• 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!

Audibility thresholds of amp and DAC measurements

flipflop

Addicted to Fun and Learning
Joined
Feb 22, 2018
Messages
927
Likes
1,240
Introduction
A recurring theme on ASR is whether or not the various measured qualities of the devices are audible. In this post, I'll present some clear and visual thresholds for when those imperfections can be considered a potentially audible concern. I will not explain the basics of amp/DAC measurements.
Feel free to comment if you have any issues with the content of this post, be it technical, grammatical, or maybe just something that's poorly communicated.

Orange marks strict limits while green marks lenient limits.
Strict thresholds guarantee that anything measuring better will be transparent.
Lenient thresholds are only lenient relatively to the strict thresholds. They are also "strict" from an audible point of view. Devices measuring worse than the lenient thresholds might still sound fully transparent to you or anyone else for that matter.


Frequency range of human hearing
Humans cannot hear sounds of every frequency. The range of hearing for a healthy young person is 20 to 20,000 hertz.
When pure sine waves are reproduced under ideal conditions and at very high volume, a human listener will be able to identify tones as low as 12 Hz (Olson, Harry F. (1967). Music, Physics and Engineering. p. 249). Below 10 Hz it is possible to perceive the single cycles of the sound, along with a sensation of pressure at the eardrums.
For these reasons, the audibility thresholds will deal with frequencies from 0 to 20 kHz.


Dynamic range, linearity
CDs, with undithered audio, have a maximum dynamic range of 96 dB (16 bits). Studies, such as the one conducted by Meyer and Moran, have shown 96 dB of dynamic range is transparent for any normal listening conditions.
120 dB (20 bits) of dynamic range is required for transparency. Anything below -120 dBFS is inaudible. Amir wrote an in-depth article about the subject.

06.png

Allo Boss V 1.2 is unable to reproduce undithered audio from CDs (with content that makes full use of the 16 bits) transparently.

05.png

miniDSP SHD lands between the two targets.

04.png

SMSL VMV D1 has more dynamic range than the required 120 dB for guaranteed transparency.


03.png

Benchmark DAC1 USB doesn't reach the -96 dBFS target.

02.png

Orchard Audio GALA remains flat up to -96 dBFS, but losses linearity before reaching -120.

01.png

Auralic Vega measures perfectly at -120 dBFS, but due to the analyzer itself introducing some noise, it doesn't show the ideal flat line, which it should be.


FFT
NwAvGuy's Heaphone Amp Measurement Recommendations - In this article, NwAvGuy provides some guidelines for good amplifier design. He is "an electrical engineer by education (BSEE) and career", not a psychoacoustician, and he does not cite his sources, so his limits will be treated as lenient.
NwAvGuy says that "noise needs to be -85dB below the signal to be inaudible which works out to only 0.005% THD+N. But music masks distortion so 0.01% (-80dB) is considered acceptable." Since I make no assumptions about the listener or listening material, I'll disregard the -80 dB threshold and go with -85 dB.
His guideline for distortion is <0.05% equivalent to -66 dBFS.

Disclaimer: The FFT spectrum shows the noise floor as being lower than it actually is. This is called “FFT gain” and it allows us to see distortion products that may be well buried inside the noise floor of the amp or DAC. The gain can be as much as 30 dB to 40 dB lower than reality. Comments on the noise floors for the following graphs are for illustrative purposes only.

10.png

Audio-gd NFB-28.28 has a "Massive 3rd Harmonic Distortion" spike that puts it above all 3 thresholds.

07.png

Parks Audio Puffin fails to meet the strict THD+N criteria and the lenient noise criteria, but stays below the lenient distortion limit.

09.png

JDS Labs The Element stays below the lenient thresholds, but doesn't stay below the strict threshold.

08.png

Allo Katana measures well enough to stay below all 3 thresholds.


SINAD
SINAD groups distortion and noise into one figure, meaning the lenient noise threshold of -85 dBFS will be used, because it's not clear whether noise or distortion dominates. When converting THD+N to SINAD, one can simply remove the minus sign and 'FS' like so: 85 dB.

31.png

The 4 DACs to the left of the orange line manage a signal over noise and distortion of 120 dB or more, exceeding the strict threshold. Totaldac d1-six, and all the other DACs to the right of it, don't make the cut for the lenient threshold.


IMD, THD+N vs frequency
For IMD, the same distortion thresholds as for FFT apply: -120 dBFS for the strict and -66 dBFS for the lenient.

12.png

Audio-gd NFB-28.28 exceeds the lenient threshold and therefore also the strict one.

11.png

Musical Fidelity V90-DAC stays between the 2 limits.

I've not come across a device that measures better than the strict threshold.


Like with SINAD, it's not always possible to tell the distortion and noise apart in THD+N vs frequency plots, so we have to go with the lenient noise threshold again.

13.png

Schiit Yggdrasil doesn't manage to stay below the lenient threshold.

14.png

miniDSP SHD and Benchmark DAC3 measure below the lenient threshold, but above the strict one.

No device measures below the strict threshold on this test.


Crosstalk
-60 dB is NwAvGuy's guideline for crosstalk, so half (numerically) of the strict limit of -120 dB.
As crosstalk measurements are still scarce on ASR, I wasn't able to found one breaching the lenient threshold.

17.png

Massdrop THX AAA 789 exceeds the strict threshold at all 3 gain settings, but also stays below the lenient threshold.

18.png

Gustard A20H stays below both thresholds.


Jitter
Jitter is only found in DACs, so I will stray away from the amp guidelines of NwAvGuy for this section. Thankfully, he wrote another article about the subject where he introduces a threshold for jitter:
19.png

Despite "all the research [he's] done", he doesn't cite his sources here either, so this limit will also be treated as lenient.
NwAvGuy uses a sample rate of 44.1 kHz and a main tone of 11.025 kHz. Amir uses a sample rate of 48 kHz, hence a main tone of 12 kHz. I've adjusted the threshold to fit Amir's graphs.

20.png

SMSL Sanskrit 10th exceeds both thresholds regardless of input.

21.png

Topping DX3 Pro exceeds the strict threshold, but not the lenient threshold.

22.png

Khadas Tone Board doesn't exceed either threshold.


Frequency response, channel balance
Going back to NwAvGuy's amp guidelines, he recommends a maximum of 0.5 dB deviation (from 0) in the frequency response.
Psychoacoustics: Facts and Models by Hugo Fastl and Eberhard Zwicker is not a very quotable book, but on pages 180-181 it makes it clear that a change in SPL of less than 0.2 dB can be heard by humans.
In 'Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms' Floyd Toole says the following: "The simplest deviation from flat is probably a spectral tilt. There is some evidence that we can detect slopes of about 0.1 dB/octave, which translates into a 1 dB tilt from 20 Hz to 20 kHz — not much."
0.1 dB is therefore the strict limit.

24.png

The HP Z series laptop's frequency response takes a nosedive right before 20 kHz and exceeds both thresholds.

25.png

Topping DX3 Pro exceeds the strict threshold in both the bass and treble, but stays within the lenient.

26.png

SpeaKa USB DAC stays within both thresholds.

The channel balance threshold from NwAvGuy mirrors the frequency response threshold: if the FR rises by 0.5 dB in one area and drops by 0.5 dB in another, the total discrepancy will be 1 dB, which will be the lenient threshold for channel balance.
The strict thresholds remains 0.1 dB.

27.png

TEAC HA-P50 exceeds both thresholds.

28.png

Neurochrome HP-1 exceeds the strict threshold, but not the lenient threshold before running out of power.

29.png

Sabaj Da3 doesn't exceed either threshold before running out of power.


Output impedance
NwAvGuy's guideline for output impedance is based on the 1/8th rule, placing it at 2 ohms with a worst case scenario of 16 ohms headphones.
The basis for the 1/8th rule is that it allows up to a 1 dB variation in the frequency response. For a 0.1 dB variation a 1/100th rule is derived, meaning the output impedance should be 0.16 ohms or less with a 16 ohms load.
Keep in mind the thresholds vary with load impedance and can be disregarded entirely with headphones that have a flat impedance curve.

30.png

All the amps to the left of Auralic Gemini 2000 exceed the lenient threshold.


Recap of thresholds
Lenient
Dynamic range, linearity: 96 dB
THD, IMD: -66 dBFS / 0.05%
Noise: -85 dBFS / 0.005%
SINAD: 85 dB
Crosstalk: -60 dBFS
Jitter: -110 dBFS, -100 dBFS around the main tone
Frequency response: ±0.5 dB
Channel balance: 1 dB
Output impedance: 2 ohms

Strict
Dynamic range, linearity, SINAD: 120 dB
THD, IMD, noise, crosstalk, jitter: -120 dBFS / 0.0001%
Frequency response, channel balance: ±0.1 dB
Output impedance: 0.16 ohms


Changelog
  • UPDATE #1: Elaborated on the meaning of the thresholds.
  • UPDATE #2: Based on feedback from @solderdude and @restorer-john, I've decided to get rid of the THD+N vs power graphs.
  • The lenient threshold line for the IMD measurement of Musical Fidelity V90-DAC was mistakenly 10 dB lower than it should've been. It has now been fixed.
  • UPDATE #3: Dithering was unaccounted for in a couple of descriptions of CD audio. The sentences have been rephrased to accurately reflect the fact that 96 dB is not the maximum dynamic range of 16 bit audio. Thanks to @bennetng and @Francis Vaughan for pointing this out.
  • UPDATE #4: A strict threshold for output impedance has been introduced. The graph has not been updated for practical reasons. Big thanks to @dc655321 for helping with the math.
  • UPDATE #5: Added SINAD to the 'Recap of thresholds' section.
  • UPDATE #6: As pointed out by @daftcombo, the lenient thresholds on the THD+N vs frequency graphs were incorrect. This has now been fixed.
  • UPDATE #7: Added SINAD to the main section. Added a disclaimer to the part about FFT. Moved updates to a new changelog section at the bottom.
 
Last edited:

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
Wow, this is really great work and well beyond call of duty. :) I will promote it to an article for home page.

The only correction I would offer is that for the jitter threshold, there is very strong masking around the main tone so for the "strict" one, you still want to have the pointing up center part of the threshold that you have from NWAVGUY.
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,597
Likes
2,235
1 dB difference is 10**(-1/20) = 0.89 (1 - 1/8 = 0.875).
0.1 dB is 10**(-0.1/20) = 0.99. So, maybe a 1/100 rule for your output impedance criterion?
Seems (way) over specified to me.

Personally, I have to concentrate like a mofo to hear a 1 dB difference...

I appreciate all your efforts in collating this info.
But, do you think that 120 dB is perhaps too "strict"?
 

confucius_zero

Addicted to Fun and Learning
Joined
Nov 22, 2018
Messages
541
Likes
345
Holy moly smokes, that's some master work flipflop. Thank you so much for helping us further understand the measurements!
 

confucius_zero

Addicted to Fun and Learning
Joined
Nov 22, 2018
Messages
541
Likes
345
So if I understand this right:

strict = not cool from a measurement/implementation/engineering machine standpoint, but we can't hear it since it goes beyond human "golden ears" hearing ability of 120db
lenient = not cool and we might hear it with critical listening.

So that ess hump is under lenient but over strick = we can't hear it.

?
 
Last edited:

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
strict = not cool from a measurement/implementation/engineering machine standpoint, but we might not hear it unless we got "golden ears"
I would put it as "provably inaudible." In other words, no gray area.
 

confucius_zero

Addicted to Fun and Learning
Joined
Nov 22, 2018
Messages
541
Likes
345

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
640
Likes
2,397
Looks really good! I am with @dc655321 I struggle hard to hear a 1 dB difference... Also, I think there should be some words on the actual testing of the audibility threshold in relation to "how" one goes about it. Meaning that masking (all types) come into play and mess with the actual threshold values during normal listening levels.

For example, here I ABX two samples of music where I change one of the files with a broad base eq and ABX the two files "at reference level" i.e. 83 dB SPL whether headphones or speakers, the best I could do is 1 dB change: https://www.computeraudiophile.com/...gital-Audio-–-Bit-Perfect-Audibility-Testing/

Also, I changed the bit depth in JRiver Media Center, which has a bit depth simulator, that you all can try yourselves if you have JRiver and in comparison to a non-altered file during an ABX test, using music, the best I could do is 12 bits. So -72 dBFS down when compared to the "unaltered" file. Note that this is ABX comparison to a non-altered file, both playing a reference level.

What I am saying is adding some context around "how" one goes about performing audibility testing. Having someone cranking the volume to max in the dead of night with headphones on and saying I can hear noise or distortion or whatever, is not a valid test...
 

Zilfallion

Member
Joined
Feb 16, 2018
Messages
82
Likes
65
And this is part of the reason I'm still happy with my Element even after it just measured okay, and not really looking for an upgrade anytime soon unless I end up needing more features out of my DAC/Amp.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
Just a note: an exhaustive analysis and proof simply does not exist for any of this. As such, take them as guidelines which are better than having none.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Introduction
A recurring theme on ASR is whether or not the various measured qualities of the devices are audible. In this post, I'll present some clear and visual thresholds for when those imperfections can be considered a potentially audible concern. I will not explain the basics of amp/DAC measurements.
Feel free to comment if you have any issues with the content of this post, be it technical, grammatical, or maybe just something that's poorly communicated.
Orange marks strict limits while green marks lenient limits.


Frequency range of human hearing
Humans cannot hear sounds of every frequency. The range of hearing for a healthy young person is 20 to 20,000 hertz.
When pure sine waves are reproduced under ideal conditions and at very high volume, a human listener will be able to identify tones as low as 12 Hz (Olson, Harry F. (1967). Music, Physics and Engineering. p. 249.). Below 10 Hz it is possible to perceive the single cycles of the sound, along with a sensation of pressure at the eardrums.
For these reasons, the audibility thresholds will deal with frequencies from 0 to 20 kHz.


Dynamic range, linearity
120 dB (20 bits) of dynamic range is required for transparency. Anything below -120 dBFS is inaudible.
CDs have a maximum dynamic range of 96 dB (16 bits). Studies, such as the one conducted by Meyer and Moran, have shown 96 dB of dynamic range is transparent for any normal listening conditions.
Amir wrote an in-depth article about the subject.

View attachment 18959
Allo Boss V 1.2 is unable to reproduce audio from CDs (with content that makes full use of the 16 bits) transparently.

View attachment 18960
miniDSP SHD lands between the two targets.

View attachment 18961
SMSL VMV D1 has more dynamic range than the required 120 dB for guaranteed transparency.


View attachment 18963
Benchmark DAC1 USB doesn't reach the -96 dBFS target.

View attachment 18964
Orchard Audio GALA remains flat up to -96 dBFS, but losses linearity before reaching -120.

View attachment 18965
Auralic Vega measures perfectly at -120 dBFS, but due to the analyzer itself introducing some noise, it doesn't show the ideal flat line, which it should be.


FFT
NwAvGuy's Heaphone Amp Measurement Recommendations - In this article, NwAvGuy provides some guidelines for good amplifier design. He is "an electrical engineer by education (BSEE) and career", not a psychoacoustician, and he does not cite his sources, so his limits will be treated as lenient.
NwAvGuy says that "noise needs to be -85dB below the signal to be inaudible which works out to only 0.005% THD+N. But music masks distortion so 0.01% (-80dB) is considered acceptable." Since I make no assumptions about the listener or listening material, I'll disregard the -80 dB threshold and go with -85 dB.
His guideline for distortion is <0.05% equivalent to -66 dB.

View attachment 18966
Audio-gd NFB-28.28 has a "Massive 3rd Harmonic Distortion" spike that puts it above all 3 thresholds.

View attachment 18967
Parks Audio Puffin fails to meet the strict THD+N criteria and the lenient noise criteria, but stays below the lenient distortion limit.

View attachment 18969
JDS Labs The Element stays below the lenient thresholds, but doesn't stay below the strict threshold.

View attachment 18970
Allo Katana measures well enough to stay below all 3 thresholds.


Since SINAD and THD+N Ratio in the dashboard view don't reveal anything not already shown by the FFT spectrum, I won't elaborate on them.


IMD, THD+N vs frequency, THD+N vs power
For IMD, the same distortion thresholds as for FFT apply: -120 dBFS for the strict and -66 dBFS for the lenient.

View attachment 18971
Audio-gd NFB-28.28 exceeds the lenient threshold and therefore also the strict one.

View attachment 18972
Musical Fidelity V90-DAC stays between the 2 limits.

I've not come across a device that measures better than the strict threshold.


In THD+N vs frequency plots, it's not always possible to tell the distortion and noise apart, so the lenient noise threshold of -85 dBFS / 0.005% THD+N will be used.

View attachment 18973
Behringer doesn't manage to stay below the lenient threshold. The strict threshold is off the chart, so it hasn't been marked.

View attachment 18974
miniDSP SHD and Benchmark DAC3 measure below the lenient threshold, but above the strict one.

No device measures below the strict threshold on this test.


THD+N vs power is the same deal as THD+N vs Frequency, so it will be brief.

View attachment 18975
View attachment 18976
Massdrop THX AAA 789 measures below the lenient threshold (before clipping sets in) with a 300 ohms load and above both thresholds with a 33 ohms load.

No device measures below the strict threshold on this test.


Crosstalk
-60 dB is NwAvGuy's guideline for crosstalk, so half (numerically) of the strict limit of -120 dB.
As crosstalk measurements are still scarce on ASR, I wasn't able to found one breaching the lenient threshold.

View attachment 18977
Massdrop THX AAA 789 exceeds the strict threshold at all 3 gain settings, but also stays below the lenient threshold.

View attachment 18978
Gustard A20H stays below both thresholds.


Jitter
Jitter is only found in DACs, so I will stray away from the amp guidelines of NwAvGuy for this section. Thankfully, he wrote another article about the subject where he introduces a threshold for jitter:
View attachment 18990

Despite "all the research [he's] done", he doesn't cite his sources here either, so this limit will also be treated as lenient.
NwAvGuy uses a sample rate of 44.1 kHz and a main tone of 11.025 kHz. Amir uses a sample rate of 48 kHz, hence a main tone of 12 kHz. I've adjusted the threshold to fit Amir's graphs.

View attachment 18980
SMSL Sanskrit 10th exceeds both thresholds regardless of input.

View attachment 18981
Topping DX3 Pro exceeds the strict threshold, but not the lenient threshold.

View attachment 18982
Khadas Tone Board doesn't exceed either threshold.


Frequency response, channel balance
Going back to NwAvGuy's amp guidelines, he recommends a maximum of 0.5 dB deviation (from 0) in the frequency response.
Psychoacoustics: Facts and Models by Hugo Fastl and Eberhard Zwicker is not a very quotable book, but on pages 180-181 it makes it clear that a change in SPL of less than 0.2 dB can be heard by humans.
In 'Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms' Floyd Toole says the following: "The simplest deviation from flat is probably a spectral tilt. There is some evidence that we can detect slopes of about 0.1 dB/octave, which translates into a 1 dB tilt from 20 Hz to 20 kHz — not much."
0.1 dB is therefore the strict limit.

View attachment 18984
The HP Z series laptop's frequency response takes a nosedive right before 20 kHz and exceeds both thresholds.

View attachment 18985
Topping DX3 Pro exceeds the strict threshold in both the bass and treble, but stays within the lenient.

View attachment 18986
SpeaKa USB DAC stays within both thresholds.

The channel balance threshold from NwAvGuy mirrors the frequency response threshold: if the FR rises by 0.5 dB in one area and drops by 0.5 dB in another, the total discrepancy will be 1 dB, which will be the lenient threshold for channel balance.
The strict thresholds remains 0.1 dB.

View attachment 18987
TEAC HA-P50 exceeds both thresholds.

View attachment 18988
Neurochrome HP-1 exceeds the strict threshold, but not the lenient threshold before running out of power.

View attachment 18989
Sabaj Da3 doesn't exceed either threshold before running out of power.


Output impedance
NwAvGuy's guideline for output impedance is based on the 1/8th rule, placing it at 2 ohms with a worst case scenario of 16 ohms headphones.
The basis for the 1/8th rule is that it allows up to a 1 dB variation in the frequency response. I can't get his math to make sense, but I would like to make a similar rule for a 0.1 dB variation at some point. Right now, there won't be a strict threshold. The math can be found in the 'tech section' of this article: http://nwavguy.blogspot.com/2011/02/headphone-amp-impedance.html

View attachment 18991
All the amps to the left of Auralic Gemini 2000 exceed the lenient threshold.


Recap of thresholds
Strict
Dynamic range, linearity: 120 dB / 20 bits
THD, IMD, noise, crosstalk, jitter: -120 dBFS / 0.0001%
Frequency response, channel balance: ±0.1 dB
Output impedance: TBD

Lenient
Dynamic range, linearity: 96 dB / 16 bits
THD, IMD: -66 dBFS / 0.05%
Noise: -85 dBFS / 0.005%
Crosstalk: -60 dBFS
Jitter: -110 dBFS, -100 dBFS around the main tone
Frequency response: ±0.5 dB
Channel balance: 1 dB
Output impedance: 2 ohms

Impressive and informative recap. However, let us not forget about electro-mecahnical devices (usually called speakers) that actually create sound.
Would you care to comment on their typical THD/IMD figures?

Also, regarding the noise level that can actually be heard, this thead may actually reveal that -120dB is way to strict and that dynamic range of good old RBCD is actually satisfactory.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,440
Likes
9,100
Location
Suffolk UK
I date from a time when analogue measurements was all there was, and my own transparency criteria come from that background.

Firstly, all my transparency criteria are done in the presence of programme material, speech and music, not test tones.

On that basis, even 1% THD or IMD is very hard to hear, if not impossible, so any amplifier or source that has less than 0.1% THD (-60dB) at all frequencies, levels and permitted loads will be transparent as far as distortion goes. Studio quality tape recorders have 3% THD on peaks, and there are superb recordings done that way.

A frequency response variation of even 3dB is hard to hear (again on speech or music, not noise or other test signals) so a 1dB variation is inaudible.

Noise rather depends on the dynamic range of the signal, but even only -30dB is pretty much inaudible, although clearly, in the quiet bits that -30dB needs to be a lot lower than in the loud bits, but overall -60dB is quiet enough not to be noticeable if the volume control is arranged for normal listening levels on peaks.

As to crosstalk, even 20dB is quite enough to give a good stereo image. One issue with crosstalk is that the distortion in the crosstalk should be low, as it's quite easy for the distortion in the non-speaking channel due to crosstalk in the speaking channel exceeds the distortion of that channel when measured individually.

As to jitter, I've never heard it on a digital source. In analogue, the equivalent is Wow and Flutter, but again, on music or speech 0.1% weighted is inaudible except perhaps on long steady notes, like flute or piano, which in the case particularly of the flute, get close to sine waves.

Thresholds can be lower using test tones or noise, but as far as listening to programme material goes, I've not seen any evidence that the above criteria are inadequate.

S.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Thresholds can be lower using test tones or noise, but as far as listening to programme material goes, I've not seen any evidence that the above criteria are inadequate.

S.

I cannot agree more. And yet, we are constantly being bombarded by some ridiculously lower threshols. Manufacturers use them to compete and many folks are claiming to hear difference between them - not in blind tests though..
 
OP
F

flipflop

Addicted to Fun and Learning
Joined
Feb 22, 2018
Messages
927
Likes
1,240
Thanks @flipflop for the great effort involved in putting this together, I encourage those participating in this thread to reply in kind.
Thank you, Thomas. I personally want to thank @amirm, @Floyd Toole, Zwicker, Fastl, NwAvGuy, and everyone else I cited for providing content and information that I could stitch together.

I will promote it to an article for home page.
You're putting a lot of pressure on me. Thank you nonetheless :)
The only correction I would offer is that for the jitter threshold, there is very strong masking around the main tone so for the "strict" one, you still want to have the pointing up center part of the threshold that you have from NWAVGUY.
What you're saying seems intuitive, but I'll have to ask you to provide a source. Without exact numbers, it would be an estimation that I wouldn't feel comfortable making.

1 dB difference is 10**(-1/20) = 0.89 (1 - 1/8 = 0.875).
0.1 dB is 10**(-0.1/20) = 0.99. So, maybe a 1/100 rule for your output impedance criterion?
I might implement that if no one objects to your math.
Seems (way) over specified to me.
Right, because it's based on a threshold that's derived from extreme listening tests.
Personally, I have to concentrate like a mofo to hear a 1 dB difference...
A 1 dB low-Q peak can be more audible than e.g. a 3 dB high-Q peak.
Toole and Olive arrived at a 0.5 dB figure in one of their studies, which I haven't been able to get my hands on. I believe they used 'normal' music in their testing.
But, do you think that 120 dB is perhaps too "strict"?
Absolutely. I follow the lenient thresholds for everything except for FR and channel balance.

Looks really good! I am with @dc655321 I struggle hard to hear a 1 dB difference... Also, I think there should be some words on the actual testing of the audibility threshold in relation to "how" one goes about it. Meaning that masking (all types) come into play and mess with the actual threshold values during normal listening levels.
I consciously chose to ignore masking, except for the main tones, because it makes assumptions about the playback material.
What I am saying is adding some context around "how" one goes about performing audibility testing. Having someone cranking the volume to max in the dead of night with headphones on and saying I can hear noise or distortion or whatever, is not a valid test...
What your own listening tests reveal might only apply to yourself. The sources I've referred to are way more authoritative on the matter.

Would you care to comment on their typical THD/IMD figures?
Not sure what kind of comment you're looking for. Distortion ratios from real headphone loads are higher than those measured by the dummy loads.
This thread is gaining a lot more traction than I thought it would, so I just want to make it clear that I'm not a professional in any area related to audio or electronics.
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Great summary but have some questions. Here is one.

index.php


Why would the limit not be sloping like the THD plot but remain horizontal as in the plot above over the entire power output range ?
At 1uW even a 120dB@1mW IEM headphone produces just 90dB SPL.
Even very efficient 'normal' headphones would merely produce about 70dB SPL
Why would the 'strict' line need to be 120dB all the way down to nV territory in this case ?
The same goes for the lenient line btw.
It would be safe to assume nobody on this planet can hear -30dBA let alone -50dBA.

I think the realistic dynamic range for music reproduction is about 70dB at normal to realisticly loud levels.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Not sure what kind of comment you're looking for. Distortion ratios from real headphone loads are higher than those measured by the dummy loads.
This thread is gaining a lot more traction than I thought it would, so I just want to make it clear that I'm not a professional in any area related to audio or electronics.

How do you expect to hear THD of 0.05% or even 0.005% when loudspeakers are having THD of app 0.5% (above app 200Hz)?

Btw, right now I'm listening to my tube amp with THD of 0.2% and SNR of 70dB and cannot hear neither that THD nor noise when switching to my Rotel amp which has far better specs than that. How do you explain that?
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,597
Likes
2,235
I might implement that if no one objects to your math.

Please don't "implement" that.
I was trying to call attention to the insignificance.

A 1 dB low-Q peak can be more audible than e.g. a 3 dB high-Q peak.
Toole and Olive arrived at a 0.5 dB figure in one of their studies, which I haven't been able to get my hands on. I believe they used 'normal' music in their testing.

My personal experience was from listening to a single tone: 1 dB is difficult to discern.

A smaller change in dB over a broader range of frequencies (low-Q) may be more noticeable as the difference in available sound energy can significantly greater than over a narrow-Q, higher dB shift. One may notice tonal shifts like this (music becomes slightly more bass-heavy, or treble-oriented, etc). But, at what center frequency and over what range is this 0.5 dB shift reliably audible ?
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
640
Likes
2,397
mitchco said:
Looks really good! I am with @dc655321 I struggle hard to hear a 1 dB difference... Also, I think there should be some words on the actual testing of the audibility threshold in relation to "how" one goes about it. Meaning that masking (all types) come into play and mess with the actual threshold values during normal listening levels.

I consciously chose to ignore masking, except for the main tones, because it makes assumptions about the playback material.

mitchco said:
What I am saying is adding some context around "how" one goes about performing audibility testing. Having someone cranking the volume to max in the dead of night with headphones on and saying I can hear noise or distortion or whatever, is not a valid test...

What your own listening tests reveal might only apply to yourself. The sources I've referred to are way more authoritative on the matter.
----------------
Maybe I am missing something completely, but if you are talking about audibility threshold of what we humans can hear or cannot hear, you cannot ignore masking, as that is part of the human condition and applies to all of us.

My listening tests don't apply just to myself, as several have reported similar when they took the same ABX tests I performed. I spent 10 years as a pro recording and music engineer and others in the industry who do this for a living have come up with similar results... Many of these audibility tests, with similar results as mine, can be found at Hydrogen Audio... https://hydrogenaud.io/

No offence to @amirm or @flipflop Fantastic that measurements are being made, but there does need to be some guidelines as to how one is testing an audibility threshold that means something to the average listener. Something that they can actually download and listen to which usually means music. That way folks can ABX something transient in nature, and not steady state tones... and determine their own audibility threshold...

@Ethan Winer has an artefact audibility comparisons that folks can download and try: http://ethanwiner.com/audibility.html

Measurements are meaningless if we don't know what is audible and what is not... And yah, there is a bell curve that would apply to most of us :)

Edit - lol obviously I don't know how to multi quote
 
Top Bottom