Wharfedale Diamond 220 Budget Speaker Review

[amirm]

Have you tried using the calculated EQ that people can build from your measurements for such a speaker? Some of those EQ settings seem to produce amazing looking improvements.

I have not but it is clear that in most cases EQ significantly improves the performance of these speakers. To some extent we have to decide if no-EQ version even matters!

 

[amirm]

Is there a way to assign some type of scale(s) to the conclusions of speaker tests? Or even a panther chart with a $-$$$$ indicator (since actual prices vary) so one knows what reviews to revisit when deciding on a speaker purchase? If I had missed this review and then went to the preference chart when buying speakers I would not have given the wharfedales a second look.

We have to leave some benefit for those who read every review, no?

Yeh, it is an imperfect system to be sure. Maybe members can create digests for these reviews.

 

[Soniclife]

I have not but it is clear that in most cases EQ significantly improves the performance of these speakers. To some extent we have to decide if no-EQ version even matters!

It's a good philosophical question, I'm leaning towards the no EQ impression isn't very important, but we need more evidence that EQ works. Correcting speakers above the transition frequency is tricky, our mono in room mics are not the same as what we hear, if ready made EQ could be downloaded that corrected above the end user may get to optimise the hardest part easily.

 

[MZKM]

Is there a way to assign some type of scale(s) to the conclusions of speaker tests? Or even a panther chart with a $-$$$$ indicator (since actual prices vary) so one knows what reviews to revisit when deciding on a speaker purchase? If I had missed this review and then went to the preference chart when buying speakers I would not have given the wharfedales a second look.

We have to leave some benefit for those who read every review, no?

Yeh, it is an imperfect system to be sure. Maybe members can create digests for these reviews.

https://www.audiosciencereview.com/forum/index.php?pages/SpeakerTestData/

Includes whether it was recommended. It is not always/currently up to date though.

I can add this data to the spreadsheet list if people wish. Maybe I'll just go by the panther type, as it'd be a bit confusing to parse otherwise. If you want a chart though, I don't want to make things too cluttered, so maybe just a Performance: Price graph that excludes anything that got a headless panther.

Headless: 1
Shrug: 2
Relax: 3
Golf/Soccer: 4

 

[infinitesymphony]

Have you tried using the calculated EQ that people can build from your measurements for such a speaker? Some of those EQ settings seem to produce amazing looking improvements.

This seems like one possible objective approach to determining how well a speaker might respond to EQ. Measure the speaker, calculate EQ compensation and apply, remeasure. Or if remeasuring is not an option, a less accurate approach might be to use the predicted post-EQ frequency response to calculate a post-EQ preference score.

My understanding is that @Maiky76 used this method on the Edifier R1280T to bring its preference score from 1.97 to 5.80 in theory:

https://www.audiosciencereview.com/...280t-powered-speaker-review.16112/post-520312

 

[vkvedam]

Excellent, I was not expecting a golfing panther though which makes it even interesting

 

[sweetchaos]

I can add this data to the spreadsheet list if people wish. Maybe I'll just go by the panther type, as it'd be a bit confusing to parse otherwise. If you want a chart though, I don't want to make things too cluttered, so maybe just a Performance: Price graph that excludes anything that got a headless panther.

Headless: 0
Shrug: 1
Relax: 2
Golf/Soccer: 3

Yes to both ideas!

 

[tuga]

My first speakers were Wharfedale Diamond IIIs... That was over 30 years ago.

 

[Soniclife]

Measure the speaker, calculate EQ compensation and apply, remeasure.

I don't see the need to measure again, their isn't doubt that EQ does what is expected anechoic, only how it's perceived as an improvement.

 

[Xyrium]

Strange that the utter lack of bass didn't bother you. Is it me or distorsion is worse at 86 than at 96 dB (nasty 5th peak replaced by way less harmful 2nd and 3rd staying the same) ?
I'd really like to see how the comparable Dali Spektor 2 fares.

Good point, though the distortion at both levels seems fairly poor.

Subjectively, it's beginning to dawn on me that some folks might like to hear some additional frequencies outside of the original source, since several speakers with higher than optimal distortion have been reviewed well by several people on the forum.

 

[richard12511]

I have not but it is clear that in most cases EQ significantly improves the performance of these speakers. To some extent we have to decide if no-EQ version even matters!

An important decision, imo, and something I've always wondered myself. I know I bring up the Revel outdoor speaker a lot, but that speaker is a perfect speaker to highlight the with EQ vs without EQ decision. Tonally that speaker was a mess, but it's directivity is some of the best we've seen(even better than the M105/M106 imo). It doesn't surprise me that after EQ that speaker might leapfrog many speakers that are tonally better out of the box. EQ can fix tonality issues, but not directivity issues. If I'm confident that I'll be using EQ, maybe I should be focusing most of my attention on the DI and beamwidth graphs.

 

[MZKM]

This seems like one possible objective approach to determining how well a speaker might respond to EQ. Measure the speaker, calculate EQ compensation and apply, remeasure. Or if remeasuring is not an option, a less accurate approach might be to use the predicted post-EQ frequency response to calculate a post-EQ preference score.

My understanding is that @Maiky76 used this method on the Edifier R1280T to bring its preference score from 1.97 to 5.80 in theory:

https://www.audiosciencereview.com/...280t-powered-speaker-review.16112/post-520312

Unless using a speaker with built-in EQ capabilities or using an amplifier that has the EQ capabilities, the Klippel cannot measure EQ.

There are some issues with an EQ scores as well:
• Not everyone has access to the same degree of freedom, that Edifier EA has 11 bands for instance, so people (say using a MiniDSP) are limited to 10 bands.
• The EQ needs to be realistic, no boosting -10dB dips. Also, the distortion of the speaker ne

Is it me or distorsion is worse at 86 than at 96 dB (nasty 5th peak replaced by way less harmful 2nd and 3rd staying the same) ?

THD is a ratio, so the absolute SPL of that peak at 86dB will be higher at 96dB, but the increase will not be 1:1 and this is a lower %.

 

[weasels]

My gut reaction is that I'd want to see a remeasure after EQ is applied. The typical scientific method is to test a hypothesis - in this case the hypothesis is that EQ is correcting flaws in the original implementation. Measurement is how that hypothesis is proved.

 

[MZKM]

An important decision, imo, and something I've always wondered myself. I know I bring up the Revel outdoor speaker a lot, but that speaker is a perfect speaker to highlight the with EQ vs without EQ decision. Tonally that speaker was a mess, but it's directivity is some of the best we've seen(even better than the M105/M106 imo). It doesn't surprise me that after EQ that speaker might leapfrog many speakers that are tonally better out of the box. EQ can fix tonality issues, but not directivity issues. If I'm confident that I'll be using EQ, maybe I should be focusing most of my attention on the DI and beamwidth graphs.

I have a Tonal Balance chart for recent speakers, which is a weighted average of the on-axis & normalized PIR graph, I can add the DI charts onto that chart. So then you can see both what the estimated tonal balance is as well as dispersion.

 

[infinitesymphony]

@Soniclife & @MZKM

All the better if we could simply calculate a post-EQ preference score using either the Harman target curve or other selectable curves. Presumably there are limits to how much EQ a speaker can take before something else happens, like additional harmonic distortion or changes in directivity (or even phase distortion depending on how the EQ is implemented, right?). Otherwise we would all be EQing our speakers to perform like Genelecs. Would a formula be able to predict these distortion or other volume/balance-dependent factors?

 

[MZKM]

Would a formula be able to predict these distortion or other volume/balance-dependent factors?

There is no way to predict distortion.

Directivity is fixed. I take it you mean perceived dispersion (sound stage).

 

[infinitesymphony]

There is no way to predict distortion.

What if we had several measurement curves (e.g. distortion at different volumes) to interpolate between?

 

[Francis Vaughan]

What if we had several measurement curves (e.g. distortion at different volumes) to interpolate between?

What does that tell us? Toole writes eloquently about the problem. Yes distortion matters, but we still have little to no understanding of what makes for benign versus objectionable versus euphonic distortion.
A metric of level dependence would be interesting but mostly of academic value. Level dependence itself has a time dependence component. Frequency response often has similar issues and they would be useful.

 

[ROOSKIE]

@Soniclife & @MZKM

All the better if we could simply calculate a post-EQ preference score using either the Harman target curve or other selectable curves. Presumably there are limits to how much EQ a speaker can take before something else happens, like additional harmonic distortion or changes in directivity (or even phase distortion depending on how the EQ is implemented, right?). Otherwise we would all be EQing our speakers to perform like Genelecs. Would a formula be able to predict these distortion or other volume/balance-dependent factors?

I almost always subtract (cut peaks) when using PEQ and at most add 1-2db of boost. Of course only being willing to add 0-2db of boost requires either accepting some faults or using speakers that are already so accurate that 0-2db of boost is plenty.
Harmonic distortion should be generally unaffected using these parameters. (in fact may often be better as I usually use a high pass filter and even do this with towers as at a certain point below port tuning the speaker is completely unhinged and useless.

If you are PEQing Amir's measurements you can generally tell where PEQ will be affective with regard to directivity error and phase. In room EQ being used for room compensation is a complex bag of possibilities but I have only had great results so far.

Many speakers can be PEQ'ed to be Genelec's. That is literally all they are doing. You dealing with an active speaker that has PEQ built into the DSP or analog circuits. Genelec does do a lot with cabinet shaping as well but beyond that many speakers can indeed sound incredible with PEQ.
Passive systems usually also have shaping EQ in the crossover circuits as well but due to the design and component parameters usually the shaping is much less than with actives and DSP based designs.

There are very few speakers made that do not already have some presentation of EQ in the speaker off the shelf.

I highly recommend you gat a speaker like the Infinity R162 and PEQ it up. (and use room compensation for the bass/midbass. You will not be unimpressed and with the R162 in particular this site has a handful of filters already published in that thread. (so easy to start there)

My gut reaction is that I'd want to see a remeasure after EQ is applied. The typical scientific method is to test a hypothesis - in this case the hypothesis is that EQ is correcting flaws in the original implementation. Measurement is how that hypothesis is proved.

While would absolutely love if amir did retest a few choice systems with the PEQ applied there is still an option. Since you will ultimately listen in your room and can measure for very little get a decent mic and use REW. Implement the PEQ and measure and listen. Use the waving mic method to create a listening window (or make multiple measures but you need to use a window not a single sho.
I have completely stopped not using PEQ. While I am a less is more guy the little PEQ tweaks are seriously where the end game is now. Forget about paying Genelec and Neuman to do it for you and just get into it. (of course this does require time so no shame if that is not available)

It's a good philosophical question, I'm leaning towards the no EQ impression isn't very important, but we need more evidence that EQ works. Correcting speakers above the transition frequency is tricky, our mono in room mics are not the same as what we hear, if ready made EQ could be downloaded that corrected above the end user may get to optimise the hardest part easily.

I agree that NO PEQ is old energy like records and bookshelf mounted speakers (in actual shelves). *note is okay to enjoy records and speaker sup on shelves but they are not cutting edge hifi, they are fun though *
PEQ is contemporary HIFI plain and simple.
Graphic EQ and Shaping filters in analog networks(crossover EQ) are also useful.
While I agree that a mono mic is not what we hear yet that is what nearly all measurements are taken with in all of HIFI. What can we do? Using a mono mic on many speakers now and creating a couple sets of active DIYers I can say that I have some experience. My mic is the jam and PEQ is the jam, you can mess things up of course and it is a skill to apply it, yet I see noting that can singlehandedly improve a stereo system more.

 

[ROOSKIE]

Unless using a speaker with built-in EQ capabilities or using an amplifier that has the EQ capabilities, the Klippel cannot measure EQ.

There are some issues with an EQ scores as well:
• Not everyone has access to the same degree of freedom, that Edifier EA has 11 bands for instance, so people (say using a MiniDSP) are limited to 10 bands.
.

I don't understand why a minidsp can not be in the Kipple chain. Doesn't make sense to me but I have not seen the set-up.

Speaking of miniDSP, many of the models allow for advanced filters and at they all at least have a 10 band on the input and 10 more on the output so 20 per channel for a 2 channel set-up. I would never use that many though.