Objectivist or Subjectivist? Give Me a Break

Tim,

I wish I knew it in terms I could describe objectively;) in terms that would please you. Perhaps than I would retire and start a loudspeaker business.

But when I listen to the XLF I know they sound a lot better than my old JMLab Grande Utopia or the MAXX3, to refer to speakers with comparable bass and size. And the difference is not due to FR. I have told this before - I am a good friend of the first Dynaudio distributor in my country. At that time Dynaudio had excellent (passive, sorry) kits. We built two pair using their accurate plans. One in Corian the other in MDF. They sounded completely different.

No need to please me, micro. They sound better to you. That's good and that's what matters. It is more important than frequency response. It might even be frequency response. And even that doesn't matter if you like it better.

Tim
 
Everything was exactly the same except the materials of the box - one was MDF the other Corian. Even the internal frames were of the same size. No screws were used in both - only special glues.

Hello Micro

Did you measure them or drop an accelerometer on the cabinets to see why??

Rob
 
I tend to believe that cabinet material has to do with the overall sound of a speaker system .. This said , I also think it would translate in FR variations as well ... Did you measure both speakers and found the FR to be the same?
 
I tend to believe that cabinet material has to do with the overall sound of a speaker system .. This said , I also think it would translate in FR variations as well ... Did you measure both speakers and found the FR to be the same?

No, thirty years ago I did not have access to anything other than a Fluke voltmeter and an oscilloscope. :) But when we taped them they sounded very different.

But the use of different materials was abundantly referred in articles, even by Dynaudio. The effects in FR were said to be minimal, as the energy emitted was very low. However they could show sometimes in the impedance plots. It was reported that ear is particularly sensitive these changes in sound, as the cabinet contribution was mostly due to delayed resonances. All this AFAIR.
 
Hello Micro

However they could show sometimes in the impedance plots.

You had a resonance if it showed up there. Depending on the Q and amplitude it could very well be audible.

Rob
 
Cabinet resonances have been discussed for years, to the point that Stereophile actually measures the cabinet's output. I am surprised there was a large difference, but if the materials were different thicknesses as well (thin resonates more) it's certainly possible. I have also seen FR changes because builders kept the exterior dimensions the same and ended up with slightly different internal volume, and due to differences is assembly and bracing. Cabinets may vibrate (resonate) at certain frequencies and that can be audible; the amount and frequency depends upon the cabinet material and volume.

My 0.000001 cents (microcent)
 
Let's not be evasive, micro :), what matters more than frequency response?

Tim
It's not so much what maters what more than frequency response. Certainly distortion , if we are talking about speakers remains a serious problem. Probably why most marketing departments want to ignore it. Chosing a set osf parameters present the following dllema. Once you optimize one measurement then the others become more improtant. Is ruler flat frequency response to be puirsued at all cost. Or is it even necessary? Certainly many manufacutrers fall on bith sides of that equation.
 
The venerable LS3/5A speakers came from the R&D department of BBC where they studied & measured the effects of the resonance of the cabinet. Early research models showed a better performance from a lossy type resonate where the back panel was screwed rather than glued in place but the production models were mostly glued for reasons of consistency. They also studied (& measured) the effects of the composition of the internal struts finally settling for Solid beach braces & beech ply for the cabinets. The crossover is also one of the most studied & measured of it's kind at the time. Interesting to look at these studies now!
 
I wish I knew it in terms I could describe objectively;) in terms that would please you. Perhaps than I would retire and start a loudspeaker business.

But when I listen to the XLF I know they sound a lot better than my old JMLab Grande Utopia or the MAXX3, to refer to speakers with comparable bass and size. And the difference is not due to FR. I have told this before - I am a good friend of the first Dynaudio distributor in my country. At that time Dynaudio had excellent (passive, sorry) kits. We built two pair using their accurate plans. One in Corian the other in MDF. They sounded completely different.

I'll have a go : energy storage. Stored energy results in colourations which may only affect FR imperceptibly.
 
Why do you insist manipulating my words? My words are "there are somethings that matter more than just the simple on axis FR.".
This does not mean "has nothing to do".

BTW, I should have added I was focusing mainly on FR above the transition zone - it is the most important zone for speaker evaluation, in room bass response is mainly dominated by room.

No, not manipulating your words, sorry.

You are clear, at least to me, that these 'other' things are more important than simple FR. That is it in a nutshell, more important. Yet, when asked to be defined, this thing that is MORE important somehow evades description? We never seem to find out just what IT is, just that it is there somehow, and more important than FR.

If it overrides all else, I cannot grasp how it has never yet been identified. Just a small conceptual problem I have with the notion, that's all.

In any case, you seem m to have missed the implications of what has followed the post you quoted, some further implications (unconsciously) raised by yourself.

Let's go back to your basic sticking point, that simple FR does not seem to correlate with your perception (therefore justifying 'throwing out' FR as having great importance), you missed the bit about the bibles measurements being (perhaps) of little real world significance. Yes, I saw the bit about above the transition zone but I fear that still does not cover the whole scene.

Before I address that, any thoughts about how misleading the bibles measurements might be, about how those (poor-if true, ie they do not in any real way reflect what people might reasonably hear) measurements themselves contribute to this stark divide in audio?? That because those bible measurements are 'essentially useless' not only are they 'worthless' but themselves cause or at least strengthen this underlying concept of measurements having no value?

Is that not a bit interesting as a topic of discussion even on a 'what if' level? Put another way, what if you received 'proper' measurements that now DO coincide with perception, would the thoughts on the value of measurements change in the audiophile world?

Perhaps this is at the bottom of the divide, those of us 'diyers' if we can use that label (cause we do get in and measure) that HAVE found the measurements that correlate with perception (for us), no wonder we are bemused when people say measurements are worthless or not very helpful. That the measurements those statements are based on ARE worthless goes a long way in turning the light on.

Anyway, back to the transition zone. When we look at artificial divides like that we miss an extremely important point. Basically, it does not work that way. There is no possible way you can seperate the entire package, and evaluate the region above the transition zone 'as if it stood alone'. Whatever happens below the transition zone impacts and effects dramatically the perception above it.

You can't separate the two. Seriously, remove a bass hump (say, just for illustration) and you do NOT hear 'just the bass hump gone'. Yes, of course it has gone and yes the bass does change....but so does the perception of the rest of the frequency range.

I can't stress that enough, we have not done a thing to anything above the bass region yet it sounds very different. It can (and is) a very dramatic demonstration to someone who does not know that, just turn off the room correction (and just note in passing the bass changes) but listen to the rest of the spectrum. It too changes dramatically.

You could look at it a few ways I guess, recognise that the system is a package and find it, whilst dramatic, not that very mysterious a phenomenon.

Or, 'seeing that the FR above the transition zone' did not change, yet the perception did, start to feel there are undiscovered measurements yet to be found by science.

I rather think that instead I found out more about my own perceptions rather than a new thing to be discovered by science.:)



Now, rather than get upset by thinking I twist your words, can you have a look at the way you phrase things and see if that contributes even a little to the problem?

We built two pair using their accurate plans. One in Corian the other in MDF. They sounded completely different.

Back to the completely different again. Seriously, what does that mean? Completely different? The same as two dac chips out of the same batch? That sort of completely different? Or the completely different as a sandwich made of ham vs a sandwich made of peanut butter. That sort of completely different?

We used to gag about with my girls when they were young, they'd talk about something that was exactly the same but completely different. So it became a standard...'they were exactly the same yet completely different' Lotsa fun.

Anyway, can you at least acknowledge that that sort of rhetoric at the very least confuses the issue? I doubt the majority of people would describe your scenario as producing 'completely different'.

I get and accept that you heard a difference, and I get and accept that it was real. In other words, it is not the usual 'you must have been imagining things or subject to some sort of uncontrolled bias'. I also accept that those differences might not show up on a simple on axis FR measurement (even tho as you admit you have no way of actually checking that).

But, here is the but...always one eh! Your claim is that this difference is MORE important than FR. How do you know?? You seem to be conflating 'having heard a difference' with *us* saying there is no difference if FR is the same. A simple Fr only measures that, FR. It does not measure other things. I mean obvious but hey, it needed to be said.

So here is the thing...change very slightly (1db dip from 1-5k say) the Fr of one...now compare.

Did you do that? No I don't think so.

See, your statement that this 'thing' is MORE important than FR can not be substantiated without experiments such as the above. Alternatively, have the FR change on a dial, move it till the differences equals the difference the 'thing' makes. Then you have quantified the two.

But none of this happened, all that happened was you heard a difference. THAT is what defined your stance, and it seems to me to be fighting a different proposition than the one we give. You seem to think that we say 'FR is all and everything', so if you heard a difference (and you are assuming the FR is the same anyway, we will never know) you have disproved the ''FR is all and everything'.

Well we don't say that! You are disproving something that is not claimed.

We might (and probably do) say that 'FR is the most important single parameter'..something I doubt Sean would disagree with....yet your anecdote above did nothing to show one way or other if that is true. You did not deliberately introduce a small change in FR to judge it's audibility, yet because you did hear a difference you feel it justified to throw out the false hypothesis. We don't say 'FR is all and everything'.

Compare two pairs speakers, one with the 'thing' yet 'identical FR', the other without the 'thing' yet measurably different Fr.

Which set produces a greater 'completely different'??

That is how you prove or disprove your claim of 'greater importance than FR'.
 
I'll have a go : energy storage. Stored energy results in colourations which may only affect FR imperceptibly.

What is stored is eventually released. My DIY friends have demonstrated this with different types of materials on OB designs of the same size, shape and driver placement. It is audible and measurable. What is preferred is a different matter.
 
One source of confusion, I think, is the audio world's love for frequency domain measurements while tending to ignore the time domain. This results in measurements like THD of steady state signals being extrapolated into a measure of a system's overall accuracy. It results in ideas like "phase doesn't matter", ignoring the effect that less-than accurate phase response has on transients. Another example is the idea that, as a room has a time domain impulse response from which a form of frequency response can be derived, people conclude that they can 'correct' their systems by applying frequency domain EQ. It leaves room for all sorts of conjecture and debate in the gaps that remain e.g. is the reason why one amplifier sounds better than another because of instantaneous thermal effects on the output devices that show up on transients in real music that conventional THD/IMD measurements cannot capture? Does feedback cause instability on transients in real music that just does not show up with test tones?

I wonder if this is partly historical: the most obvious measure of a system's accuracy would be to simply subtract the input from the output and see what remains, but at a very early stage in audio evolution this would be found to be impossible. Using the frequency domain would be an obvious way of doing sensitive measurements that could accommodate inevitable delays and phase shifts, and so these factors eventually came to be ignored by many people.

Are we at a stage where the endless debate could be resolved by aiming for systems that were truly accurate in both the time and frequency domains? We could consign the "phase doesn't matter" argument to history by pre-processing the signal digitally to correct for the system's phase shifts. Likewise in speakers we can correct for impulse response, group delay etc. in anechoic conditions. (I am less convinced by digital room correction). We could get to the stage where we really can subtract input from output to see what remains. It would be fascinating to see a system that behaved perfectly on test signals fall apart when presented with real music - as the audiophiles might have claimed previously but were never able to prove.
 
One source of confusion, I think, is the audio world's love for frequency domain measurements while tending to ignore the time domain.

In addition to all the usual prefaces (IMO, IME, 2c etc etc) I must stress that my thoughts on this relate to speakers only, have not the foggiest if any of this applies to electronics (amps and such).

So with that caveat, and only from my experience, funnily enough any time domain do show up in FR. Some (not me, not good enough) might be able to recognise them from the FR? All I can do is change the time and see the corresponding change in FR. Depending on frequency it can vary dramatically (lots of change in the treble for small amounts, need much more in the bass to see a change)

It results in ideas like "phase doesn't matter", ignoring the effect that less-than accurate phase response has on transients.

Phase (I think) is a variation of timing...the above was mainly to do with 'time alignment', when the signals arrive from each driver. One concept about correcting phase which I can think with (and is true in my experience) is that phase in addition to affecting transients (all in phase does have more of a jump factor, cause 'everything arrives simultaneously') has an awful lot to do with the ambiance, the space encoded on the recording. The better the phase the greater the space.

Another example is the idea that, as a room has a time domain impulse response from which a form of frequency response can be derived, people conclude that they can 'correct' their systems by applying frequency domain EQ.

AFAICT, the maths/science on that is reasonably rock solid....that eq DOES work on bass/room induced anomalies, as long as that anomaly is minimum phase. That also leaves space for it having no effect on non minimum phase problems. I hope those with better knowledge of that area can amplify or correct me.

Leaving that aside, there is also a lot of debate on how much phase is audible in the bass. Most ignore it there, 'most' data says it does not matter, but in my experience I do find a worthwhile improvement. Having said that, it does seem to be an outlier from what I have read. Still, if it does not make an audible difference and it is in my imagination, then the reverse also applies. It can't hurt it and as long as I think it's better then no harm, no foul.:D

Are we at a stage where the endless debate could be resolved by aiming for systems that were truly accurate in both the time and frequency domains? We could consign the "phase doesn't matter" argument to history by pre-processing the signal digitally to correct for the system's phase shifts. Likewise in speakers we can correct for impulse response, group delay etc. in anechoic conditions. (I am less convinced by digital room correction). We could get to the stage where we really can subtract input from output to see what remains. It would be fascinating to see a system that behaved perfectly on test signals fall apart when presented with real music - as the audiophiles might have claimed previously but were never able to prove.

I'm not sure that the phrase 'phase does not matter' is accepted across the boards?? Is it?

In any case, it is not too hard (as I have done) to at least demonstrate the effect of phase correction has (on speakers...as I said I have no idea about 'electronics'). Each is then able to make up their own mind about it's benefits or not.

Re digital room correction, is that from experience? or from 'theory'?
 
Yep - so energy storage in a system results in smearing and we hear smearing as colouration. Its easily seen in time domain measurements - waterfall (or CSD) plots. Not at all easy to spot on FR plots.

Yessir! Agree 100%
 
Re digital room correction, is that from experience? or from 'theory'?

Mainly theory (or at least my own sort-of theories!). I have dabbled with room correction but never got anywhere with it - maybe a lack of persistence or poor technique. But my experience is that I can 'hear through' the room to the original signal pretty well in most circumstances, and I could imagine that this is just a natural ability that humans have - it certainly seem effortless. The 'theory' part is that a partial time domain correction or a half baked frequency domain correction may be the worst of all worlds, upsetting our natural ability to filter out the room while corrupting the signal itself.
 
Raffles

I tend to agree with you that we may have to look for other measurements and/or measurements protocols. I also believe like you that the emphasis on steady state signal is not warranted by the transients-based nature of music. Much need to be done in this area. I hear differences between amps and those are hard to put in FR terms for example...
I do not agree with you on DRC though. I have recently started to pay attention to DRC. It requires a good handle on the DRC system used and truly of acoustics.. You have to understand your room,its limitation, the capabilities of the DRC itself and its limitations.. Steep learning curves.. DRC is not a cure-all and definitely but absolutely NOT plug-and-play. In the wrong hands DRC easily become plug-and-run away :)
 

Terryj,

5645 characters including spaces - congratulations you won by exhaustion of the opponent.

My sentence was short - and it must be considered within reasonable variations of FR. I present my experience and opinions with sometimes too short statements thinking mainly in those who are interested in reading them. I do not shield them against possible extreme interpretations to attack them or syllogistic attacks.

Also we should think about those who are always so happy when they find some substance to feed the "audiophile hyperbole" argument, using my sometimes extreme wording. Why taking away this pleasure from them? :)
 
Last edited:
Yep - so energy storage in a system results in smearing and we hear smearing as coloration. Its easily seen in time domain measurements - waterfall (or CSD) plots. Not at all easy to spot on FR plots.

Hello Opus11


Yes but part of the problem with these measurements is that if the waterfall isn't quite as clean can you hear it?? There is difficulty determining just how audible a measured artifact is just by looking at either an ETC or Waterfall. The emphasis of frequency response is simply there because it is easily measured and there is very good correlation with how audible changes are.

From my own experience I now use aquaplas damped beryllium and aluminum diaphragms in my compression drivers. I started with undamped titanium and when I switched over to damped titanium it's clearly audible. It's a time domain change and very difficult to measure, where it shows up is in an ETC plot and you need a anechoic chamber to see it.

In a home measurement system there is too much noise to see it. So you can run your conventional FR measurements all day tyring to find it and never see what can be plainly heard. You have to look in the right place and have measurement capability that most manufacturers don't.

As far as phase not mattering I laugh every time I see that. Go flipping the leads on your drivers if you think phase doesn't matter. We are much less sensitive to phase changes unless they are gross. If they are gross they effect the Frequency Domain where we are more sensitive and that's how we hear them.

Rob:)
 
Last edited:
Teryj,

5645 characters including spaces - congratulations you won by exhaustion of the opponent.

so exhausted you need to leave letters out if you persist with 'forum monikers'?

You entire rebuttal consisted of counting characters (including spaces)??

that's it?

Now, count up the posted spelling errors (few I hope) and you start to get an idea of how much bloody work goes into my posts!! Let me assure you, there are plenty of typos that I have to correct.

My sentence was short - and it must be considered within reasonable variations of FR. I present my experience and opinions with sometimes too short statements thinking mainly in those who are interested in reading them. I do not shield them against possible extreme interpretations to attack them or syllogistic attacks.

Spare me.

please.

Are you capable of answering questions? You seem to constantly throw the challenges out, I have yet to get a straight response to anything.

You are the one who said 'they sounded completely different', and you take ME to task for trying to get a handle on what you mean? Give me a break.

Reasonable variations of FR? You said they (dynaudio) assured you they 'were the same'. Just how much wriggle room do you claim for your statements?

No, you don't actually simply present your experience...you continually go on to claim science needs to find new measurements. The existing ones don't cut the mustard. You certainty is based on fleeting anecdotal 'evidence' from years ago, when even you admit that you have no way of even confirming the basic starting point, that the two indeed have identical FR responses.

You ask *us* to be more scientific, to embrace the notion that there are things currently not understood, so let's try this in reverse. Go back to your experience that 'proves' there are things more important than FR...be scientific (as you wish *us* to be)...give us your own estimate of the error bars inherent in your experience.

Tell us, do you think you could hear the difference between corian and mdf if the Fr was NOT the same? If so, why so. If not, why not. Have the intellectual capacity to critique your own statements for validity, tell us where it on safe ground, tell us where you think it might be on shaky grounds.



Rob, hope we are not confusing phase with polarity?

Go flipping the leads on your drivers if you think phase doesn't matter
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing