Volume matching a myth

[mudlark]

It's clear, surely, that when comparing equipment you should reduce the variables to a minimum. When comparing speakers with a different sensitivity a multimeter won't help that much. Anything else up the system then the ouput can be measured as long as you don't short the speaker cables in your enthusiasm to measure things.

Measuring at 1kHz is chosen because the human hearing system is sensitive at that point and the frequency response is reasonably level. Most systems are designed to have a good performance at 1kHz.

 

[sergeauckland]

It's clear, surely, that when comparing equipment you should reduce the variables to a minimum. When comparing speakers with a different sensitivity a multimeter won't help that much. Anything else up the system then the ouput can be measured as long as you don't short the speaker cables in your enthusiasm to measure things.

Measuring at 1kHz is chosen because the human hearing system is sensitive at that point and the frequency response is reasonably level. Most systems are designed to have a good performance at 1kHz.

The problem with comparing loudspeakers is that their frequency response is very variable, compared with electronics. Consequently, level matching on loudspeakers will depend on the frequency it's done at, whereas with electronics it very seldom is. With loudspeakers, it's usual to measure the sensitivity using noise with a frequency response shaped more like music, typically A weighted noise.

However, looking at the anechoic frequency response of a loudspeaker will tell one a lot about how it's likely to sound.

S.

 

[sonddek]

Measuring at 1kHz is chosen because the human hearing system is sensitive at that point and the frequency response is reasonably level. Most systems are designed to have a good performance at 1kHz.

Sure, but many components don't have a perfectly flat response, and rising treble, and/or bass, could make 1kHz level-matched systems vary in apparent volume. What then? In fact, the 1kHz level-match suggests that it's not hard to design kit which seduces the unwary by having a dip around 1kHz.

 

[sergeauckland]

Sure, but many components don't have a perfectly flat response, and rising treble, and/or bass, could make 1kHz level-matched systems vary in apparent volume. What then? In fact, the 1kHz level-match suggests that it's not hard to design kit which seduces the unwary by having a dip around 1kHz.

Cartridges and loudspeakers are the only components I can think of that don't necessarily have a sensibly flat frequency reponse. If an amplifier is so heroically ill conceived as to have a rising treble response, then it's hardly in the HiFi category.

S.

 

[spxy]

OK - but technically it's not possible to level match. For example, when matching a more and less dynamic source, should you level match for the peaks, the rms, or the noise floor? And in which part of the frequency spectrum? Level matching the treble simply isn't the same thing as level-matching the bass on most equipment. What's your taste in level matching? Personally I think the most realistic way to level match is to set the volume by ear to the level at which you want to listen. That might be slightly different for different devices, but at least it's a reasonable approximation to how you will use the gear in the real laboratory of life.

The better gear might make you less inclined to change volume, wherever it's set.

Yes this is a very good point...music is dynamic, and some kit is more dynamic than others

 

[Thighsofsteel]

I put this to the e-meter loving Scientologists (joke): if you need to level match to 0.1 dB using some arbitrary definition of level matching, to distinguish two pieces of kit, then you should probably buy the prettier or cheaper one.

No one is saying that. What we are saying is, that tell a real difference between X and Y you remove as many variables as possible to ensure the differences are real not imagined.

If I was to spend x numbers of pounds on an item I want to be sure I am getting something for my money. That is all

 

[sergeauckland]

Yes this is a very good point...music is dynamic, and some kit is more dynamic than others

Not that I know of. Dynamics comes from the music, not the kit. The kit is linear dynamically so there's no difference to the dynamics between items. Unless you have a different definition of dynamics.

As to level-matching, it can't be done on programme material, it has to be done with tone or steady noise as level matching has to be done closely, or its not worth doing at all.

S.

 

[Pete the Feet]

Not that I know of. Dynamics comes from the music, not the kit. The kit is linear dynamically so there's no difference to the dynamics between items. Unless you have a different definition of dynamics.

As to level-matching, it can't be done on programme material, it has to be done with tone or steady noise as level matching has to be done closely, or its not worth doing at all.

S.

I am in agreement with the sound level comparison as the first thing a salesman usually does is crank it up to impress.
The question is how would you physically do it?
If testing a source, dac or cables, would you send a 1 khz test tone then put your voltmeter across the speaker cables or do you need to be more specific and put it across the individual drivers, and then check each channel for balance?

Sent from my mobile, excuse all typos

 

[Errol]

I'm waiting and waiting for someone to say ....but it has to be unseen! You cannot listen to music without one!
A blindfold that is.......SM
Errol.

 

[Mescalito]

Sure, but many components don't have a perfectly flat response, and rising treble, and/or bass, could make 1kHz level-matched systems vary in apparent volume. What then? In fact, the 1kHz level-match suggests that it's not hard to design kit which seduces the unwary by having a dip around 1kHz.

In which case, they cannot by any stretch of the imagination be called "hi-fi".
At best, they are sound effect generators.

Chris

 

[Steven Toy]

Dynamics indeed comes from the music to be reproduced as realistically as possible via the equipment. No equipment is yet absolutely transparent in terms of dynamics. Some are closer than others though.

A flat frequency response to dynamics is like saying this orange is sufficiently orangy to be an apple.

 

[Darth Vader]

Not that I know of. Dynamics comes from the music, not the kit. The kit is linear dynamically so there's no difference to the dynamics between items. Unless you have a different definition of dynamics.

As to level-matching, it can't be done on programme material, it has to be done with tone or steady noise as level matching has to be done closely, or its not worth doing at all.

S.

Yes the dynamics comes from the music but much kit cannot provide the same thus resulting in a none realistic sound.

It took me years to get the dynamics, sound and timbre of a piano from vinly/CD compared to my joanna in then next room. Its now close.........

Cheers,

DV

PS Sound balancing is a waste of time. If something is better it slaps me in the face. I have often told HiFi salesmen that I don't want to do 'wine tasting' with HiFi. If you have to listen hard to tell a difference then it just doesn't matter.

 

[johncee]

PS Sound balancing is a waste of time. If something is better it slaps me in the face. I have often told HiFi salesmen that I don't want to do 'wine tasting' with HiFi. If you have to listen hard to tell a difference then it just doesn't matter.

Absolutely.

 

[sergeauckland]

Dynamics indeed comes from the music to be reproduced as realistically as possible via the equipment. No equipment is yet absolutely transparent in terms of dynamics. Some are closer than others though.

A flat frequency response to dynamics is like saying this orange is sufficiently orangy to be an apple.

Yes it is. If it wasn't, then that would be picked up as non-linearity, which would be seen as harmonic or intermodulation distortion.

Edit: A flat frequency response isn't necessary to carry dynamics.

S.

 

[Darth Vader]

Yes it is. If it wasn't, then that would be picked up as non-linearity, which would be seen as harmonic or intermodulation distortion.

Edit: A flat frequency response isn't necessary to carry dynamics.

S.

Too true. It also partly explains why some amplifiers that actually measure better in this area can actually sound worse when listening. A good reason not to rely soley on measured specs.

Cheers,

DV

 

[Thighsofsteel]

Too true. It also partly explains why some amplifiers that actually measure better in this area can actually sound worse when listening. A good reason not to rely soley on measured specs.

Cheers,

DV

Or it could be you have heard that particular brand isnt very good and hey presto you dont like it

 

[RJohan]

Not that I know of. Dynamics comes from the music, not the kit. The kit is linear dynamically so there's no difference to the dynamics between items. Unless you have a different definition of dynamics.

Sorry, that is not how it is. The good old American magazine 'High Fidelity' (long gone now) measured how many dB's loudspeakers put out at some different amount's of input power. I think it was 1 w, the loudspeakers rated power and, in short bursts, what it could take without going up in smoke. It was then easy to read these figures and (even if you didn't know the fact beforehand) make the conclusion that different speakers compresses the signal to a differentl amount at high levels.
Lets say we do a level match of two speaker pairs at a rather high level (100 dB) where one of speakers compresses 3 dB more than the other (it can easily be that much of a difference, trust me). Some of the music used for the test is much softer than that, we can asume, and hovers around 80 dB. The speakers that compressed 3 dB mores at 100 dB will most probably compress much less at 80 dB and therefore sound a bit louder than the less compressed ones!
The music has, as we know, ever changing dynamics, it means that we simply can't level match loudspeakers (at least). They all compresses a bit differently at different levels and different frequencies. This is of course one of the major quality differences with loudspeakers!

From my experience, something that sounds bad at a low volume will only sound even worse at higher one.

JohanR

 

[Pete the Feet]

From my experience, something that sounds bad at a low volume will only sound even worse at higher one.
JohanR

Some speakers don't come on song until you crank them up a bit though.

 

[sergeauckland]

Sorry, that is not how it is. The good old American magazine 'High Fidelity' (long gone now) measured how many dB's loudspeakers put out at some different amount's of input power. I think it was 1 w, the loudspeakers rated power and, in short bursts, what it could take without going up in smoke. It was then easy to read these figures and (even if you didn't know the fact beforehand) make the conclusion that different speakers compresses the signal to a differentl amount at high levels.
Lets say we do a level match of two speaker pairs at a rather high level (100 dB) where one of speakers compresses 3 dB more than the other (it can easily be that much of a difference, trust me). Some of the music used for the test is much softer than that, we can asume, and hovers around 80 dB. The speakers that compressed 3 dB mores at 100 dB will most probably compress much less at 80 dB and therefore sound a bit louder than the less compressed ones!
The music has, as we know, ever changing dynamics, it means that we simply can't level match loudspeakers (at least). They all compresses a bit differently at different levels and different frequencies. This is of course one of the major quality differences with loudspeakers!

From my experience, something that sounds bad at a low volume will only sound even worse at higher one.

JohanR

Hang on, once loudspeakers get into compression, they're outside their linear region and shouldn't be used there. It's like an amplifier clipping. Clipping an amplifier takes it outside the linear region, o shouldn;t be used that way.

Note also, that with loudspeakers, it's usually the tweeter than does the compressing, so running a loudspeakerino comrpession not only affects the dynamics, but also the frequency response.

If you want to judge HiFi equipment by what they sound like when abused, that's not how I evaluate HiFi, or anything else for that matter.

S

 

[rubaiyat_23]

It's a question of degree, really.

Every wide-bandwidth, wide-dynamic range system (video, audio, or otherwise) operates within a box where the noise/distortion floor is (of course) at the bottom, and the maximum energy pressure level is at the top.

Use of the "Master Set" method of loudspeaker placement insures that equal pressurization of the room is created by both (stereo) loudspeakers to maximize the usable frequency and dynamic ranges.