• 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!

Can you choose a great speaker by its frequency response?

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
3,517
Likes
7,027
Location
Stow, Ohio USA
A little bit of fun here as was exploring Klippel data for what makes a better speaker. To keep this post simpler for now, am just going to share simple on-axis frequency response graphs of 3 highly regarded speakers that Amir has measured. While we know there is more to consider, on-axis linearity is generally considered to be a key metric. So let's see how well members can discern speakers based on it...

To make it a bit more challenging, the plots are +/- 3dB. If I used the usual SPL spread, it will be easier to figure out which speakers were chosen (as I suspect some of you may do anyway). All are plotted with the same frequency range. All source data is from Amir's Klippel testing.

So here they are:

Speaker 1

1651925470806.png


Speaker 2

1651925731711.png


Speaker 3
1651925791304.png


All of these speakers are unique in some way from a design perspective relative to each other but did not give much consideration to how much they each cost. As I suspect with many of you, my first takeaway is how forgiving our hearing must be for these to be some of the better speakers on the market. For that matter, this also seems likely why manufacturers publish a tolerance spec rather than a graph!

So please post your preferred speaker and why. Also, include why you did not consider the other 2 speakers to be as good and why.

Depending on the responses, I plan to reveal the identities of each speaker by the start of next week. Just to be clear as to my impartiality here, I do not own and have not heard any of these speakers.

Have fun!

Rick
 
Last edited:
D

Deleted member 46664

Guest
The scaling on the vertical axis matters ... but it is obscured. If that's 3db per division those are all pretty good speakers. If it's 10db per division I wouldn't buy any of them.

Also based on major peaks and valleys either there's a lot of room-effect in those charts or the first two are likely the same speaker.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
The vertical scaling is sooo narrow that it's not possible to make a good judgement, I think. Information is just not enough from this single perspective. But I'm just going to throw in my guess that the "smoothest looking" with the most bass extension (1st graph) will be more preferred. Last plot looks the "worst" of the set, I suppose.
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
3,517
Likes
7,027
Location
Stow, Ohio USA
Could you annotate the ordinate of the FRs please?

As mentioned, the intent was to obscure (somewhat), but if it helps, treat the major y-axis divisions as 0.5 dB. So, each small division is about 0.1 dB
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
3,517
Likes
7,027
Location
Stow, Ohio USA
The scaling on the vertical axis matters ... but it is obscured. If that's 3db per division those are all pretty good speakers. If it's 10db per division I wouldn't buy any of them.

Also based on major peaks and valleys either there's a lot of room-effect in those charts or the first two are likely the same speaker.

No tricks here and all measurements are from Amir's Klippel. :)
 

Darkscience

Active Member
Joined
Dec 15, 2020
Messages
109
Likes
102
I like speaker one, I guess? I would need to see what the speaker looks like, aesthetics is high on my list of needs. Price is irrelevant to me, (except on the higher end above my budget, then I can't consider that speaker).
 
D

Deleted member 46664

Guest
As mentioned, the intent was to obscure (somewhat), but if it helps, treat the major y-axis divisions as 0.5 dB. So, each small division is about 0.1 dB

Zoomed in that tightly ... all three speakers will likely sound about the same since you have no more than +- 4db of variation.
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
3,517
Likes
7,027
Location
Stow, Ohio USA
Zoomed in that tightly ... all three speakers will likely sound about the same since you have no more than +- 4db of variation.

As stated, I have not heard any of them. From Amir's reviews, am very sure they are quite different.

Have to agree with the earlier comment, their bass performance is clearly different. :)
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,180
While we know there is more to consider, on-axis linearity is generally considered to be a key metric.
I don't agree, especially at this level zooming, without having off axis / directivities measurements you cannot choose a loudspeaker as for example on some designs you have some dips on axis that disappear off axis and are not significantly audible. In the other case FR problems can be easily EQed if the directivity is smooth but you cannot correct directivity problems. Also in my experience in upmost cases loudspeakers with a smooth directivity and some on-axis ripples sound better then the other way around, which is reflected also in the Harman score. Even LW is correlated better to perceived sound than on-axis measurements.
So coming back to the title of this thread "Can you choose a great speaker by its frequency response?" my answer is "No, if it is in that small tolerance band and if off-axis/directivites are not shown." Of course if the FR responses of some loudspeaker had a much higher tolerance band or were dropping off much earlier, the chances to be not preferred would be significant.
 
Last edited:

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,424
Likes
7,940
Location
Brussels, Belgium
I don't agree, especially at this level zooming, without having off axis / directivities measurements you cannot choose a loudspeaker as for example on some designs you have some dips on axis that disappear off axis and are not significantly audible. In the other case FR problems can be easily EQed if the directivity is smooth but you cannot correct directivity problems. Also in my experience in upmost cases loudspeakers with a smooth directivity and some on-axis ripples sound better then the other way around, which is reflected also in the Harman score. Even LW is correlated better to perceived sound than on-axis measurements, sorry.
While all these comments are valid, it's very clear that your comments are based on the assumption that the on-axis is already well behaved.

a miss-behaved on-axis with a lot of resonances always results in a miss-behaved off-axis. but a good-behaved on-axis doesn't always means a good behaved off-axis.

As a result on-axis / listening window is the biggest determinant of sound quality.
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,180
While all these comments are valid, it's very clear that your comments are based on the assumption that the on-axis is already well behaved.
Which seems to be the case here as like he writes the zooming is very high.

a miss-behaved on-axis with a lot of resonances always results in a miss-behaved off-axis.
Yes, as said that doesn't seem to be the case here though, also small ripples due to diffraction and symmetry effects (especially on coaxial driver) can get smaller off-axis or the more important LW.

As a result on-axis / listening window is the biggest determinant of sound quality.
LW shares in my experience the top place together with directivity, the higher the percentage of direct sound at the listening position there is the higher the influence of the first and vise versa.
 

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,462
Location
Land O’ Lakes, FL
Speaker 1 has deeper bass but it’s shelved down, so even though it doesn’t fall off till later, it may sound bass-lite.

Speaker 2 rolls off normally, but the response is tilted, so it could sound treble-lite or even boomy in the midbass.

Just going off the on-axis response, Speaker 3 doesn’t seem to have any benefit over either.
 
Top Bottom