• 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!

KEF LS50 Bookshelf Speaker Review

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
This review made me register and I'm skipping formal introduction. I doubt anyone's interested, but if you are, I'll do it later.

I have a question which will ask of you to speculate, so if you have a wooden pencil lying around, please bite on it. ;-)
To ask this question I can't hold nothing back so I'll say I'm extremely satisfied with these speakers and most of these comments how it's just another unremarkable speaker don't bother me. I'm holding onto mine until death or lottery.

But what I really can't comprehend is this lack of bass you all talk about. (???????) When I was shopping for bookshelves no other had lows nowhere near what these spouted (Revel included, perhaps it was a poor set up in the shop). First of all these speakers venture into tactile bass. Me sitting 9 feet from them and vibrations enter through couch (then my ass, sorry to say) and shake my guts at 80dB measured.

All my guests drop their jaw when they hear the lows and most of them state it's too much and want them lowered. (No DSP, Pure Direct on an Yamaha ax-890 110Wpch into 8). I have them foot and a half from the back wall and 5 feet from side walls.

And now for the walls (as Oclee stated himself about the difference in US and EU houses and in room response). I have these speakers in a 22 square meter room (240 square feet) in front of a very hard, bare concrete wall. It is 5,5 meters long (16-17 feet). Simple formula speaks of a bump in 60Hz region.

But is this enough to explain so much bass? I'm sharing a (not so precise) spectrogram of these speakers playing a song that goes bellow 30Hz. You can see the activity in the lower region, but on the bottom half, take a look at four red stripes (upper left corner of the lower half of the image) extending well beyond 50Hz. That is the EDM heavy and low beat that is played in this song.

One thing that makes my visitors laugh is when I say I plan to get a sub.

Could you give your opinions and assumptions on this huge difference you see in these measurements and the output I'm getting at my place, please?
Screenshot_20200614-135106.png
 

whazzup

Addicted to Fun and Learning
Joined
Feb 19, 2020
Messages
575
Likes
486
This review made me register and I'm skipping formal introduction. I doubt anyone's interested, but if you are, I'll do it later.

I have a question which will ask of you to speculate, so if you have a wooden pencil lying around, please bite on it. ;-)
To ask this question I can't hold nothing back so I'll say I'm extremely satisfied with these speakers and most of these comments how it's just another unremarkable speaker don't bother me. I'm holding onto mine until death or lottery.

But what I really can't comprehend is this lack of bass you all talk about. (???????) When I was shopping for bookshelves no other had lows nowhere near what these spouted (Revel included, perhaps it was a poor set up in the shop). First of all these speakers venture into tactile bass. Me sitting 9 feet from them and vibrations enter through couch (then my ass, sorry to say) and shake my guts at 80dB measured.

All my guests drop their jaw when they hear the lows and most of them state it's too much and want them lowered. (No DSP, Pure Direct on an Yamaha ax-890 110Wpch into 8). I have them foot and a half from the back wall and 5 feet from side walls.

And now for the walls (as Oclee stated himself about the difference in US and EU houses and in room response). I have these speakers in a 22 square meter room (240 square feet) in front of a very hard, bare concrete wall. It is 5,5 meters long (16-17 feet). Simple formula speaks of a bump in 60Hz region.

But is this enough to explain so much bass? I'm sharing a (not so precise) spectrogram of these speakers playing a song that goes bellow 30Hz. You can see the activity in the lower region, but on the bottom half, take a look at four red stripes (upper left corner of the lower half of the image) extending well beyond 50Hz. That is the EDM heavy and low beat that is played in this song.

One thing that makes my visitors laugh is when I say I plan to get a sub.

Could you give your opinions and assumptions on this huge difference you see in these measurements and the output I'm getting at my place, please?View attachment 76261

My bad, saw that you listed the amplifier already. The measurements are close mic-ed? And you have pink noise measurements rather than a specific song?
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,186
This review made me register and I'm skipping formal introduction. I doubt anyone's interested, but if you are, I'll do it later.

I have a question which will ask of you to speculate, so if you have a wooden pencil lying around, please bite on it. ;-)
To ask this question I can't hold nothing back so I'll say I'm extremely satisfied with these speakers and most of these comments how it's just another unremarkable speaker don't bother me. I'm holding onto mine until death or lottery.

But what I really can't comprehend is this lack of bass you all talk about. (???????) When I was shopping for bookshelves no other had lows nowhere near what these spouted (Revel included, perhaps it was a poor set up in the shop). First of all these speakers venture into tactile bass. Me sitting 9 feet from them and vibrations enter through couch (then my ass, sorry to say) and shake my guts at 80dB measured.

All my guests drop their jaw when they hear the lows and most of them state it's too much and want them lowered. (No DSP, Pure Direct on an Yamaha ax-890 110Wpch into 8). I have them foot and a half from the back wall and 5 feet from side walls.

And now for the walls (as Oclee stated himself about the difference in US and EU houses and in room response). I have these speakers in a 22 square meter room (240 square feet) in front of a very hard, bare concrete wall. It is 5,5 meters long (16-17 feet). Simple formula speaks of a bump in 60Hz region.

But is this enough to explain so much bass? I'm sharing a (not so precise) spectrogram of these speakers playing a song that goes bellow 30Hz. You can see the activity in the lower region, but on the bottom half, take a look at four red stripes (upper left corner of the lower half of the image) extending well beyond 50Hz. That is the EDM heavy and low beat that is played in this song.

One thing that makes my visitors laugh is when I say I plan to get a sub.

Could you give your opinions and assumptions on this huge difference you see in these measurements and the output I'm getting at my place, please?View attachment 76261
It is normal that with the lowest room mode, room gain and gain through placing loudspeakers close to solid surfaces you get a much enhanced bass compared to the anechoic measurements, here is a moving microphone measurement of my LS50 around my listening position when I used them as my desktop system, you can see they go down linearly till aproximately 33 Hz!

1596364331600.png
 

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
My bad, saw that you listed the amplifier already. The measurements are close mic-ed? And you have pink noise measurements rather than a specific song?
I'm sorry, but no, I don't. The mic was at my LP. I can do a pink noise. Will this reveal low in room response?

All cards on the table, I don't have any measuring gear, this is just my smartphone app. I do, however, have the pink noise on a Sheffield Test CD.

Sorry, how close should a mic be? One third octave or full bandwidth?
 
Last edited:

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
It is normal that with the lowest room mode, room gain and gain through placing loudspeakers close to solid surfaces you get a much enhanced bass compared to the anechoic measurements, here is a moving microphone measurement of my LS50 around my listening position when I used them as my desktop system, you can see they go down linearly till aproximately 33 Hz!

View attachment 76271
So all these things combined will reinforce the lows enough for this listening experience I and most of my guests are having? In other words, I had some "luck in the draw"? A speaker with more lows might have been too much and I could've ended up REQing them of?

Would this also mean that a sub would provide smoother and deeper lows rather than more lows?
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,186
So all these things combined will reinforce the lows enough for this listening experience I and most of my guests are having? In other words, I had some "luck in the draw"? A speaker with more lows might have been too much and I could've ended up REQing them of?
Exactly, in such a placement a loudspeaker with deeper linear bass often needs some EQ to reduce bass.

Would this also mean that a sub would provide smoother and deeper lows rather than more lows?
The biggest advantage of a well integrated sub is that you take the lower bass away from the satellites, making them playing with less distortions and higher SPL limits. Also with subwoofers you have more degrees of freedom in placement being able to compensate more problems like SBIR and with many subwoofers you can also have more constant bass at different locations in the room or even use them against room modes
 

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
Here:
Screenshot_20200802-125220.png

You should read it. I see less since I know less. This is full bandwidth pink noise.
 
Last edited:

whazzup

Addicted to Fun and Learning
Joined
Feb 19, 2020
Messages
575
Likes
486
@whazzup what does this pink noise tell you? Do you see anything different?

Note that I don't have the LS50. Though from the graph, there's a clear dropoff from ~130hz onwards, which is more in line (though still different) with the measurements on the 1st page. As for your earlier graph in #422, I suppose your recorded song has a lot of bass for the graph to show that way?
 

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
Note that I don't have the LS50. Though from the graph, there's a clear dropoff from ~130hz onwards, which is more in line (though still different) with the measurements on the 1st page. As for your earlier graph in #422, I suppose your recorded song has a lot of bass for the graph to show that way?
Well, thank you for your comment. Don't mind me saying but I get the feeling that the comparison you're making would make more sense if I was trying to prove my pair has more bass than the tested pair.

But I wasn't. I was trying to say that in my room they perform better in the low region. That's why I delibratly posted the graph of a bass heavy number to try and demonstrate that they do deliver when asked to.

Thank you for your input, though. I think I got it covered now.
 

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
One of my main speakers in my media room lost a mid driver last month, so I set up my LS50s on stands in there for the time being. With limited time invested in placement via measurement and ear, this is where they currently sit. Below 400Hz is room modes. I will work on placement more when I have time to see if I can improve the 750Hz dip, then I will apply some EQ to smooth out the bass. As you can see, LS50s are not bad at all, if you can leverage the room.


KEF LS50 Media Room No EQ.png
 

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
^ And this is the measured corrected response without subs after optimizing placement and applying room EQ. This is with the same PEQ applied to both channels using EQ APO.

LS50 Stereo MMM Corrected.png


I have applied per-channel PEQ in my Yamaha AVR and MiniDSP OpenDRC-DI and dialed it in even better, but those are more difficult pathways to measure. I have also integrated both subs by ear. I may find time to measure the whole system this weekend and optimize the xover point and EQ for the addition of the subs.

As much as I like these, I think the end game is to lay one on its side as a center channel and get KEF R5s as the stereo pair. I just like listening to towers without subs.
 

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
^ And this is the measured corrected response without subs after optimizing placement and applying room EQ. This is with the same PEQ applied to both channels using EQ APO.

View attachment 82556

I have applied per-channel PEQ in my Yamaha AVR and MiniDSP OpenDRC-DI and dialed it in even better, but those are more difficult pathways to measure. I have also integrated both subs by ear. I may find time to measure the whole system this weekend and optimize the xover point and EQ for the addition of the subs.

As much as I like these, I think the end game is to lay one on its side as a center channel and get KEF R5s as the stereo pair. I just like listening to towers without subs.
You've done some nice job there! What do you think is the 3db dip at 60Hz?
 

Steve Dallas

Major Contributor
Joined
May 28, 2020
Messages
1,201
Likes
2,784
Location
A Whole Other Country
You've done some nice job there! What do you think is the 3db dip at 60Hz?

Thanks!

Room modes. My room is nearly a cube. I have added some bass traps, which helped a little, but we all know how difficult taming bass frequencies in a room with traps is. I might be able to do something with that on the MiniDSP side, since it supports FIR.

The original 750Hz dip was solved by experimenting with positioning. The PEQ filters are all cuts in the bass frequencies.

Since I shot that measurement, I reduced the cuts below 300Hz by 1dB to warm things up a bit. The peaks down there are now all a little above the target line, which conforms more to the Harmon curve, I believe.
 
Last edited:

Maiky76

Senior Member
Joined
May 28, 2020
Messages
440
Likes
3,706
Location
French, living in China
Hi,

@thewas_ asked me to have a look at his EQ and run my optimizer.
here are the results:
Score with No EQ: 4.6
For AB comparison the EQed version need to be boosted and the LF Boost may result in a reduced dynamic range.

The initial fitting by @thewas_ is pretty good.
Only minor changes after the optimization resulting a slightly higher score.
Mostly due to smoother PIR and ON which is not easy to do by hand.
the Final gain no EQ vs Optimized EQ is quite decent.

Code:
TheWas_ EQ:
Score: 6.16

Type      Freq      Gain     Q
PEQ       67.2,     3.00,   2.00,...
PEQ      170.0,    -2.00,   1.00,...
PEQ      540.0,     1.60,   5.00,...
PEQ      830.0,    -2.40,   2.50,...
PEQ     1240.0,    -1.20,   8.00,...
PEQ     1650.0,     2.50,   3.00,...
PEQ     2640.0,    -2.80,   2.00,...
PEQ     4620.0,    -3.40,   3.00,...

Score after Optimization: 6.36

Type      Freq      Gain     Q
PEQ       65.0,     3.00,   1.78,...
PEQ      166.5,    -2.10,   0.88,...
PEQ      550.0,     1.40,   3.72,...
PEQ      830.0,    -2.40,   1.83,...
PEQ     1224.0,    -1.34,   9.45,...
PEQ     1609.0,     2.11,   2.65,...
PEQ     2764.0,    -2.80,   1.35,...
PEQ     4747.0,    -3.40,   3.00,...

EQ design
Kef LS50 Thewas vs Optimized EQ Design.png



Kef LS50 Thewas EQ Spinorama.png
Kef LS50 Thewas Optimized EQ Spinorama.png
Kef LS50 Thewas vs Optimized PIR-LW-ON Zoom.png
Kef LS50 Thewas vs Optimized Tonal.png
Kef LS50 Thewas vs Optimized Radar.png

No EQ vs Optimized
Kef LS50 No EQ vs Thewas Optimized Radar.png
 

Attachments

  • Kef LS50 Thewas EQ Design.png
    Kef LS50 Thewas EQ Design.png
    183.8 KB · Views: 179

Maiky76

Senior Member
Joined
May 28, 2020
Messages
440
Likes
3,706
Location
French, living in China
Hi,

@thewas_ asked me to have a look at his EQ and run my optimizer.
here are the results:
Score with No EQ: 4.6
For AB comparison the EQed version need to be boosted and the LF Boost may result in a reduced dynamic range.

The initial fitting by @thewas_ is pretty good.
Only minor changes after the optimization resulting a slightly higher score.
Mostly due to smoother PIR and ON which is not easy to do by hand.
the Final gain no EQ vs Optimized EQ is quite decent.

Code:
TheWas_ EQ:
Score: 6.16

Type      Freq      Gain     Q
PEQ       67.2,     3.00,   2.00,...
PEQ      170.0,    -2.00,   1.00,...
PEQ      540.0,     1.60,   5.00,...
PEQ      830.0,    -2.40,   2.50,...
PEQ     1240.0,    -1.20,   8.00,...
PEQ     1650.0,     2.50,   3.00,...
PEQ     2640.0,    -2.80,   2.00,...
PEQ     4620.0,    -3.40,   3.00,...

Score after Optimization: 6.36

Type      Freq      Gain     Q
PEQ       65.0,     3.00,   1.78,...
PEQ      166.5,    -2.10,   0.88,...
PEQ      550.0,     1.40,   3.72,...
PEQ      830.0,    -2.40,   1.83,...
PEQ     1224.0,    -1.34,   9.45,...
PEQ     1609.0,     2.11,   2.65,...
PEQ     2764.0,    -2.80,   1.35,...
PEQ     4747.0,    -3.40,   3.00,...

EQ designView attachment 84181


View attachment 84182View attachment 84180View attachment 84177View attachment 84178View attachment 84179
No EQ vs Optimized
View attachment 84176

A new version is going to be available.
Some preliminary data here:
https://www.audiosciencereview.com/...echnology-mat-coming.16133/page-2#post-522531
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,186
Hi,

@thewas_ asked me to have a look at his EQ and run my optimizer.
here are the results:
Score with No EQ: 4.6
For AB comparison the EQed version need to be boosted and the LF Boost may result in a reduced dynamic range.

The initial fitting by @thewas_ is pretty good.
Only minor changes after the optimization resulting a slightly higher score.
Mostly due to smoother PIR and ON which is not easy to do by hand.
the Final gain no EQ vs Optimized EQ is quite decent.

Code:
TheWas_ EQ:
Score: 6.16

Type      Freq      Gain     Q
PEQ       67.2,     3.00,   2.00,...
PEQ      170.0,    -2.00,   1.00,...
PEQ      540.0,     1.60,   5.00,...
PEQ      830.0,    -2.40,   2.50,...
PEQ     1240.0,    -1.20,   8.00,...
PEQ     1650.0,     2.50,   3.00,...
PEQ     2640.0,    -2.80,   2.00,...
PEQ     4620.0,    -3.40,   3.00,...

Score after Optimization: 6.36

Type      Freq      Gain     Q
PEQ       65.0,     3.00,   1.78,...
PEQ      166.5,    -2.10,   0.88,...
PEQ      550.0,     1.40,   3.72,...
PEQ      830.0,    -2.40,   1.83,...
PEQ     1224.0,    -1.34,   9.45,...
PEQ     1609.0,     2.11,   2.65,...
PEQ     2764.0,    -2.80,   1.35,...
PEQ     4747.0,    -3.40,   3.00,...

EQ designView attachment 84181


View attachment 84182View attachment 84180View attachment 84177View attachment 84178View attachment 84179
No EQ vs Optimized
View attachment 84176
I have now listened to your Matlab optimised EQ for several hours and must say its the best sounding one I ever had on my LS50s, including the ones based on listening position measurements, its the first time in 9 years I like my LS50 sound even more than the one of my very good Neumann KH120, so big respect both to you for your coding but also to the Harman score which seems to work really good at least at this case!
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,186
I imported the listening window measurements of the new LS50 Wireless II and used them as a target to adapt the existing 8 EQ filters of the ASR measured passive LS50 listening window, match quite well:

1601295174268.png


They are not very different to the existing ones of Maiky76 and mine except they have a bit more upper bass and lower mids, which give a tad warmer sound, here are the filter coefficients if someone wants to test them:

Code:
Filter  1: ON  PK       Fc   70.00 Hz  Gain   3.00 dB  Q  1.780
Filter  2: ON  PK       Fc   151.4 Hz  Gain  -1.60 dB  Q  2.704
Filter  3: ON  PK       Fc   565.4 Hz  Gain   3.00 dB  Q  1.000
Filter  4: ON  PK       Fc   755.0 Hz  Gain  -3.20 dB  Q  1.246
Filter  5: ON  PK       Fc    1254 Hz  Gain  -1.40 dB  Q  6.629
Filter  6: ON  PK       Fc    1592 Hz  Gain   3.00 dB  Q  1.981
Filter  7: ON  PK       Fc    2747 Hz  Gain  -3.40 dB  Q  1.032
Filter  8: ON  PK       Fc    4692 Hz  Gain  -2.40 dB  Q  4.273
 

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,297
Likes
5,075
Location
Nashville
I imported the listening window measurements of the new LS50 Wireless II and used them as a target to adapt the existing 8 EQ filters of the ASR measured passive LS50 listening window, match quite well:

View attachment 85168

They are not very different to the existing ones of Maiky76 and mine except they have a bit more upper bass and lower mids, which give a tad warmer sound, here are the filter coefficients if someone wants to test them:

Code:
Filter  1: ON  PK       Fc   70.00 Hz  Gain   3.00 dB  Q  1.780
Filter  2: ON  PK       Fc   151.4 Hz  Gain  -1.60 dB  Q  2.704
Filter  3: ON  PK       Fc   565.4 Hz  Gain   3.00 dB  Q  1.000
Filter  4: ON  PK       Fc   755.0 Hz  Gain  -3.20 dB  Q  1.246
Filter  5: ON  PK       Fc    1254 Hz  Gain  -1.40 dB  Q  6.629
Filter  6: ON  PK       Fc    1592 Hz  Gain   3.00 dB  Q  1.981
Filter  7: ON  PK       Fc    2747 Hz  Gain  -3.40 dB  Q  1.032
Filter  8: ON  PK       Fc    4692 Hz  Gain  -2.40 dB  Q  4.273
I have both this one and Maiky76's optimum eq for my OG LS50s on my miniDSP 2x 4 HD with an SB2000 crossed over at 100 HZ w 24 db per octave Linkwitz-Reilly high/low pass, and I think I prefer Maiky76's eq. It just seems smoother on first listen.
 
Last edited:

killdozzer

Major Contributor
Joined
Aug 2, 2020
Messages
1,615
Likes
1,628
Location
Zagreb
I'm not all that graph-savy. Can someone point out if there's a measurement in this thread that concerns IMD in LS50?
 
Top Bottom