• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

What measurements number in the dashboard leads to a "warm and intimate" sound vs "colder more technical/clinical" sound?

confucius_zero

Addicted to Fun and Learning
Joined
Nov 22, 2018
Messages
541
Likes
345
It is true that higher sinad lead to something more separated more defined and true to the source. However, I wonder which measurement data can define the "personality" of an audio device when it comes to warm vs cold? The frequency response?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,492
Location
Seattle Area
The frequency response and output impedance in the case of headphone amplifier. Even then the direct connection to those made up audiophile terms is not great.
 

PaulD

Senior Member
Joined
Jul 22, 2018
Messages
453
Likes
1,341
Location
Other
It is true that higher sinad lead to something more separated more defined and true to the source. However, I wonder which measurement data can define the "personality" of an audio device when it comes to warm vs cold? The frequency response?
I would suggest that the terms "warm" and "cold" have more to do with other perceptions than the sound, and are a result of cognitive bias.

Several years ago I changed the LED on the front of an amplifier from blue to yellow. Several people told me the amplifier sounded "warmer", and I'm sure they genuinely thought that. It sounded exactly the same of course, but the perception of "sound personality" probably has little to do with accuracy...
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
It is true that higher sinad lead to something more separated more defined and true to the source. However, I wonder which measurement data can define the "personality" of an audio device when it comes to warm vs cold? The frequency response?

voice-color.png


Warm/warmth – a general downward tilt in the frequency range between 300Hz and 3kHz. The opposite is cold.

Cold – a general upward tilt in the frequency range between 300Hz and 3kHz. The opposite is warmth.

Warm or cold is usually describing a general tilt of the frequency response in the frequency response shown above.
It does not need to be sloping in a flat line as in the example. It gives more or less a 'general direction'.
The angle of the tilt determines the amount of 'warmth/cold'.
Amplifiers and DACs do not sound warm or cold. Such is a subjective found idea that is not demonstrable in level matched non-sighted tests.

That said. As Amir already mentioned. In some cases a higher output impedance may well skew the frequency response towards a darker (or muddier) sound.
 
Last edited:

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
Tie the definitions down first.
Nerd.png
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
So how will this curve sound? Note the green line is the actual result. Warm, cool, or how?
1573020094963.png

Looks less radical in linear scale.
1573020153788.png
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
There are more descriptors than warm/cool so it does not have to fall under warm/cool... pick one :D
The most fun part is that a lot of people tend to use descriptors below to describe the sound of DACs and amplifiers while objectively these all are 'flat' or 'rolled off'.

Airy – Describes the sense of space and openness of the music. Best heard in cymbals where the shimmer is realistic. This mostly has a relation with treble extension. It should be mentioned that not all well extended headphones sound 'airy' and some not so well extended headphones can still sound sort of 'airy'. The superlative (exaggerated) form af 'airy' sound is 'ethereal' where the sound feels very 'fragile' and overly 'delicate'.

AnalyticalA term used to describe a product that produces a high level of detail in reference to the music being played back. In technical terms this means the tonal response is between 'flat' and slightly elevated above 3kHz.

Balance – This could mean balance between channels or sometimes people use the term to describe Tonal Balance.
Balance between channels means the stereo image is right in the middle. In some cases balance can be correct for a certain part of the frequency range and not for higher or lower frequencies. This can be spotted in measurements. In sound it can lead to a fuzzy position of a stereo image.
Also one sided seal loss with headphones can cause this when bass is less prominent in the leaky side. On measurements with perfect seal this won't be obvious.

Bassy – This is the lower end frequency of the human hearing. Bass can be measured in quantity (heaviness) and quality. Other bass descriptors are muddy, punchy, rubbery, boomy and fat.
The bass area is between 25Hz and 300Hz but harmonics can go as high as 5kHz. Peaks and dips in the frequency range between 20Hz and 500Hz can make the bass sound different from realistic. For descriptors and the frequency range they fall into see the chart above.
The opposite of bassy is Lean

Bass (Lows) – can be divided 4 ranges: sub-bass (10Hz-25Hz) this is not audible as tones or notes but deep rumbles in movies are an example. When it is absent most people won't miss it. That is once you heard what bass extension does it can add to the musical experience. For instance with Piano music where pedals are used this lowest part can make the piano sound more real.
Lower bass (25Hz-50Hz) is the part of the bass where the deepest musical notes are. A small and gentle roll-off usually isn't very audible. A substantial roll-off (-10dB) is quite audible.
Bass (50Hz-120Hz) is the part where most of the 'body' of bass instruments is. Sharp dips or peaks and emphasis or subdued parts here can color the bass, make it sound boomy, exaggerated, full, tight, dis-attached or muddy. The best sounding headphones/speakers have a flat response in the area. Some folks prefer substantially more bass than others. When traveling a bit more bass usually is welcome. High-bass (120Hz-250Hz) is where the harmonics of most bass sound producing instruments are. A small peak there may make bass sound tighter. Too much (>5dB) can make bass sound muddy and boomy. A dip in this area can make bass instruments sound 'wrong' or 'disattached'. This is where bass appears to be 'apart' from the music. Some people actually prefer a dip, certainly when the lower frequencies are elevated.

Bloat – Bloat is present usually in the mid bass. When frequencies below 400Hz are lifted opposite the mids and highs the bass can sound 'bloated' and overdone. When there is a dip around 150Hz this is the opposite of bloated and the bass sounds disattached.
Bloated bass often also causes 'bloom' in mens voices. In this case the male voice sounds unnaturally bassy/full and lacks clarity.
When only the part below 150Hz is boosted the bass doesn't sound 'bloated'.

Bright/Brightness – A small emphasis of upper mids. The frequency band between 1kHz and 7kHz. Brightness is a wider bandwidth form of Clarity + Detail, when exaggerated even more it becomes Shrill.

Clarity – A few dB emphasis between 1kHz and 3kHz can add some clarity for voices and instruments. The opposite (a dip) of clarity (seen a lot in headphones) could be described as Laid back, Sweet, lacking Attack. An exaggerated form is shrill, cold, hard.

Coarse – is another word for Grainy or Gritty and describes uneven treble response with deep dips and peaks. Note that measurements made with measurement gear using a fake Pinna can show substantial peaks and dips above 5kHz caused by the fake Pinna but may not indicate roughness nor should one try to EQ this out.

Cold – a general upward tilt in the frequency range between 300Hz and 3kHz. The opposite is warmth.

Congestion – Something sounds congested when it the sound is muffled and not very dynamic. It usually is a combination of emphasis in warmth and laid back sound / lacking in clarity.

Compressed – This can mean 2 things. There is compressed data and compressed music. Compressed data means that a file has been shrunk in size. Most people have heard of ZIP and RAR where one or more files are smaller in size so they can be sent as an attachment. This is lossless compression where the file can be reconstructed to its original size and then can be used. In Audio there are many lossless formats. The biggest difference with something like ZIP is that a ZIP file needs to be 'unpacked' before one can use the file (song or album) where audio files are de-compressed while the song is being played back. One can also compress files by leaving out data which can not be recovered later. The most well known lossy compression files are JPG (for pictures) and MP3 for music but there are many more formats. During the compression (encoding) one can usually select how much one throws away. More compression = more loss of fidelity.
When music is compressed the louder parts in a recording and lowered in volume and/or softer parts of a recording are amplified. Usually this is done in studios to get a more even sound. There is a fine line between this being done craft-fully (usually for each instrument its own compression) and it being overdone in dramatic ways. This sadly is very common for most popular recordings. Search for 'loudness wars'.
When a headphone sounds compressed it usually is bit bloated sounding and lacks realism/clarity/dynamics.

Crisp – is another description of having clarity, an exaggerated form is shrill. The opposite is laid back.

Dark/Darkness – Describes a gradual downward slope of frequencies above 3kHz often combined with an emphasis on bass/lower mids. An exaggerated form is Muffled. The opposite is Shrill

Decay – How a sound/note/resonance fades away over time. This is not always visible in the frequency response but could be visible in Cumulative Spectral Decay (Waterfall) plots. When there is some clarity and no mid bass emphasis most people describe the decay as fast.

Depth – How far away the instruments spacing is from back to front. In headphones this is not always perceived as such. Sometimes a gradual downward slope between 300Hz and 10kHz can also give a sense of depth imitating a greater distance from an actual music performance.

Detail – When something sounds detailed the frequency band between 1kHz and 3kHz is usually audibly flat or neutral and is more or less flat between 3kHz and 10kHz or has some mild emphasis (a few dB) or a peak in that part of the frequency range. An exaggerated form is Edgy, Sharpness, Sibliance.

Dynamic – When a headphone sounds dynamic it sounds 'lively' and instruments and voices 'pop-out' of clearly. It is the opposite of a relaxed, muffled or compressed sound. When a recording is truly dynamic there is a substantial difference between soft sounds and peaks.
You need to play the music loud to hear small details. Funnily enough when you can hear small details already at lower playback levels the sound is actually compressed yet many say, incorrectly, this is a dynamic sound.
The word dynamic is also used to describe a type of headphone driver (the most common type) where a magnet and voice-coil is used to create sound.
Then there is also dynamic range in an electrical sense. This describes how many difference in decibel there is between the signal and the noise and distortion products.
And finally Dynamic Range as in DR-rating. This describes how the difference between the average levels and peak levels in a recording.
Usually the recordings with DR ratings between 10 and 20 sound more realistic than recordings between 2 and 5.

Edgy or Etched – Between 3kHz and 7kHz, as a subset of brightness and Detailed, an exaggerated form of brightness/detail is Edgy, Etched, Sharpness, Sibliance. The opposite (a dip) can be described as lacking Finesse, dis-attached Treble.

Ethereal - Describes a slight treble elevated sound that feels 'fragile' and slightly unnatural but otherwise neutral and pleasant.

Finesse – This describes fine and delicate details heard in music. When one can hear small details clearly, realistic and well defined as in not 'coarse'.
The opposite of finesse is grainy.

Forward – When voices and instruments pop-out. This usually is caused by a mild peak in the 1kHz to 2kHz range. A small peak is usually not very detrimental to the sound at all. The opposite of forward is 'laid-back'.

Flat – When the frequency response does not deviate in the audible range something is considered flat.
When this word is used with headphones or speakers one usually finds that it doesn't color the sound (too much). Strangely enough not everyone that describes a sound as flat it actually is flat. So one man's 'flat' can be another ones bassy, boring, midrangy, or sharp sound.

Fun – This often describes a sound with elevated treble (>5kHz) and bass (<100Hz), also called 'U shaped' or 'bathtub shaped' sound.

Grainy – Also called 'gritty' and describes a sound that lacks finesse. The finer details are rendered a bit coarse. Often this is caused by lots of sharp peaks and deep dips in the treble area above 6kHz.

Harsh – This describes a sound similar as grainy but more concentrated in the midrange from 1kHz to 5kHz rather than in the treble range. Often distortion levels >1% in the midrange and or sharp peaks in that frequency range are the cause.

Highs – Also called treble, these are the upper frequencies above 5kHz.

Imaging – When a headphone or speaker images well it is very easy to pinpoint where instruments are and that image is stable and 'sharp'. This requires a flat frequency response, matched between two drivers, and detailed recordings. Slightly angled headphone drivers can have better 'imaging' as they make use of the ear-shape.

Laid back – describes a relaxed sound signature and is caused by an audible and rather wide dip around 2-4kHz. It is the opposite of forward and dynamic. Upper mids and lower treble is under-emphasized and is found to be pleasant for longer listening sessions when the dip is just slight (between 5 and 10dB).

Lush – is similar to Laid back. It describes a rich, warm-pillowy sound usually with slightly elevated lower frequencies and rather wide dip around 2-4kHz with a good treble quality that is a bit higher in level than when said to be laid back.

Microphonics – sensitivity to touch. It could be present in some electronic components but in the case of headphones it refers to mechanically 'conducted' sounds of the cable. When one touches the earcups of a headphone one can hear this quite clearly. Depending on how a cable is coupled to the cup touching the cable does a similar thing but attenuated compared to touching the cups directly.
Cloth covered, thick and rigid cables usually are worse in this aspect then softer and thinner smooth surface cables. When microphonics are bad you can hear a cable rubbing against clothes when moving when music isn't even playing that loud. Some are annoyed others simply don't mind.

Mids/Midrange – Frequencies between roughly 250Hz and 4kHz. One can divide this in lower mids (250Hz - 500Hz) where the 'body' of voices and instrumens is. The mids (500Hz - 1500Hz) and upper mids (1.5kHz to 4kHz). In the upper mids the 'clarity' of voices and instruments as well as 'attack' of musical instruments can be found.

Muddy – Describes the sound of elevated lower mids/upper bass where the bass is not clearly defined and 'blends in' with the rest of the music in an unnaturally elevated way.

Muffled – sound is when it would appear as though the sound is coming from a speaker with a wool blanket draped over it. Lacking in clarity and treble. Frequencies above 2kHz are subdued and or rolled-off.

Natural – When a headphone sounds natural all instruments and voices are reproduced in a realistic manner without coloration, a correct stereo-image and with a dynamic sound.

Neutral – and neutral are closely the same but neutral is more about the reproduction of sound not being colored. All frequencies are reproduced at the proper level.
There can be concensus about how a headphone should measure to be considered neutral.
However, there are many people stating a headphone or speaker sounds neutral to them but in reality is quite colored. The term neutral used in subjective reviews thus can be considered a meaningless term unless comparisons are made to known neutral headphones.

Openness – Has little to do with a headphone being open or closed. It usually refers to a detailed and neutral sounding headphone where instruments have a clear separation. Headphones with a good clarity are often described as being ‘open’ sounding.

Punch – powerful bass and drum hits. A small emphasis around 100-150Hz can relatively boost harmonics of bass drum/bass.

Rolled-off - Sound can be rolled-off in the lows and in the treble or both. Rolled-off bass is lacking in lower bass and can have slightly subdued bass as well. When the treble is rolled off frequencies above 10kHz usually are low in amplitude. When this happens usually the sound is not considered airy and lacks sparkle.

Sharpness – An emphasis between 2kHz and 6kHz can give instruments and voices the impression of being ‘highlighted’ and sharply detailed.

Shrill – An exaggerated form of sharpness usually with an emphasis between 4kHz and 10kHz and is unpleasant, shreaky.

Sibilant / sibilance – A sharp peak in the frequency band between 5kHz and 8kHz can emphasize the ‘s’ sounds in words. A high peak (>5dB) can sound piercing and sharp and is usually fatiguing and unpleasant during longer listening sessions. A lower peak between 3dB and 5dB can create an illusion of ‘detail’ in music and ‘sharply’ defined instruments. In the long run also fatiguing.

Signature – the overall tonal balance of a headphone. One could describe a signature in many ways as found in this article.

Soft – can be used to describe the loudness (amplitude as in how loud the sound is) or the ‘texture’ of the treble. For the latter it can describe the amplitude opposite the bass and mids or define treble quality. When treble is not splashy elevated, sibilant or coarse but ‘sweet’ and pleasant it is often described as soft. When the frequency response is quite ‘flat’ between 6kHz and 15kHz without sharp peaks and dips the treble quality usually is good and depending on the relative level can be said to be soft or sweet.

Soundstage – Describes in 3d terms (height, width and depth) where one perceives recorded instruments. For headphones it is very rare to hear instruments ‘projected’ clearly meters in front of you without special recordings or software. In the vast majority of cases sounds are between left and right ear only. For headphones instead of soundstage the word headstage is used.
This describes how well instruments are defined between left and right and how ‘sharp’ they can be pinpointed. One has to realize the stereo image is created in the studio during mixing/mastering process and is artificial in most cases. For headphones there is crossfeed which makes some recordings easier to listen to but this too is just a ‘trick’.

Sparkle - Usually caused by a slight emphasis in the (upper) treble and describes vibrant treble.

Spatial – describes how sounds can be heard all around you. Requires digital trickery to pull this of with headphones. Usually to get spatial sound multiple speakers are required placed around the listening position. Related terms: crossfeed and soundstage/headstage.

Sweet – Corresponds with ‘lush’. It describes a ‘soft’ sound signature usually with slightly elevated lows and soft (not grainy or splashy) highs.

Timbre – The tone of a note from an instrument. Timbre is determined by the ratio between the fundamental tone and its harmonics. Headphones with a ‘flat’ tonal response should have the proper timbre. When the tonal balance is not flat timbre can be affected and changed. This can sometime be for the better or worse and depends on the recording. Do note that in studios timbre of each individual instrument is often adjusted to fit in the recording. This can’t be undone.

Tonal Balance – describes how 'flat' a headphone is. When the bass is boosted the tonal balance is bassy, when treble is boosted the tonal balance is 'bright'. When a headphone is described as tonally balanced no specific frequency bands are popping out or are subdued.
The real snag here is that some owners may feel a headphone is tonally balanced but in reality it may be bassy, bass-shy, warm, cold, midrangy or bright for instance. Tonally balanced = realistic sounding with no emphasis on anything.
Objectively tonal balance is difficult to prove headphone measurements can differ substantially between test-rigs.

Transparent – Similar to clarity it is a clean clear open and detailed quality.

Warm/warmth – a general downward tilt in the frequency range between 300Hz and 3kHz. The opposite is cold.
 
Last edited:

Mtbf

Active Member
Joined
Jul 8, 2019
Messages
148
Likes
200
There are more descriptors than warm/cool so it does not have to fall under warm/cool... pick one :D
The most fun part is that a lot of people tend to use descriptors below to describe the sound of DACs and amplifiers while objectively these all are 'flat' or 'rolled off'.
I once bought myself a pair of new high end speakers. At low volumes the sound was fine, but every time I raised the volume above a certain level the sound became a bit harsh. This didn’t happen at all with my previous transmission line speakers, those I could play very load without any loss of fidelity. I was puzzled at the time but had to conclude that my amplifier couldn’t handle the impedance curve of my new speakers. The harshness completely took away the joy of listening, so I had to buy myself a new amplifier. Since then I limit my interest to amplifiers that double the output power going from 8 to 4 (to 2) ohms (which still doesn’t guarantee no differences in sound quality).

What do you think could have been the cause of this harshness at increasing levels? Distortion or a change in frequency response?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Could be a lack of output power (clipping levels reached) combined with a lower sensitivity or impedance of the speakers forcing the amp into current limiting... hard to say without actually looking at the output waveform during the harshness (and comparing it to the input signal)
 

whazzup

Addicted to Fun and Learning
Joined
Feb 19, 2020
Messages
575
Likes
486
@solderdude amazing list of definitions!
The soundstage, I've always wondered how speakers pull it off (especially high-end ones in hifi demo rooms), to trick the listener into that sense of space. Is the crucial factor a pair of speakers with the flatest possible in-room frequency response? Or is it more of a product of sophisticated mixing techniques (of the individual music tracks)?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I think it depends on the speakers, the room, the recording itself and the brain.
 

whazzup

Addicted to Fun and Learning
Joined
Feb 19, 2020
Messages
575
Likes
486
I think it depends on the speakers, the room, the recording itself and the brain.
Thanks, realised there're some earlier threads discussing this. Will go read up first instead of cluttering this thread.
 

Dumdum

Senior Member
Joined
Dec 13, 2019
Messages
338
Likes
219
Location
Nottinghamshire, UK
From the above...

Imaging – When a headphone or speaker images well it is very easy to pinpoint where instruments are and that image is stable and 'sharp'. This requires a flat frequency response and detailed sound. Slightly angled headphone drivers can have better 'imaging' as they make use of the ear-shape.

I would challenge this... for imaging to be sharp it requires a matched frequency response between two drivers, not a flat frequency response
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
thanks...changed it to:

Imaging – When a headphone or speaker images well it is very easy to pinpoint where instruments are and that image is stable and 'sharp'. This requires a flat frequency response, matched between two drivers, and detailed recordings. Slightly angled headphone drivers can have better 'imaging' as they make use of the ear-shape.
 

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
Last edited:

30 Ounce

Active Member
Joined
Apr 9, 2019
Messages
155
Likes
217
Higher order harmonic distortions (5th, 6th, 7th etc.) can cause a cold analytical sound often associated with transistor amps. Lower order (2nd, 3rd) can add a fullness (warmth) and sometimes pleasing effect often found in tube amps. It is difficult to hear the lower order as they blend well with the fundamental frequency where the higher order being so far away from the fundamental frequency can be easier to hear at much lower levels.
 
Last edited:

mhardy6647

Grand Contributor
Joined
Dec 12, 2019
Messages
11,217
Likes
24,181
Remember that the second harmonic is one octave above the fundamental. Second harmonic distortion is essentially euphonic by definition.
Third harmonic distortion, however, isn't likely to sound good (I don't think -- we don't have a piano here any more).

E.g., if the fundamental is A (440 Hz, a/k/a A4); the second harmonic is 880 Hz (A5); the third harmonic is 1320 Hz (which is pretty close to, but not exactly E6 @ 1318.51 Hz)
Frequencies based on the piano tuning taken from https://www.ece.iastate.edu/~alexs/classes/2016_Spring_575/HW/HW5/files/piano-key-freq-wikipedia.pdf -- I am seriously not a musician and get all herumschleudert when it comes to the various flavors of scales (well tempered, EDIT: equal tempered [derp :oops:], etc.).

My point, if I have one :rolleyes:, is that some combinations of frequencies sound good (think: chords) -- and some (generally) don't -- I mean, unless you're a fan of some of the avant-garde composers ;)).
Plus, of course, all generalizations are false, including this one. :cool:
 
Last edited:

Francis Vaughan

Addicted to Fun and Learning
Forum Donor
Joined
Dec 6, 2018
Messages
933
Likes
4,697
Location
Adelaide Australia
Harmonic theory get even worse, and when you start talking about pianos, something alluded to in the above link. IMHO, this is something that is overlooked is talking about distortion effects. It isn't for nothing that the piano has often been considered one of the hardest instruments to record and reproduce successfully. The physicas of a piano make it impossible to create an instrument that is in tune with itself. This is a wfetched problem, and underlies the art involved in tuning a piano. On a piano string (and to a lesser extent on other strings instruments) the harmonics are not exact. This is because the string has non-zero width. The result is that the effective length of a string varies for each harmonic, and thus the harmonics are not exact integer multiples of the fundamental. This is hellish, as a it becomes intrinsically impossible to tune a piano such that the harmonics line up.
Next there is the myth of equal temperament. Everyone gets taught that in the modern world everything uses equal temperament, and this allows free modulation between keys, something critical for any advanced music. Pity is that is isn't really true. Almost no instrument actually manages equal temperament, and in practice when physically realised, advantage is taken of the physics of the instrument to intonate it to something other than a perfect equal temperament. Fretless instruments don't and never have cared. Pianos are tuned with a stretched tuning. Octaves are wider than an exact 2:1 ratio. This sounds better, as it pulls the errant harmonics closer. The larger a piano (ie the longer the strings) the less stretch is needed, and on short pianos the stretch can be quite severe. Next, as the equally tepered scale is oftern termed, it makes all notes sound equally bad. It is common to tweak the exact pitches so that common keys are favoured over uncommon ones. There are a whole family of different tunings to choose from. Each with its own benefits and flaws.

So, what does this mean for reproduction of music? Most importantly it means that simplistic ideas if the effects of harmonic distortion are not valid. Low order harmonic distortion products may be considered benign, and even pleasant. But they will affect the reproduction of music in audible ways. Something that is worth emphasising - most people don't realise how harsh many instruments sound live. Indeed they may have a real preference for the smoothness of recorded music over the live sound. Notions that spectral masking will cover up these issues are also invalid. Spectral masking may actually be the cause of changes in the timbre of instruments in the presence of harmonic distortion. It may cover up inharmoniticy, and paint the sound with an inaccurate, and maybe unwelcome, rose gloss. (Preference ratings are just that, that are not accuracy ratings. You do not want to track, mix, or master on speakers or electronics that yield a preferred sound as opposed to an accurate one.) Patrons of live music may have different expectations and preferences in sound to people whose main exposure to music is from recordings.
 

majingotan

Major Contributor
Forum Donor
Joined
Feb 13, 2018
Messages
1,511
Likes
1,781
Location
Laguna, Philippines
Harmonic theory get even worse, and when you start talking about pianos, something alluded to in the above link. IMHO, this is something that is overlooked is talking about distortion effects. It isn't for nothing that the piano has often been considered one of the hardest instruments to record and reproduce successfully. The physicas of a piano make it impossible to create an instrument that is in tune with itself. This is a wfetched problem, and underlies the art involved in tuning a piano. On a piano string (and to a lesser extent on other strings instruments) the harmonics are not exact. This is because the string has non-zero width. The result is that the effective length of a string varies for each harmonic, and thus the harmonics are not exact integer multiples of the fundamental. This is hellish, as a it becomes intrinsically impossible to tune a piano such that the harmonics line up.
Next there is the myth of equal temperament. Everyone gets taught that in the modern world everything uses equal temperament, and this allows free modulation between keys, something critical for any advanced music. Pity is that is isn't really true. Almost no instrument actually manages equal temperament, and in practice when physically realised, advantage is taken of the physics of the instrument to intonate it to something other than a perfect equal temperament. Fretless instruments don't and never have cared. Pianos are tuned with a stretched tuning. Octaves are wider than an exact 2:1 ratio. This sounds better, as it pulls the errant harmonics closer. The larger a piano (ie the longer the strings) the less stretch is needed, and on short pianos the stretch can be quite severe. Next, as the equally tepered scale is oftern termed, it makes all notes sound equally bad. It is common to tweak the exact pitches so that common keys are favoured over uncommon ones. There are a whole family of different tunings to choose from. Each with its own benefits and flaws.

So, what does this mean for reproduction of music? Most importantly it means that simplistic ideas if the effects of harmonic distortion are not valid. Low order harmonic distortion products may be considered benign, and even pleasant. But they will affect the reproduction of music in audible ways. Something that is worth emphasising - most people don't realise how harsh many instruments sound live. Indeed they may have a real preference for the smoothness of recorded music over the live sound. Notions that spectral masking will cover up these issues are also invalid. Spectral masking may actually be the cause of changes in the timbre of instruments in the presence of harmonic distortion. It may cover up inharmoniticy, and paint the sound with an inaccurate, and maybe unwelcome, rose gloss. (Preference ratings are just that, that are not accuracy ratings. You do not want to track, mix, or master on speakers or electronics that yield a preferred sound as opposed to an accurate one.) Patrons of live music may have different expectations and preferences in sound to people whose main exposure to music is from recordings.

IMO piano recording these days is already a solved issue. I've heard modern recordings with lifelike piano sound reproduction and a lot of older recordings that sound terrible. Then I research the gears that have been used with crappy piano sound and found that really sound distorted are the ones that use tube microphones in the recording chain. The outlier of mastering engineering technique (e.g. excessive compression) is omitted in my statement above.
 

Speedskater

Major Contributor
Joined
Mar 5, 2016
Messages
1,610
Likes
1,333
Location
Cleveland, Ohio USA
Higher order harmonic distortions (5th, 6th, 7th etc.) can cause a cold analytical sound often associated with transistor amps. Lower order (2nd, 3rd) can add a fullness (warmth) and sometimes pleasing effect often found in tube amps.
For ancient solid-state amplifiers this may well be true, but in modern 'accurate' solid-state amps all harmonic distortions are orders of magnitude below what is audible. Now tube amps on the other hand are often designed with intentional high harmonic distortion for a euphonic sound.
 
Top Bottom