Turk 182

measurements - do they really matter that much ?

Recommended Posts

1 hour ago, MartinC said:

Actually that LS50 data does raise an important point: measurement data is only useful if it is accurate, and inaccurate data is worse than no data at all.

For example, earlier this week it took me a while to persuade someone that some small nearfield monitors weren't actually giving a flat output down to 26 Hz no matter what a measured amplitude response might suggest.

As usual for threads of this type we have conflation of specifications and measurements of all types. For myself I have used microphone measurements to significantly improve the sound of my system, by optimising loudspeaker position, subwoofer integration and applying room mode compensating parametric EQ. I flat out could not have achieved results that were as good without measurements.

The problem with any measurements on loudspeakers is they are rarely measured at normal/high listening levels.  The problem then being that most speakers can sound great at low volumes then when you turn the wick up you hear all sorts of problems.  So in many ways the measurements are just an indication but you have to listen at the levels you like your music played at to determine if you like em or not.

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites
Super Dealer
Posted (edited)

Depends upon who is measuring, often speakers are measured at 80,85,90,95 dB with associated distortion.

Keith

Edited by PuritéAudio

Share this post


Link to post
Share on other sites
10 hours ago, Bodgit said:

Measurements are for people with rulers.

Only rulers that can still get straight. No tape measures please :D

Share this post


Link to post
Share on other sites
Posted (edited)
13 hours ago, Turk 182 said:

as the title states measurements do they really matter that much ?

i have read that for example one amplifier may "measure" better than another but may not actually sound better to a certain listener !

personally i'm not to fussed about how something measures as not being technical it means nothing to me anyway...

my approach is simply to plug a new bit of kit in (hardware, cable or accessory) listen over some time and if i like it keep it if not send it back !

(its the method i will use when auditioning kit for a new system i will be getting soon).

i also believe, right or wrong, that not all aspects of a system can currently be measured any way - what are you thoughts ?

This is my opinion:

- Each measurement will quantify a particular aspect of performance. A single measurement in isolation is not able to characterise overal performance.

- A comprehensive set of measurements will provide a reasonable picture of an equipment's performance (how accurately it handles the signal being fed into it). It will say a lot about an equipment's sound to those who know what each one of them means and how it correlates with listening. For those who don't, it won't provide much of a clue as to whether they'll will like its sound, or in other words if it will sound good to them.

- "Measures better" doesn't always equate to "sounds better". The later is related to taste, and taste varies from listener to listener because it is subject to the listener's taste in music and in sonic presentation, as well as his experience/culture of live and reproduced sound, etc.

I am not a designer, but as an end user I find measurements very useful for shortlisting and to help me track down possible causes of problems I have identified through listening assessment which I try do perform from an objectivist perspective (a bit like measuring, but with my ears).
I have been working on finding reasonable correlation between my listening experience and measurements of certain qualities I dislike in speakers and, together with previous listening experience with a particular brand, driver material, topology, or technology, I believe that I am not missing out on the speakers that don't make it into the list. There are only so many speakers one can listen to anyway.
For both purposes, shortlisting and identifying problems, I find measurements a lot more informative and useful than magazine reviews and user reports. Experience has taught me that it's far more effective to rely on measurements than on reviews when deciding what's worth listening to.
 
For me, listening and measurements are both indispensable and complementary.
Equipment must sound right and it must perform well in the test bench.
Usually equipment that measures well over a comprehensive set of measurements sounds right to me. If it doesn't then there must be something that the measurements aren't picking up, which is why I listen (to pick up those things).
 
I do make my own measurements but only to assist with the positioning of loudspeakers and listening spot, and more recently for EQ'ing of the sub-100Hz range.
I did mess very briefly and superficialy (due to limited knowledge and funds) with a pair of two-way horn + bin box speakers several years ago and whilst I've since expanded my theoretical knowledge it is still restricted...and theoretical.
 
I've given up on reviews and reports long ago so, when I am shortlisting speakers, if things that I can potentially dislike show up in the measurements I tend to give those speakers a pass. I also use impressions from listening experience of particular technologies or topologies or brands as another source of information. This is why I have been trying to correlate my listening experiences with available measurements as best as I possible can.
 
There are many online sources of measurements (Stereophile, Soundstage, AVHub, Hi-Fi News in English, Hi-Fi, Stereoplay, Audio, in German, Audio Review in Italian, Revue du Son and Stereo Prestige Image in French, Audio in Polish, etc.).
I try to compare the results from different sources because some of them don't use the best methods and not all perform the same particular set of measurements. But a few do produce quite consistent and reliable results. One must be aware of potential traps, such as Stereophile's anechoic response which is not measured on-axis but over a very wide listening window and it's definitely not anechoic and it produces a 6dB artifact in the lower bass (though to be fair this is explained in a piece called Measuring Loudspeakers).
 
Finally, it is just as important to know at least some of the basics of how equipment works, and also a bit about the acoustics (how speaker and listening spot positioning affect the end result).
 
I can only say this approach has been far more fruitful than the more tradional subjectivist way of trial-and-error with no real sense of control over one's choices.
Edited by tuga
  • Like 2
  • Upvote 1

Share this post


Link to post
Share on other sites
14 minutes ago, uzzy said:

The problem with any measurements on loudspeakers is they are rarely measured at normal/high listening levels.  The problem then being that most speakers can sound great at low volumes then when you turn the wick up you hear all sorts of problems.  So in many ways the measurements are just an indication but you have to listen at the levels you like your music played at to determine if you like em or not.

It depends what is trying to be measured how significant level is in my experience. For an amplitude and phase response it makes virtually no difference for example. Distortion figures vary much more obviously though.

When I make my own measurements I can obviously choose to use whatever level I want.

Personally I've never experienced any really obvious problems that were only apparent at higher levels. I think by far the biggest variations in sound quality with volume are the result of how our hearing sensitivity varies.

Share this post


Link to post
Share on other sites

I often wonder if Measurists, when dating, asked their dates/potential girlfriends/potential wives if they could measure their willy and then make a judgement if its good enough for the job based solely on the tape measure. Bet they didn't. 

Share this post


Link to post
Share on other sites
1 minute ago, tuga said:

I can only say this approach has been far more fruitful than the more tradional subjectivist way of trial-and-error with no real sense of control over one's choices.

I chose the Gale 401s 42 years ago because they sounded great and I never realised they measured well until after I bought them. I then bought an amp 40 years ago because it sounded great and drove loudspeakers that dropped below 4ohms and had a healthy output (namely the Gales).  As it turned out it measures great too but I never knew that when I chose it.  I got a pair of Rogers Pro 9TL for £50 and they turfed out the Gales, buggered if I know how they measured but I guess they might have been lumpy down the very bottom end but god they were fun.  I could go on but I won't .

I think it is horses for courses - I trust my ears and as it turned out they must have measured quite well (as the equipment measurements after purchase were not disappointing).  If however, you wish to use measurements as a benchmark and to help with your choices that is fine.

At then end of the day I hope none of us (whichever process we use) will end up with something they do not enjoy listening to music through, as at the end of the day it is enjoying the music that counts.

Share this post


Link to post
Share on other sites
29 minutes ago, uzzy said:

The problem with any measurements on loudspeakers is they are rarely measured at normal/high listening levels.  The problem then being that most speakers can sound great at low volumes then when you turn the wick up you hear all sorts of problems.  So in many ways the measurements are just an indication but you have to listen at the levels you like your music played at to determine if you like em or not.

Turning the wick up increases distortion and affects the balance. Speakers which sound good at very low levels will exaggerate frequency extremes (like listening at loud volume with the loudness circuit activated).

Soundstage often measures distortion at different SPLs:

thd_90db.gif

THD+N @ 90dB, 50Hz - 10kHz (measured @ 2m)

.

thd_95db.gif

THD+N @ 95dB, 50Hz - 10kHz (measured @ 2m)

.

linearity_90db.gif

Deviation from linearity - Difference @ 90dB, 50Hz - 20kHz (measured @ 2m)

.

linearity_95db.gif

Deviation from linearity - Difference @ 95dB, 50Hz - 20kHz (measured @ 2m)

  • Like 2

Share this post


Link to post
Share on other sites
1 minute ago, uzzy said:

I chose the Gale 401s 42 years ago because they sounded great and I never realised they measured well until after I bought them. I then bought an amp 40 years ago because it sounded great and drove loudspeakers that dropped below 4ohms and had a healthy output (namely the Gales).  As it turned out it measures great too but I never knew that when I chose it.  I got a pair of Rogers Pro 9TL for £50 and they turfed out the Gales, buggered if I know how they measured but I guess they might have been lumpy down the very bottom end but god they were fun.  I could go on but I won't .

I think it is horses for courses - I trust my ears and as it turned out they must have measured quite well (as the equipment measurements after purchase were not disappointing).  If however, you wish to use measurements as a benchmark and to help with your choices that is fine.

At then end of the day I hope none of us (whichever process we use) will end up with something they do not enjoy listening to music through, as at the end of the day it is enjoying the music that counts.

You are a man of impeccable taste. :^

  • Haha 1

Share this post


Link to post
Share on other sites
10 minutes ago, MartinC said:

Personally I've never experienced any really obvious problems that were only apparent at higher levels. I think by far the biggest variations in sound quality with volume are the result of how our hearing sensitivity varies.

I beg to differ - I have heard loads of loudspeakers that sound great at low volumes but when you turn the wick up start to sound hard and display sonics that my ears do not like.  Someone remarked in another post about the JBLs he has - where they sound as easy on the ear at low volumes as they do at high volumes.  

That is what my ears have told me but I do appreciate others may have a different view.  

Share this post


Link to post
Share on other sites
Just now, uzzy said:

I beg to differ - I have heard loads of loudspeakers that sound great at low volumes but when you turn the wick up start to sound hard and display sonics that my ears do not like.  Someone remarked in another post about the JBLs he has - where they sound as easy on the ear at low volumes as they do at high volumes.  

That is what my ears have told me but I do appreciate others may have a different view.  

I didn't say the sound doesn't change. Rather than the variability in hearing response is at the very least a major reason for this.

To prove my point about amplitude and phase responses not changing much here are some examples from my sub that I was able to find quickly:

1642190951_SubcompressiontestAmplitudeResponse.jpg.3c0dd0d0b031dc92069c3177f2a4c138.jpg

1657011505_SubcompressiontestPhaseResponse.jpg.b0b56c6e213166415ce07e849e930bc2.jpg

Share this post


Link to post
Share on other sites
Just now, MartinC said:

I didn't say the sound doesn't change. Rather than the variability in hearing response is at the very least a major reason for this.

I think Tuga's explanation kind of confirmed what I said about changes with volume - I rest my case but it is just my opinion using the measurement tools on each side of my head  :D 

Share this post


Link to post
Share on other sites

The ear is less sensitive to frequency extremes at low levels:

800px-Lindos1.svg.png

https://en.wikipedia.org/wiki/Equal-loudness_contour

  • Upvote 1

Share this post


Link to post
Share on other sites
1 minute ago, uzzy said:

I think Tuga's explanation kind of confirmed what I said about changes with volume - I rest my case but it is just my opinion using the measurement tools on each side of my head  :D 

Errr, what Tuga and I have said are completely in agreement.

Share this post


Link to post
Share on other sites
Posted (edited)
19 minutes ago, tuga said:

Turning the wick up increases distortion and affects the balance. Speakers which sound good at very low levels will exaggerate frequency extremes (like listening at loud volume with the loudness circuit activated).

4 minutes ago, MartinC said:

Errr, what Tuga and I have said are completely in agreement.

Ermmm - if yuo say so :D  I thought you were disagreeing with me - seems you agree with me then lol 

Edited by uzzy

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.