Procedures

PRE-CONCEPTIONS (to a subtle and confusing subject)

After diving into listening over several years or even decades, certain patterns emerge, which can be useful in identifying exactly how we should listen to loudspeakers, as opposed to music.

We all know how to listen to music, but how about loudspeakers?

First undoubtedly, we need some wide experience from the different sorts of loudspeakers out there.

Classifying these types, we should be aware by now, that there are different transducers for producing excellent treble, mid range and bass, but be careful - looking purely at tone is naive ! Consider fundamentally that a good loudspeaker should be reproducing the signal into it without distortion not falling into the common trap that a loudspeaker is good because it sounds so!

What if the signal into it (basically recording) is dull, full of turntable crackle and without bass, compared to the original performers - live?

A reverberating, ultra boomy loudspeaker without extended treble reproduction would 'sound' better than a full range reference speaker, right?

I think I have made my point by now. To evaluate a loudspeaker, it is impossible to do it with anything other than a perfectly marvelous recording, that in it's headphone experience is full of realism, drama(stereo staging where the balance swings left/right), and detailed bass and treble plus giving a feeling of intimacy with the vocalist who after all is inches from the mic which is inches from your ear.

Sensible loudspeakers should improve this experience by adding the reverberation of the room into a sort of surround sound feeling, but not losing any detail, realism, stereo precision (imaging)

Remember that most recordings take into account you are going to be playing them through loudspeakers several feet away, so they close-mic' (which also allows the tracks to be separately mixed later).

This close mic'ing is not really helpful to the reproduction, so the engineer usually adds some electronic reverb/delay/presence effects on the desk during mix down. Consequently but arbitrarily, the engineer concerned makes some judgment about YOUR hifi/room combination regarding it's inherent reverberation and tonality, based upon their studio monitors!

I cannot stress this reality enough, and it's importance. In an ideal recording, the musician is singing/playing a non-amplified instrument; the microphones are placed EXACTLY where where the performer is (or pair )

 

UNDERSTAND THE PROBLEMS LOUDSPEAKERS CANNOT SOLVE

The Loudspeakers function is empirically to reproduce the sound recorded at the microphone, so they need to be positioned in your listening room where the mic's would have been during the recording. This is generally impossible, and furthermore pointless, as even if they were, the mic's pick up sound from a point, whereas a loudspeaker radiates sound in a radiation pattern which is not necessarily the radiation pattern of the original source. e.g. a snare drum radiates sound omnidirectionally, but a loudspeaker radiates almost all sound within 90 degrees.

The problem with mic's is that they don't record direction, only quantum, and we all know how mathematically incomplete that information is(vectors have direction and magnitude).

Since we don't have direction information, we cant reproduce it, and the poor old loudspeaker simply splurges the magnitude information over a wide enough angle to include both the listener, and incidentally enough of the room that a reverberrant effect is generated around the listener to fool them that the sound is not just coming from the cones, but is less definable than that and therefore, possibly could be...live??

Hence all this rubbish forces us to concede that all loudspeakers are attempting the impossible, and to reproduce a convincing performance, must use assumption, tricks and the bias of listeners taste in instruments/recording/mixdown recipes to make an effective loudspeaker for them to get the best out of their listening pleasure?

However, today there are many fine recordings, and as decades of reasonable hifi have entered the homes of the masses, recordings are less and less adulterated, and with particularly the advent of hiss-less CDs and subwoofers, many recordings are both full-range and also dramatic. This opens the door for a new type of loudspeaker market, based upon these modern less adulterated recordings.

Does this mean that all old loudspeakers are probably hopeless at reproducing modern recordings, and that all full-range/low reverberation loudspeakers are probably hopeless at making old recordings sound good?

Yes!

Put an early elvis or beatles 45rpm single through CD quality loudspeakers, and then try them through some 6 inch single driver/reflex cabinet made in the era - youll see the difference.

 

So what procedures we invent should suit the criteria we seek to evaluate, and this is not to so much say what is best, but what is what!.

By now you should have accepted then that a) perfect reproduction with 2 loudspeakers is impossible b) tonal distortion, resonances and room effects are often beneficial and not necessarily evil.

 

THE CRITERIA I SUGGEST YOU USE FIRST

So 'mean to suit ends'; we wish to identify, evaluate and describe loudspeakers under the following characteristics;

Realism/Fidelity

Imaging

Tone (frequency range and distortion vs frequency)

Reverberation (incl reflex effects etc)

Efficiency (incl frequency vs efficiency)

Directionality (room effects)

We will see unfortunately that each of these characteristics are affected and involved with all the others, so cannot be taken as a definitive description.

 

EVALUATION PRACTICE

Realism/Fidelity, mainly vocals are what we use to decide if a reproduction is real, but other instruments too. Non-natural/amplified/synthesised/electric sounds are a a poor guide as noone can really remeber what they sounded like in the performance. Amplified instruments like electric guitars are sounding through 10-24 inch cones usually, and that is the instrument (cone) not the few grams of wire vibrating quietly between fret and bridge. to reproduce such sounds effectively you would need such a cone (10-24inch) in your lounge floorstanders - forget it unless you want a P.A system in your lounge and not a hifi, ignore amplified instrument reproduction for lounge hifi. Ask yourself "is this real or is it recorded" and mark it 1-10 between the two extremes.Close your eyes while doing this, this is an absolute must.

Imaging. within the music, instruments should appear as they were, i.e. sound coming from one place both vertically and horizontally constant. Is the Bass drum position is vague or precise, are any pan-effects from the guitars etc perfect or do they seem to travel through tunnels in the room coming out behind you unexpectedly. Do guitars seem crisp or blurred. Do dramatic shifts left/right happen as they were recorded. Do vertical shifts happen as they werent recorded (cant have been). The human ear and brain discern directional sourcing of sounds to suggest if they are real or fake, and thus only a precisely imaged reproduction from a stereo might fool you a recorded performance is live, any multiple source sound or variation in directionality will convince you otherwise.

Tone, at it's simplest would involve playing different (recorded perfectly)tones through the loudspeaker and evaluating the fidelity of each throughout the range. You may note how the close-mic'd 's' and 't' sounds are missing/present, how a drum beat and bass guitar note is blurred together/simultaneous but separate, the guitars sound gutsy/flat or the keyboards sound fantastic/dull. using your memory of natural sounds is helpful here as the human brain is an expert at not only speech recognition but speech realism (i.e. live/reproduced with distortion). The same is true of string and woodwind instruments to a large extent. You can gauge the 'woodiness' of chello, the 'cat-guttiness' of violins, the 'fleshiness' of drum skins and so on. Listening for deep bass is of course only possible with such recordings such as timpanies or bass guitars/bass drums/ possibly lower registers of grand piano etc, but beware the engineer! if it's an unfamiliar recording, you probably ought to listen to the recording in headphones first, or better still use a spectrum analyzer to ensure the bass notes are at a realistic performance level relative to the general performance level. In older recordings, the engineers reduced the low bass to avoid saturating the weak electronic amplifiers of the day, and to eliminate background rumblings present in the recording studios (London underground/passing lorries etc)

Reverberation is more subtle, either by playing a swept sine slowly through, and noting the holes and peaks in the loudness, and where the sound becomes ragged or blurred. Any loudspeaker to a greater or lesser degree has a natural harmonic frequency at which it will start to vibrate and effectively multiply the loudness of that frequency. poorly designed cabinets, floppy cone material (paper) and un-braced construction in the cabinet will lead to worse 'farting' and 'buzzing' than loudspeakers with well vented cabinets, efficient cones, and braced cabinets. Most loudspeakers are designed to 'lose control' and vibrate in the region of 100-300hz roughly equating to where the bass drum finishes and the bass guitar starts - giving a "rocky" loudspeaker "for it's size"; this is more commonly engineered by deliberately inducing resonance by reflex port tuning at this wavelength, at the cost somewhat, of the rest of the range's realism.

Efficiency is even more subtle, as only the original performance would provide an adequate reference of the ratio between loud and quiet decibels at each note, and the loudspeaker may not be reproducing this range in volumes but it would be hard to detect.

Evaluating directionality is tricky unless you are listening in a field. The bouncing of sound off objects and walls is not all bad; this is what normally happens in an auditorium. The listener evaluates the directionality by moving their listening position, noting the level change, and also the change in reproduction observed with different rooms. fortunately the human ear allows us to subconsciously determine direction of origin for some sound, and hence we can with practice in this, distinguish how much sound is emanating directly/from the room indirectly, from a loudspeaker.

 

The most important things to have in your mind while you are evaluating by listening are

(objective) Is the amp, room and positioning giving the loudspeaker the best chance.

(objective)What did it sound like on headphones or the best reference loudspeakers you tried to date

(subjective)Which recordings would it be good/poor at reproducing

In all the above, assume nothing, experiment with everything (in a deliberate and measured way though)

You should at the end of your half day or half minute have evaluated your subject loudspeaker in a useful categorized way.

By the way, don't bother evaluating your own loudspeakers on your own, as they will always come out "the best"!, always get others, especially your competitirs and cynical audiophiles.

 

Finally though, remeber that every genuine breakthrough in history always is accompanied by two things:

    fierce criticism, and

    those that claim it never happened......

 Is the world flat?, did men land on the moon? Has a new type of loudspeker been made that puts all others to shame?

[Resolution] [Product] [Sales] [Advertisements] [Forum] [Reviews]

Please contact our Webmaster with questions or comments.
© Copyright 2004.  All rights reserved.