• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Wharfedale Diamond 12.1 Review (Speaker)

witwald

Active Member
Joined
Jul 23, 2019
Messages
196
Likes
183
That kind of dip is quite popular in many designs and it's beneficial for many forward sounding records.
There are also distant sounding recordings that will not benefit from such a dip, being made to sound even more distant. It would seem that a wide built-in dip such as this one, about 2dB in magnitude centered on 2kHz and covering two octaves, isn't really going to help things from an accuracy perspective. Application of a parametric equaliser would allow a user to adjust the sound to suit each recording if so desired. This speaker's approach doesn't have that flexibility. Shouldn't a loudspeaker try and be more faithful to the original source, rather than introducing a somewhat wide tonal imbalance?
 

witwald

Active Member
Joined
Jul 23, 2019
Messages
196
Likes
183
Well, off-topic and my personal view after years of filter construction and listening, but a dip around 2 kHz will exaggerate the comb-filtering, "stereo system error", present when you have the speakers in a stereo setup in the usual 22.5°-30° angles.
Can you elaborate a little on how it might do that? I'd like to try and simulate the effect in VituixCAD.
Thus, in my view any dips should occur 3-4 kHz, and 7-8 kHz.
I don't understand why having a set of two dips would enhance the naturalness of sound reproduction. Can you please describe the concepts behind this approach?
The only way to reduce the 2 kHz stereo system dip is to have reflection "fill-in", hence the radiated lateral response must have sufficient energy in that region.
I'm not entirely sure that it is possible to do that using the same drivers. After all, the dip is being caused by the interaction of the acoustic high-passed and low-passed responses. For some measurement axes the dip may no longer be present, due to the crossover topology in use, but if that is the case then it may be better to simply design the crossover to be less sensitive to that type of effect. Relying on the radiated lateral response to "fill-in" the dip seems fraught with uncertainty owing to the variability of typical listening rooms.
 
Last edited:

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
1,922
Likes
7,285
Location
NYC
I'm not sure that's something you can really do in VCAD, but this image from Toole's book(chapter 7) explains it:

Screenshot_20210922-184125.png
 

witwald

Active Member
Joined
Jul 23, 2019
Messages
196
Likes
183
Code:
EQ for Wharfedale Diamond 12.1 computed from ASR data
Preference Score 5.4 with EQ 5.8
Generated from http://github.com/pierreaubert/spinorama/generate_peqs.py v0.11
Dated: 2021-09-22-05:52:02

Preamp: -2.4 dB

Filter  1: ON PK Fc  1685 Hz Gain +2.52 dB Q 4.87
Filter  2: ON PK Fc  2346 Hz Gain +1.25 dB Q 6.00
Filter  3: ON PK Fc   465 Hz Gain -0.60 dB Q 0.05
Filter  4: ON PK Fc  2773 Hz Gain +1.41 dB Q 6.00
Filter  5: ON PK Fc  1207 Hz Gain +0.93 dB Q 6.00
Filter  6: ON PK Fc   542 Hz Gain -0.65 dB Q 2.08
Filter  7: ON PK Fc  1965 Hz Gain +0.73 dB Q 6.00
Filter  8: ON PK Fc 15538 Hz Gain +0.84 dB Q 6.00
Filter  9: ON PK Fc  4016 Hz Gain -0.85 dB Q 2.33
That is a lot of parametric EQ being applied to the response of this loudspeaker, some of it with relatively high Q values. Considering that the loudspeaker seems to have a wide built-in dip, about 2dB in magnitude centered on 2kHz and covering two octaves, I'd expect that a single parametric EQ filter in that region could suffice to bring up the energy to be relatively flat on axis. Wouldn't many of the other small peaks and dips be caused to a large extent by sound diffraction effects. I am concerned that EQing individual peaks and dips resulting from diffraction effects is not entirely a good idea.
 

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
1,500
Likes
1,428
May not of such an importance, but, but I find interesting that after many decades of sound reproduction, the frequency ranges are still not firmly agreed on, seams to depend on what you read, who you talk to. I'm not pointing a mistake, just odd that there is no consensus. Personnally, I've always referred to the 2K region as being in the mids, maybe upper mids, but not lower treble as mentioned in this review. It's just interesting because we are all audio saavy and discuss about this all the time but sometimes words don't mean the same thing for everybody. I know it's off topic, but what do guys think? What you think would be the most universally accepted reference for frequency ranges? Just a tought.
 

witwald

Active Member
Joined
Jul 23, 2019
Messages
196
Likes
183
Near-field measurements show that this is one of the rarer designs that keeps the internal port and cabinet resonances at bay:
View attachment 154898
It would be instructive to see the individual far-field (or approximation) filtered on-axis acoustic responses of the woofer and the tweeter in this loudspeaker, as well as the total summed response at the measuring point. As the loudspeaker has bi-amp terminals, that would not be difficult to accomplish, as it would require only three additional frequency response measurements. The results would show us how the crossover has been designed in order to integrate the outputs from the woofer and tweeter.
 

witwald

Active Member
Joined
Jul 23, 2019
Messages
196
Likes
183
I'm not sure that's something you can really do in VCAD, but this image from Toole's book(chapter 7) explains it:
View attachment 155089
Thanks for that. I think I understand what's going on. It's related to the time delay differences between sound reaching each of the listener's two ears. Hence, what has a change in frequency response of the loudspeaker got to do with this effect? The relative time differences aren't changed. All that's changed is the intensity of the sound reaching the ears. Once we get to relative level differences, then that original difference in level has been subtracted out by the differencing process. Or am I missing something?
 

InsideTheWire

Member
Joined
Apr 11, 2021
Messages
19
Likes
28
Location
Central Illinois
I bought a set of these earlier this year and I am not disappointed. I was having a hard time deciding between the 12.3s or the 12.1s. I was surprised at how good these sound in my modest system. It's nice to see that they have measured pretty nicely too!
 

MarkWinston

Senior Member
Joined
Jul 13, 2021
Messages
474
Likes
239
Finally some [American] measurements of the Evo models.
Too bad it’s not the 3-way 4.2.
Worry not as many people prefer the sound of the 4.1s over the 4.2s for whatever reason. Measurements will tell why.
 

SS55

Member
Joined
Aug 8, 2021
Messages
39
Likes
40
What do you think are the best objective measurements that capture resolution/delicate details? Is it distortion at a given SPL? How smooth the FR is?
that’s interesting! What do you think is a measurable proxy for resolution?
I would like to know this too.
If I had to guess, I would say lack of resonance which would also show up in measurements with a smoother frequency response.
 

ROOSKIE

Addicted to Fun and Learning
Joined
Feb 27, 2020
Messages
893
Likes
1,589
Location
Minneapolis
that’s interesting! What do you think is a measurable proxy for resolution?
You are touching one of the most important questions.
I really think there is currently no clear answer.
I have zero experience with this Wharfedale model, with previous Wharfedale Diamond models I have tried I felt they were not resolving speakers subjectively.
I recently used both the ELAC DBR62 and the POLK L200. They measured in room almost exactly the same. Granted an in room MMM reading from the listening position only tells you so much about a speaker, that said those speakers did not resolve equally. It wasn't that close. Neither where they equal in Dynamics, not really very close. (I preferred the Polk in everyway, it does also cost over twice as much to be fair)
I had a similar experience with the JBL A130 and the Wharfedale Diamond 225, the JBL was much more resolving and "clean" sounding even with the boosted mid's taken down with PEQ.
Anyway in terms of how to fully measure the Resolution and Dynamics of a speaker (which I find to both be very important), there seems to be a real debate over how to demonstrate it in testing.
Surely low IMD distortion and low compression would be big factors but it seems there is more.
 
Last edited:

Maiky76

Senior Member
Joined
May 28, 2020
Messages
301
Likes
2,137
Location
French, leaving in China
This is a review and detailed measurements of the Wharfedale Diamond 12.1 bookshelf speaker. It was kindly purchased by a member and drop shipped to me. It costs US $399 for a pair.

The look and feel of the 12.1 is definitely above the budget category:

View attachment 154893

The back is rather pedestrian but more beefy than other speakers in its class:

View attachment 154894

Overall a good look for the speaker.

Measurements that you are about to see were performed using the Klippel Near-field Scanner (NFS). This is a robotic measurement system that analyzes the speaker all around and is able (using advanced mathematics and dual scan) to subtract room reflections (so where I measure it doesn't matter). It also measures the speaker at close distance ("near-field") which sharply reduces the impact of room noise. Using computational acoustics, far-field response is computed and that is what I present. Both of these factors enable testing in ordinary rooms yet results that can be more accurate than an anechoic chamber.

I performed over 1000 measurement which resulted in error rate of about 1%. Clean high frequency response is responsible for ease of measurement in this regard.

Reference axis is approximately the center of the tweeter. Grill was not used.

Wharfedale Diamond 12.1 Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker is and how it can be used in a room. This so called spinorama shows us just about everything we need to know about the speaker with respect to tonality and some flaws:

View attachment 154895

The all important on-axis response (black) is almost flat which is great. There is a bit of a valley around 2 to 3 kHz which unfortunately gets deeper off axis making directivity rather poor there. We can see it better in early window reflections:

View attachment 154896

It takes a rather severe penalty there. Lack of a waveguide causes the woofer beamwidth to be narrower than tweeter's around the crossover frequencies. Fortunately you can use a thick rug to absorb the floor reflections like I have. And high ceilings help too with the ceiling bounce.

Putting the two together we get the expected results which is good:

View attachment 154897

Near-field measurements show that this is one of the rarer designs that keeps the internal port and cabinet resonances at bay:

View attachment 154898

We routinely see these resonances being higher in frequency and competing with the tweeter/woofer response. Not here. The port does its thing to extend low frequencies and that is it. As it should be.

Back to directivity, we see its measure better here:

View attachment 154899

We see the beam width narrowing (woofer) then widening (tweeter). One benefit here though is that beam width is 20 or so degrees wider in each direction than is typical of these speakers. There is on going discussion as to trade off here.

Here is our color 3-D map of the same:

View attachment 154900

There is that narrowing of the high frequencies (same beaming but now applied to tweeter) and I wonder if this makes the speaker less bright to my ears.

Here is vertical with a bit more allowance for height than normal:
View attachment 154901

And our 3-D directivity at three frequencies:

View attachment 154902

Someone talked me into expanding my distortion measurements down to 76 dBSPL which is making the presentation hard but here it is anyway:

View attachment 154903

During measurements I did not hear much distortion above deep bass which indicates good power handling. Here is the same as a percentage:


View attachment 154905

Edit: forgot to post the impedance:

View attachment 154937

Wharfedale 12.1 Listening Tests
I powered the speakers in my usual far field setup and started to listen. One track, then two, then three. I am not hearing much to complain about! Yet we had that directivity error and some lower treble dip. Brought out the EQ to fill those in:

View attachment 154909

This made the vocals, especially that of females, to stand out more which I liked. And added a bit of resolution to them as is typical of this type of boost. On some tracks I thought there was a bit extra brightness but overall, I liked it better with EQ than without.

Power handling was excellent. Speaker simply doesn't do what it can't do, i.e. deep bass. As a result even with a single speaker, I could fill it with a ton of watts and it kept getting louder until I got scared before it did! There was some amount of tactile bass which was surprising and welcome

At this point I was puzzled that we had some flaws in measurements yet I am not able to put my finger on anything. So I pulled out my Revel M106 speaker and played it. The M106 was definitely a step up with much better resolution when it came to delicate details. It was clearly a better experience. That speaker is four times the price though so I changed it out for Revel M16. The M16 has a boosted upper bass and this was quite audible, creating a warmer, albeit slightly muddy sound without correction for my room mode which it activates. The M16 retails for double the price but you can get it for less discounted. Still, I didn't feel that it provided much of any advantage over Wharfedale 12.1 in this quick comparison.

Conclusions
Technical we have some technical flaws in the midst of good performance otherwise in the 12.1. Objectively this stands out a lot but in listening tests in my room at least with floor absorption and high ceilings, I was not able to identify it. Indeed I was stomped in trying to find much of any fault with the speaker. Directivity errors though mean that speaker sound will more room dependent so your experience may vary somewhat from mine. Still, I think Wharfedale has done a very good job here.

Overall, I am going to recommend the Wharfedale Diamond 12.1. You could almost push me to give it the highest honors had it not been due to directivity errors. It provides a quandary with respect to audibility of such flaws versus some other trade offs.

-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Hi,

NB: new forum is a bit cumbersome...

Here is my take on the EQ.

These EQ are anechoic EQ to get the speaker right before room integration. If you able to implement these EQs you must add EQ at LF for room integration, that usually not optional… see hints there: https://www.audiosciencereview.com/...helf-speaker-review.11144/page-26#post-800725

The raw data with corrected ER and PIR:

Score no EQ: 5.3
With Sub: 7.7

Spinorama with no EQ:
  • very decent
  • maybe tuned to try to compensate for the directivity error?
Wharfedale 12.1 No EQ Spinorama.png

Directivity:

Better stay at tweeter height
Horizontally, anything goes up to 20deg a bit of toe in will help
Wharfedale 12.1 2D surface Directivity Contour Only Data.png

Wharfedale 12.1 LW better data.png


EQ design:

I have generated two EQs. The APO config files are attached.
  • The first one, labelled, LW is targeted at making the LW flat
  • The second, labelled Score, starts with the first one and adds the score as an optimization variable.
  • The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.


Score EQ LW: 5.5
with sub: 7.9

Score EQ Score: 6.0
with sub: 8.4


Wharfedale 12.1 APO EQ Score 96000Hz September222021-171254 Preamp: -2.9 dB Filter 1: ON HPQ Fc 42.13, 0.00, 1.00 Filter 2: ON PK Fc 2271.80, 1.32, 3.58 Filter 3: ON PK Fc 1215.52, 0.94, 4.93 Filter 4: ON PK Fc 1643.13, 2.56, 4.85 Filter 5: ON PK Fc 2851.01, 1.17, 4.94 Filter 6: ON PK Fc 5680.14, -0.89, 4.05 Filter 7: ON PK Fc 18872.47, 1.19, 1.47 Filter 8: ON PK Fc 9936.00, 0.43, 2.71 Wharfedale 12.1 APO EQ LW 96000Hz September232021-110531 Preamp: -2.9 dB Filter 1: ON HPQ Fc 41.68, 0.00, 0.99 Filter 2: ON PK Fc 2171.51, 1.55, 1.51 Filter 3: ON PK Fc 1197.65, 1.03, 4.99 Filter 4: ON PK Fc 1643.64, 1.65, 4.93 Filter 5: ON PK Fc 2946.68, 1.21, 4.07 Filter 6: ON PK Fc 3740.94, -0.89, 3.63 Filter 7: ON PK Fc 14485.94, 1.52, 0.26 Filter 8: ON PK Fc 7736.56, 0.73, 4.73
Wharfedale 12.1 EQ Design.png


Spinorama EQ LW
Wharfedale 12.1 LW EQ Spinorama.png


Spinorama EQ Score
Wharfedale 12.1 Score EQ Spinorama.png


Zoom PIR-LW-ON
Wharfedale 12.1 Zoom.png


Regression - Tonal
Wharfedale 12.1 Regression - Tonal.png


Radar no EQ vs EQ score
Small improvements
Wharfedale 12.1 Radar.png


The rest of the plots is attached.
 

Attachments

  • Wharfedale 12.1 Reflexion data.png
    Wharfedale 12.1 Reflexion data.png
    139.4 KB · Views: 14
  • Wharfedale 12.1 LW data.png
    Wharfedale 12.1 LW data.png
    133.9 KB · Views: 11
  • Wharfedale 12.1 2D surface Directivity Contour Data.png
    Wharfedale 12.1 2D surface Directivity Contour Data.png
    273.9 KB · Views: 11
  • Wharfedale 12.1 3D surface Vertical Directivity Data.png
    Wharfedale 12.1 3D surface Vertical Directivity Data.png
    451.1 KB · Views: 12
  • Wharfedale 12.1 3D surface Horizontal Directivity Data.png
    Wharfedale 12.1 3D surface Horizontal Directivity Data.png
    430 KB · Views: 13
  • Wharfedale 12.1 Normalized Directivity data.png
    Wharfedale 12.1 Normalized Directivity data.png
    293.3 KB · Views: 14
  • Wharfedale 12.1 Raw Directivity data.png
    Wharfedale 12.1 Raw Directivity data.png
    445.7 KB · Views: 14
  • Wharfedale 12.1 APO EQ Score 96000Hz.txt
    439 bytes · Views: 13
  • Wharfedale 12.1 LW better data.png
    Wharfedale 12.1 LW better data.png
    140.8 KB · Views: 12
  • Wharfedale 12.1 Vertical 3D Directivity data.png
    Wharfedale 12.1 Vertical 3D Directivity data.png
    405.5 KB · Views: 11
  • Wharfedale 12.1 Horizontal 3D Directivity data.png
    Wharfedale 12.1 Horizontal 3D Directivity data.png
    398.4 KB · Views: 12
  • Wharfedale 12.1 APO EQ LW 96000Hz.txt
    436 bytes · Views: 11

dm127

New Member
Joined
Sep 23, 2021
Messages
2
Likes
1
It takes a rather severe penalty there. Lack of a waveguide causes the woofer beamwidth to be narrower than tweeter's around the crossover frequencies. Fortunately you can use a thick rug to absorb the floor reflections like I have. And high ceilings help too with the ceiling bounce.
This looks a lot more like a poorly-designed crossover than any beaming. Considering that the speaker uses a 5'' woofer, there is no reason for it to start beaming before 3KHz. The fact that the dip is accentuated in vertical measurements indicates exactly this - since the drivers are more out-of-phase, the sound coming from them cancels each-other more, leading to more severe dips.

Using a waveguide would only make things worse really as you'd put the acoustic centers of the drivers further apart, making the crossover behave even more poorly. Plus you'd make the tweeter beam around the crossover frequency while the woofer wouldn't.

To properly look for directivity issues in a speaker, normally you should look at the horizontal on-axis measurements, not the vertical ones - and really, you can see that in the two horizontal/vertical 3D maps provided - there is no anomaly on the horizontal axis around the 2-3kHz region, in fact, you can see that the directivity between the woofer and tweeter is perfectly matched. On the vertical axis on the other hand, there is plenty of cancellation around the xover region.

Another sign of the poor crossover can be seen in the nearfield measurements - the woofer's breakup is not properly controlled, being just 20dB down around 5.5kHz and 30dB down at 8kHz+ means that you will definitely hear the breakup whenever listening to this speaker.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
1,323
Likes
868
Location
Sweden
Can you elaborate a little on how it might do that? I'd like to try and simulate the effect in VituixCAD.

I don't understand why having a set of two dips would enhance the naturalness of sound reproduction. Can you please describe the concepts behind this approach?

I'm not entirely sure that it is possible to do that using the same drivers. After all, the dip is being caused by the interaction of the acoustic high-passed and low-passed responses. For some measurement axes the dip may no longer be present, due to the crossover topology in use, but if that is the case then it may be better to simply design the crossover to be less sensitive to that type of effect. Relying on the radiated lateral response to "fill-in" the dip seems fraught with uncertainty owing to the variability of typical listening rooms.

OT again, this has also been discussed in many threads. Given that a preferred speaker evaluated in mono has a linear frequency response, the same speaker will sound different in stereo mode for the same mono signal. This is due to the typical head-related interference as depicted by Shirley at al in a typical reflective room. Now having a speaker with dips and peaks that follows that curve will exaggerate the timbral shift. And a speaker with inverse curve will reduce the effect.

Toole also confirm these errors but instead suggests to leave stereo and add a center channel.
 

MarkWinston

Senior Member
Joined
Jul 13, 2021
Messages
474
Likes
239
I know cone size measurements are taken at the basket, but that 5 inch is actually more like 4 or even less. When it comes to cone sizes, wharfedale measures among the smallest surface area wise. My 6 and half inch woofer on the 12.2 dont even reach 5 and a quarter. All my wharfedales have the smallest '6 and a half inch' woofer compared to the rest of my other speakers.
 

witwald

Active Member
Joined
Jul 23, 2019
Messages
196
Likes
183
Given that a preferred speaker evaluated in mono has a linear frequency response, the same speaker will sound different in stereo mode for the same mono signal. This is due to the typical head-related interference as depicted by Shirley et al in a typical reflective room.
I follow what you're saying. Thanks for the example and the explanation.
Now having a speaker with dips and peaks that follows that curve will exaggerate the timbral shift. And a speaker with inverse curve will reduce the effect.
I think that I am following.

When auditioned in mono, won't a single equalised speaker not sound the same as the unequalised speaker? So what has been achieved? Isn't it just a different sound of the mono speaker due to equalisation differences? And in mono won't a speaker equalised to be non-flat produce an incorrect unnatural response when playing back a recording of a known mono source?

As far as stereo goes, isn't the aim to faithfully reproduce the frequency content of the recorded material? Adding the previously mentioned equalisation seems to be adding modifications/distortion, not making things better. I expect that stereo mixes are monitored on stereo loudspeaker systems. Hence, the typical head-related interference is baked into the mix when it is constructed by the mixing engineer. Don't we then each hear our own version of that mix, because we each have our own head-related transfer function to contend with? We don't need extra equalisation to produce what ostensibly is just another effect of sorts.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
1,323
Likes
868
Location
Sweden
I follow what you're saying. Thanks for the example and the explanation.

I think that I am following.

When auditioned in mono, won't a single equalised speaker not sound the same as the unequalised speaker? So what has been achieved? Isn't it just a different sound of the mono speaker due to equalisation differences? And in mono won't equalised speaker reproduce an unnatural response when playing back a recording of a known mono source?

As far as stereo goes, isn't the aim to faithfully reproduce the frequency content of the recorded material? Adding the previously mentioned equalisation seems to be adding modifications/distortion, not making things better. I expect that stereo mixes are monitored on stereo loudspeaker systems. Hence, the typical head-related interference is baked into the mix when it is constructed by the mixing engineer. Don't we then each hear our own version of that mix, because we each have our own head-related transfer function to contend with?

If these things were recorded in the stereo mix, a preferred speaker evaluated in mono in a typical reflective room, the on-axis response should mimic the curve depicted by Shirley et al. But they don’t, according to current research.
 

dm127

New Member
Joined
Sep 23, 2021
Messages
2
Likes
1
Thanks to those that calculate the Spinorama after EQ, we can see that the directivity error at the crossover point isn’t that big of a deal, it is indeed mostly the dip on-axis that causes the dip in the ER/PIR/SP.

My point was that it isn't a directivity error but a crossover design issue. Either the speakers aren't correctly level-matched at the xover point, they are out of phase, or both. Boosting that region will help with a flatter response but will also increase the midrange distortion, seeing that the 2-4kHz region already has some distortion at moderate listening levels.
 
Top Bottom