoothing on OmniMic is 1/6th Octave, which is very course. In this case, it obscures what is actually a small "BBC" dip around 2800 Hz, of the type the comments suggest might be of benefit. You can see it in the OmniMic Plot of the Road Show BMR's shown in an earlier post (reproduced below) by my business partner, which was made with much less smoothing. Finally, and I mention it only because it ticks me off that Parts Express refuses to acknowledge the problem, the OmniMic calibration files aren't very accurate and usually result in a high frequency roll-off above 12 kHz that doesn't really exist. The calibration file used in the plot below is pretty good. The Bobulatort's--not so much. Anyhow, thanks again for taking the time to post. There's one more dude in Atlanta that wants to hear the BMR's, and then I'll send them bouncing up I 95 to North Carolina.


1611645583022.png
 

Bobulator

New Member
Joined
Jan 26, 2021
Messages
4
Likes
4
Dennis, Thanks for the thoughtful comments. Didn't know the Omnimic cal file doesn't properly calibrate above 12k, seems like something they could fix? Particularly as so many people seem to use it. For me however, with my... ahem... older ears I don't hear much in the top octave so it's not too much of a concern. I may be missing the last bit of joy from the RAALs, but they still sound pretty special to my ears! Just after I posted I realized I didn't mention I used 1/6 octave smoothing, sorry. I tried other more resolving settings, 1/12 and 1/24, but didn't seem to give additional comparison data, seemed to be mostly just more room noise, so I left it at 1/6. I did mention to ignore the absolute values of the curves in the lower frequencies due to room modes, just to view the curve differences. Honestly, I was surprised in the small differences in the sub 100Hz area, my objective listening experience would make me believe there was a bigger difference with the new BMR. More surprising is the difference in the BBC area dip are around 2.8 kHz you show, I didn't see any evidence of that in my measurements. Any thoughts why that might be? I'd expect the Omnimic cal to be OK in that area, or possibly no?
 

Dennis Murphy

Major Contributor
Technical Expert
Joined
Mar 17, 2020
Messages
1,071
Likes
4,535
Dennis, Thanks for the thoughtful comments. Didn't know the Omnimic cal file doesn't properly calibrate above 12k, seems like something they could fix? Particularly as so many people seem to use it. For me however, with my... ahem... older ears I don't hear much in the top octave so it's not too much of a concern. I may be missing the last bit of joy from the RAALs, but they still sound pretty special to my ears! Just after I posted I realized I didn't mention I used 1/6 octave smoothing, sorry. I tried other more resolving settings, 1/12 and 1/24, but didn't seem to give additional comparison data, seemed to be mostly just more room noise, so I left it at 1/6. I did mention to ignore the absolute values of the curves in the lower frequencies due to room modes, just to view the curve differences. Honestly, I was surprised in the small differences in the sub 100Hz area, my objective listening experience would make me believe there was a bigger difference with the new BMR. More surprising is the difference in the BBC area dip are around 2.8 kHz you show, I didn't see any evidence of that in my measurements. Any thoughts why that might be? I'd expect the Omnimic cal to be OK in that area, or possibly no?

I'm not sure why you didn't pick up on the little dip in your other measurements. It's definitely there on the Road Show pair and every other BMR. It's not deliberate. The BMR midrange driver has an on'axis diffraction dip there that's virtually impossible to eliminate with passive components. it goes away off axis and probably isn't very audible. You can see it in greater detail in Ken Lin's Praxis measurement below. That also shows the inherent response of the BMR in the 300 - 800 Hz range because the measurement is still anechoic in that region. Below 300 Hz the room response falls off rapidly because Ken did the measurement in the middle of his huge downstairs rec room space.

1611690443224.png
 

Bobulator

New Member
Joined
Jan 26, 2021
Messages
4
Likes
4
Dennis, at 1/12 octave I did get a small dip around 2.8 kHz, but not enough to form any significant BBC type dip. Also got a bump from about 1.8 to 2.2 kHz. Somewhat matches the above graph, but not sure if these little bumps are significant, as they are from a single point in space, and weren't apparent in listening.

1611719234328.png
 

Selah Audio

Active Member
Audio Company
Forum Donor
Joined
Feb 4, 2020
Messages
194
Likes
445
Dennis, at 1/12 octave I did get a small dip around 2.8 kHz, but not enough to form any significant BBC type dip. Also got a bump from about 1.8 to 2.2 kHz. Somewhat matches the above graph, but not sure if these little bumps are significant, as they are from a single point in space, and weren't apparent in listening.

View attachment 108663
I use the Omnimic and have a business associate that has one as well. Neither system has a problem above 12K. For your graph can you share what settings were used?
 

Dennis Murphy

Major Contributor
Technical Expert
Joined
Mar 17, 2020
Messages
1,071
Likes
4,535
I use the Omnimic and have a business associate that has one as well. Neither system has a problem above 12K. For your graph can you share what settings were used?
This isn't a question of settings. You load the supplied calibration file and you go with what you get. I've had 4 different Omnimics. All had the problem, although it's quite manageable in my current setup. You can see the issue in the Bobulator's plots. It's a very well-known problem that probably doesn't occur all the time. You and your friend lucked out, as did I after 4 tries.
Dennis, at 1/12 octave I did get a small dip around 2.8 kHz, but not enough to form any significant BBC type dip. Also got a bump from about 1.8 to 2.2 kHz. Somewhat matches the above graph, but not sure if these little bumps are significant, as they are from a single point in space, and weren't apparent in listening.

View attachment 108663
That's a very strange plot. I can't think of any reason why you would get that scooped out response between 200 and 1200 Hz. That doesn't look like a normal room response. Is there anything weird about the dimensions of your room?
 
Top Bottom